Symfony Profiler

Security

Token

There is no security token.

Firewall

main Name

Security enabled

Stateless

Configuration

Key	Value
provider	security.user.provider.concrete.app_user_provider
context	main
entry_point	App\Security\KbinAuthenticator
user_checker	App\Security\UserChecker
access_denied_handler	(none)
access_denied_url	(none)
authenticators	[`"two_factor" "remember_me" "App\Security\KbinAuthenticator" "App\Security\FacebookAuthenticator" "App\Security\GoogleAuthenticator" "App\Security\GithubAuthenticator" "App\Security\KeycloakAuthenticator"`]

Listeners

Listener	Duration	Response
Symfony\Component\Security\Http\Firewall\ChannelListener {#723`-map: Symfony\Component\Security\Http\AccessMap {#722 …} -logger: Monolog\Logger {#783 …} -httpPort: 80 -httpsPort: 443`}	0.00 ms	(none)
Symfony\Component\Security\Http\Firewall\ContextListener {#706-tokenStorage: Symfony\Component\Security\Core\Authentication\Token\Storage\TokenStorage {#1017 …} -sessionKey: "_security_main" -logger: Monolog\Logger {#783 …} -userProviders: Symfony\Component\DependencyInjection\Argument\RewindableGenerator {#705 …} -dispatcher: Symfony\Component\EventDispatcher\Debug\TraceableEventDispatcher {#747 …} -registered: false -trustResolver: Scheb\TwoFactorBundle\Security\Authentication\AuthenticationTrustResolver {#780 …} -sessionTrackerEnabler: Symfony\Component\Security\Core\Authentication\Token\Storage\UsageTrackingTokenStorage::enableUsageTracking(): void {#703 …}}	7.91 ms	(none)
Symfony\Component\Security\Http\Firewall\AuthenticatorManagerListener {#584`-authenticatorManager: Symfony\Component\Security\Http\Authentication\AuthenticatorManager {#595 …}`}	0.00 ms	(none)
Scheb\TwoFactorBundle\Security\Http\Firewall\TwoFactorAccessListener {#582`-twoFactorFirewallConfig: Scheb\TwoFactorBundle\Security\TwoFactor\TwoFactorFirewallConfig {#842 …} -tokenStorage: Symfony\Component\Security\Core\Authentication\Token\Storage\UsageTrackingTokenStorage {#1018 …} -twoFactorAccessDecider: Scheb\TwoFactorBundle\Security\Authorization\TwoFactorAccessDecider {#581 …}`}	0.08 ms	(none)
Symfony\Component\Security\Http\Firewall\AccessListener {#579`-tokenStorage: Symfony\Component\Security\Core\Authentication\Token\Storage\UsageTrackingTokenStorage {#1018 …} -accessDecisionManager: Symfony\Component\Security\Core\Authorization\TraceableAccessDecisionManager {#937 …} -map: Symfony\Component\Security\Http\AccessMap {#722 …}`}	0.00 ms	(none)
Symfony\Component\Security\Http\Firewall\LogoutListener {#786`-tokenStorage: Symfony\Component\Security\Core\Authentication\Token\Storage\UsageTrackingTokenStorage {#1018 …} -options: [ "csrf_parameter" => "_csrf_token" "csrf_token_id" => "logout" "logout_path" => "app_logout" ] -httpUtils: Symfony\Component\Security\Http\HttpUtils {#841 …} -csrfTokenManager: Symfony\Component\Security\Csrf\CsrfTokenManager {#1015 …} -eventDispatcher: Symfony\Component\EventDispatcher\Debug\TraceableEventDispatcher {#747 …}`}	0.00 ms	(none)

Authenticators

No authenticators have been recorded. Check previous profiles on your authentication endpoint.

Access Decision

affirmative Strategy


"Symfony\Component\Security\Core\Authorization\Voter\AuthenticatedVoter"
"Scheb\TwoFactorBundle\Security\Authorization\Voter\TwoFactorInProgressVoter"
"Symfony\Component\Security\Core\Authorization\Voter\RoleHierarchyVoter"
"Symfony\Component\Security\Core\Authorization\Voter\ExpressionVoter"
"App\Security\Voter\EntryCommentVoter"
"App\Security\Voter\EntryVoter"
"App\Security\Voter\MagazineVoter"
"App\Security\Voter\MessageThreadVoter"
"App\Security\Voter\MessageVoter"
"App\Security\Voter\NotificationVoter"
"App\Security\Voter\OAuth2UserConsentVoter"
"App\Security\Voter\PostCommentVoter"
"App\Security\Voter\PostVoter"
"App\Security\Voter\UserVoter"

#	Voter class
1	"Symfony\Component\Security\Core\Authorization\Voter\AuthenticatedVoter"
2	"Scheb\TwoFactorBundle\Security\Authorization\Voter\TwoFactorInProgressVoter"
3	"Symfony\Component\Security\Core\Authorization\Voter\RoleHierarchyVoter"
4	"Symfony\Component\Security\Core\Authorization\Voter\ExpressionVoter"
5	"App\Security\Voter\EntryCommentVoter"
6	"App\Security\Voter\EntryVoter"
7	"App\Security\Voter\MagazineVoter"
8	"App\Security\Voter\MessageThreadVoter"
9	"App\Security\Voter\MessageVoter"
10	"App\Security\Voter\NotificationVoter"
11	"App\Security\Voter\OAuth2UserConsentVoter"
12	"App\Security\Voter\PostCommentVoter"
13	"App\Security\Voter\PostVoter"
14	"App\Security\Voter\UserVoter"

Access decision log


null
"App\Security\Voter\UserVoter"
App\Entity\Entry {#2412
  +user: Proxies\__CG__\App\Entity\User {#1970 …}
  +magazine: App\Entity\Magazine {#266
    +icon: Proxies\__CG__\App\Entity\Image {#247 …}
    +name: "privacy@lemmy.ml"
    +title: "privacy"
    +description: """
      A place to discuss privacy and freedom in the digital world.\n
      ============================================================\n
      \n
      Privacy has become a very important issue in modern society, with companies and governments constantly abusing their power, more and more people are waking up to the importance of digital privacy.\n
      \n
      In this community everyone is welcome to post links and discuss topics related to privacy.\n
      \n
      ### Some Rules\n
      \n
      - Posting a link to a website containing tracking isn’t great, if contents of the website are behind a paywall maybe copy them into the post\n
      - Don’t promote proprietary software\n
      - Try to keep things on topic\n
      - If you have a question, please try searching for previous discussions, maybe it has already been answered\n
      - Reposts are fine, but should have at least a couple of weeks in between so that the post can reach a new audience\n
      - Be nice :)\n
      \n
      ### Related communities\n
      \n
      - [Lemmy.ml libre_culture](https://lemmy.ml/c/libre_culture)\n
      - [Lemmy.ml privatelife](https://lemmy.ml/c/privatelife)\n
      - [Lemmy.ml DeGoogle](https://lemmy.ml/c/degoogle)\n
      - [Lemmy.ca privacy](https://lemmy.ca/c/privacy)\n
      \n
      ### Chat rooms\n
      \n
      - [Matrix/Element]Dead\n
      - [Discord](https://discord.gg/8xfrVjvngh)\n
      \n
      much thanks to @gary_host_laptop for the logo design :)
      """
    +rules: null
    +subscriptionsCount: 1
    +entryCount: 561
    +entryCommentCount: 7684
    +postCount: 6
    +postCommentCount: 73
    +isAdult: false
    +customCss: null
    +lastActive: DateTime @1729582580 {#276
      date: 2024-10-22 09:36:20.0 +02:00
    }
    +markedForDeletionAt: null
    +tags: null
    +moderators: Doctrine\ORM\PersistentCollection {#238 …}
    +ownershipRequests: Doctrine\ORM\PersistentCollection {#234 …}
    +moderatorRequests: Doctrine\ORM\PersistentCollection {#223 …}
    +entries: Doctrine\ORM\PersistentCollection {#181 …}
    +posts: Doctrine\ORM\PersistentCollection {#139 …}
    +subscriptions: Doctrine\ORM\PersistentCollection {#201 …}
    +bans: Doctrine\ORM\PersistentCollection {#118 …}
    +reports: Doctrine\ORM\PersistentCollection {#104 …}
    +badges: Doctrine\ORM\PersistentCollection {#82 …}
    +logs: Doctrine\ORM\PersistentCollection {#72 …}
    +awards: Doctrine\ORM\PersistentCollection {#61 …}
    +categories: Doctrine\ORM\PersistentCollection {#1820 …}
    -id: 85
    +apId: "privacy@lemmy.ml"
    +apProfileId: "https://lemmy.ml/c/privacy"
    +apPublicUrl: "https://lemmy.ml/c/privacy"
    +apFollowersUrl: "https://lemmy.ml/c/privacy/followers"
    +apInboxUrl: "https://lemmy.ml/inbox"
    +apDomain: "lemmy.ml"
    +apPreferredUsername: "privacy"
    +apDiscoverable: true
    +apManuallyApprovesFollowers: null
    +privateKey: null
    +publicKey: null
    +apFetchedAt: DateTime @1729583592 {#270
      date: 2024-10-22 09:53:12.0 +02:00
    }
    +apDeletedAt: null
    +apTimeoutAt: null
    +visibility: "visible             "
    +createdAt: DateTimeImmutable @1699780272 {#272
      date: 2023-11-12 10:11:12.0 +01:00
    }
  }
  +image: Proxies\__CG__\App\Entity\Image {#1943 …}
  +domain: Proxies\__CG__\App\Entity\Domain {#1912 …}
  +slug: "Google-Researchers-Attack-Prompts-ChatGPT-to-Reveal-Its-Training-Data"
  +title: "Google Researchers’ Attack Prompts ChatGPT to Reveal Its Training Data"
  +url: "https://www.404media.co/google-researchers-attack-convinces-chatgpt-to-reveal-its-training-data/"
  +body: """
    > ChatGPT is full of sensitive private information and spits out verbatim text from CNN, Goodreads, WordPress blogs, fandom wikis, Terms of Service agreements, Stack Overflow source code, Wikipedia pages, news blogs, random internet comments, and much more.\n
    \n
    > Using this tactic, the researchers showed that there are large amounts of privately identifiable information (PII) in OpenAI’s large language models. They also showed that, on a public version of ChatGPT, the chatbot spit out large passages of text scraped verbatim from other places on the internet.\n
    \n
    > “In total, 16.9 percent of generations we tested contained memorized PII,” they wrote, which included “identifying phone and fax numbers, email and physical addresses … social media handles, URLs, and names and birthdays.”\n
    \n
    Edit: [The full paper that's referenced in the article can be found here](https://arxiv.org/pdf/2311.17035.pdf)
    """
  +type: "link"
  +lang: "en"
  +isOc: false
  +hasEmbed: false
  +commentCount: 80
  +favouriteCount: 364
  +score: 0
  +isAdult: false
  +sticky: false
  +lastActive: DateTime @1701712043 {#2410
    date: 2023-12-04 18:47:23.0 +01:00
  }
  +ip: null
  +adaAmount: 0
  +tags: null
  +mentions: null
  +comments: Doctrine\ORM\PersistentCollection {#1866 …}
  +votes: Doctrine\ORM\PersistentCollection {#1966 …}
  +reports: Doctrine\ORM\PersistentCollection {#1946 …}
  +favourites: Doctrine\ORM\PersistentCollection {#1368 …}
  +notifications: Doctrine\ORM\PersistentCollection {#2426 …}
  +badges: Doctrine\ORM\PersistentCollection {#2439 …}
  +children: []
  -id: 18235
  -titleTs: "'attack':3 'chatgpt':5 'data':10 'googl':1 'prompt':4 'research':2 'reveal':7 'train':9"
  -bodyTs: "'/pdf/2311.17035.pdf)':134 '16.9':88 'address':109 'agreement':23 'also':61 'amount':48 'articl':127 'arxiv.org':133 'arxiv.org/pdf/2311.17035.pdf)':132 'birthday':117 'blog':17,31 'chatbot':71 'chatgpt':1,69 'cnn':14 'code':27 'comment':34 'contain':94 'edit':118 'email':106 'fandom':18 'fax':104 'found':130 'full':3,120 'generat':91 'goodread':15 'handl':112 'identifi':51,101 'includ':100 'inform':7,52 'internet':33,85 'languag':58 'larg':47,57,74 'media':111 'memor':95 'model':59 'much':36 'name':115 'news':30 'number':105 'openai':55 'overflow':25 'page':29 'paper':121 'passag':75 'percent':89 'phone':102 'physic':108 'pii':53,96 'place':82 'privat':6,50 'public':66 'random':32 'referenc':124 'research':42 'scrape':78 'sensit':5 'servic':22 'show':43,62 'social':110 'sourc':26 'spit':9,72 'stack':24 'tactic':40 'term':20 'test':93 'text':12,77 'total':87 'url':113 'use':38 'verbatim':11,79 'version':67 'wiki':19 'wikipedia':28 'wordpress':16 'wrote':98"
  +cross: false
  +upVotes: 0
  +downVotes: 0
  +ranking: 1701362787
  +visibility: "visible             "
  +apId: "https://kbin.social/m/privacy@lemmy.ml/t/665071"
  +editedAt: DateTimeImmutable @1702265006 {#1850
    date: 2023-12-11 04:23:26.0 +01:00
  }
  +createdAt: DateTimeImmutable @1701276387 {#2420
    date: 2023-11-29 17:46:27.0 +01:00
  }
}
"App\Security\Voter\UserVoter"
App\Entity\Entry {#2412
  +user: Proxies\__CG__\App\Entity\User {#1970 …}
  +magazine: App\Entity\Magazine {#266
    +icon: Proxies\__CG__\App\Entity\Image {#247 …}
    +name: "privacy@lemmy.ml"
    +title: "privacy"
    +description: """
      A place to discuss privacy and freedom in the digital world.\n
      ============================================================\n
      \n
      Privacy has become a very important issue in modern society, with companies and governments constantly abusing their power, more and more people are waking up to the importance of digital privacy.\n
      \n
      In this community everyone is welcome to post links and discuss topics related to privacy.\n
      \n
      ### Some Rules\n
      \n
      - Posting a link to a website containing tracking isn’t great, if contents of the website are behind a paywall maybe copy them into the post\n
      - Don’t promote proprietary software\n
      - Try to keep things on topic\n
      - If you have a question, please try searching for previous discussions, maybe it has already been answered\n
      - Reposts are fine, but should have at least a couple of weeks in between so that the post can reach a new audience\n
      - Be nice :)\n
      \n
      ### Related communities\n
      \n
      - [Lemmy.ml libre_culture](https://lemmy.ml/c/libre_culture)\n
      - [Lemmy.ml privatelife](https://lemmy.ml/c/privatelife)\n
      - [Lemmy.ml DeGoogle](https://lemmy.ml/c/degoogle)\n
      - [Lemmy.ca privacy](https://lemmy.ca/c/privacy)\n
      \n
      ### Chat rooms\n
      \n
      - [Matrix/Element]Dead\n
      - [Discord](https://discord.gg/8xfrVjvngh)\n
      \n
      much thanks to @gary_host_laptop for the logo design :)
      """
    +rules: null
    +subscriptionsCount: 1
    +entryCount: 561
    +entryCommentCount: 7684
    +postCount: 6
    +postCommentCount: 73
    +isAdult: false
    +customCss: null
    +lastActive: DateTime @1729582580 {#276
      date: 2024-10-22 09:36:20.0 +02:00
    }
    +markedForDeletionAt: null
    +tags: null
    +moderators: Doctrine\ORM\PersistentCollection {#238 …}
    +ownershipRequests: Doctrine\ORM\PersistentCollection {#234 …}
    +moderatorRequests: Doctrine\ORM\PersistentCollection {#223 …}
    +entries: Doctrine\ORM\PersistentCollection {#181 …}
    +posts: Doctrine\ORM\PersistentCollection {#139 …}
    +subscriptions: Doctrine\ORM\PersistentCollection {#201 …}
    +bans: Doctrine\ORM\PersistentCollection {#118 …}
    +reports: Doctrine\ORM\PersistentCollection {#104 …}
    +badges: Doctrine\ORM\PersistentCollection {#82 …}
    +logs: Doctrine\ORM\PersistentCollection {#72 …}
    +awards: Doctrine\ORM\PersistentCollection {#61 …}
    +categories: Doctrine\ORM\PersistentCollection {#1820 …}
    -id: 85
    +apId: "privacy@lemmy.ml"
    +apProfileId: "https://lemmy.ml/c/privacy"
    +apPublicUrl: "https://lemmy.ml/c/privacy"
    +apFollowersUrl: "https://lemmy.ml/c/privacy/followers"
    +apInboxUrl: "https://lemmy.ml/inbox"
    +apDomain: "lemmy.ml"
    +apPreferredUsername: "privacy"
    +apDiscoverable: true
    +apManuallyApprovesFollowers: null
    +privateKey: null
    +publicKey: null
    +apFetchedAt: DateTime @1729583592 {#270
      date: 2024-10-22 09:53:12.0 +02:00
    }
    +apDeletedAt: null
    +apTimeoutAt: null
    +visibility: "visible             "
    +createdAt: DateTimeImmutable @1699780272 {#272
      date: 2023-11-12 10:11:12.0 +01:00
    }
  }
  +image: Proxies\__CG__\App\Entity\Image {#1943 …}
  +domain: Proxies\__CG__\App\Entity\Domain {#1912 …}
  +slug: "Google-Researchers-Attack-Prompts-ChatGPT-to-Reveal-Its-Training-Data"
  +title: "Google Researchers’ Attack Prompts ChatGPT to Reveal Its Training Data"
  +url: "https://www.404media.co/google-researchers-attack-convinces-chatgpt-to-reveal-its-training-data/"
  +body: """
    > ChatGPT is full of sensitive private information and spits out verbatim text from CNN, Goodreads, WordPress blogs, fandom wikis, Terms of Service agreements, Stack Overflow source code, Wikipedia pages, news blogs, random internet comments, and much more.\n
    \n
    > Using this tactic, the researchers showed that there are large amounts of privately identifiable information (PII) in OpenAI’s large language models. They also showed that, on a public version of ChatGPT, the chatbot spit out large passages of text scraped verbatim from other places on the internet.\n
    \n
    > “In total, 16.9 percent of generations we tested contained memorized PII,” they wrote, which included “identifying phone and fax numbers, email and physical addresses … social media handles, URLs, and names and birthdays.”\n
    \n
    Edit: [The full paper that's referenced in the article can be found here](https://arxiv.org/pdf/2311.17035.pdf)
    """
  +type: "link"
  +lang: "en"
  +isOc: false
  +hasEmbed: false
  +commentCount: 80
  +favouriteCount: 364
  +score: 0
  +isAdult: false
  +sticky: false
  +lastActive: DateTime @1701712043 {#2410
    date: 2023-12-04 18:47:23.0 +01:00
  }
  +ip: null
  +adaAmount: 0
  +tags: null
  +mentions: null
  +comments: Doctrine\ORM\PersistentCollection {#1866 …}
  +votes: Doctrine\ORM\PersistentCollection {#1966 …}
  +reports: Doctrine\ORM\PersistentCollection {#1946 …}
  +favourites: Doctrine\ORM\PersistentCollection {#1368 …}
  +notifications: Doctrine\ORM\PersistentCollection {#2426 …}
  +badges: Doctrine\ORM\PersistentCollection {#2439 …}
  +children: []
  -id: 18235
  -titleTs: "'attack':3 'chatgpt':5 'data':10 'googl':1 'prompt':4 'research':2 'reveal':7 'train':9"
  -bodyTs: "'/pdf/2311.17035.pdf)':134 '16.9':88 'address':109 'agreement':23 'also':61 'amount':48 'articl':127 'arxiv.org':133 'arxiv.org/pdf/2311.17035.pdf)':132 'birthday':117 'blog':17,31 'chatbot':71 'chatgpt':1,69 'cnn':14 'code':27 'comment':34 'contain':94 'edit':118 'email':106 'fandom':18 'fax':104 'found':130 'full':3,120 'generat':91 'goodread':15 'handl':112 'identifi':51,101 'includ':100 'inform':7,52 'internet':33,85 'languag':58 'larg':47,57,74 'media':111 'memor':95 'model':59 'much':36 'name':115 'news':30 'number':105 'openai':55 'overflow':25 'page':29 'paper':121 'passag':75 'percent':89 'phone':102 'physic':108 'pii':53,96 'place':82 'privat':6,50 'public':66 'random':32 'referenc':124 'research':42 'scrape':78 'sensit':5 'servic':22 'show':43,62 'social':110 'sourc':26 'spit':9,72 'stack':24 'tactic':40 'term':20 'test':93 'text':12,77 'total':87 'url':113 'use':38 'verbatim':11,79 'version':67 'wiki':19 'wikipedia':28 'wordpress':16 'wrote':98"
  +cross: false
  +upVotes: 0
  +downVotes: 0
  +ranking: 1701362787
  +visibility: "visible             "
  +apId: "https://kbin.social/m/privacy@lemmy.ml/t/665071"
  +editedAt: DateTimeImmutable @1702265006 {#1850
    date: 2023-12-11 04:23:26.0 +01:00
  }
  +createdAt: DateTimeImmutable @1701276387 {#2420
    date: 2023-11-29 17:46:27.0 +01:00
  }
}
"App\Security\Voter\UserVoter"
App\Entity\Entry {#2412
  +user: Proxies\__CG__\App\Entity\User {#1970 …}
  +magazine: App\Entity\Magazine {#266
    +icon: Proxies\__CG__\App\Entity\Image {#247 …}
    +name: "privacy@lemmy.ml"
    +title: "privacy"
    +description: """
      A place to discuss privacy and freedom in the digital world.\n
      ============================================================\n
      \n
      Privacy has become a very important issue in modern society, with companies and governments constantly abusing their power, more and more people are waking up to the importance of digital privacy.\n
      \n
      In this community everyone is welcome to post links and discuss topics related to privacy.\n
      \n
      ### Some Rules\n
      \n
      - Posting a link to a website containing tracking isn’t great, if contents of the website are behind a paywall maybe copy them into the post\n
      - Don’t promote proprietary software\n
      - Try to keep things on topic\n
      - If you have a question, please try searching for previous discussions, maybe it has already been answered\n
      - Reposts are fine, but should have at least a couple of weeks in between so that the post can reach a new audience\n
      - Be nice :)\n
      \n
      ### Related communities\n
      \n
      - [Lemmy.ml libre_culture](https://lemmy.ml/c/libre_culture)\n
      - [Lemmy.ml privatelife](https://lemmy.ml/c/privatelife)\n
      - [Lemmy.ml DeGoogle](https://lemmy.ml/c/degoogle)\n
      - [Lemmy.ca privacy](https://lemmy.ca/c/privacy)\n
      \n
      ### Chat rooms\n
      \n
      - [Matrix/Element]Dead\n
      - [Discord](https://discord.gg/8xfrVjvngh)\n
      \n
      much thanks to @gary_host_laptop for the logo design :)
      """
    +rules: null
    +subscriptionsCount: 1
    +entryCount: 561
    +entryCommentCount: 7684
    +postCount: 6
    +postCommentCount: 73
    +isAdult: false
    +customCss: null
    +lastActive: DateTime @1729582580 {#276
      date: 2024-10-22 09:36:20.0 +02:00
    }
    +markedForDeletionAt: null
    +tags: null
    +moderators: Doctrine\ORM\PersistentCollection {#238 …}
    +ownershipRequests: Doctrine\ORM\PersistentCollection {#234 …}
    +moderatorRequests: Doctrine\ORM\PersistentCollection {#223 …}
    +entries: Doctrine\ORM\PersistentCollection {#181 …}
    +posts: Doctrine\ORM\PersistentCollection {#139 …}
    +subscriptions: Doctrine\ORM\PersistentCollection {#201 …}
    +bans: Doctrine\ORM\PersistentCollection {#118 …}
    +reports: Doctrine\ORM\PersistentCollection {#104 …}
    +badges: Doctrine\ORM\PersistentCollection {#82 …}
    +logs: Doctrine\ORM\PersistentCollection {#72 …}
    +awards: Doctrine\ORM\PersistentCollection {#61 …}
    +categories: Doctrine\ORM\PersistentCollection {#1820 …}
    -id: 85
    +apId: "privacy@lemmy.ml"
    +apProfileId: "https://lemmy.ml/c/privacy"
    +apPublicUrl: "https://lemmy.ml/c/privacy"
    +apFollowersUrl: "https://lemmy.ml/c/privacy/followers"
    +apInboxUrl: "https://lemmy.ml/inbox"
    +apDomain: "lemmy.ml"
    +apPreferredUsername: "privacy"
    +apDiscoverable: true
    +apManuallyApprovesFollowers: null
    +privateKey: null
    +publicKey: null
    +apFetchedAt: DateTime @1729583592 {#270
      date: 2024-10-22 09:53:12.0 +02:00
    }
    +apDeletedAt: null
    +apTimeoutAt: null
    +visibility: "visible             "
    +createdAt: DateTimeImmutable @1699780272 {#272
      date: 2023-11-12 10:11:12.0 +01:00
    }
  }
  +image: Proxies\__CG__\App\Entity\Image {#1943 …}
  +domain: Proxies\__CG__\App\Entity\Domain {#1912 …}
  +slug: "Google-Researchers-Attack-Prompts-ChatGPT-to-Reveal-Its-Training-Data"
  +title: "Google Researchers’ Attack Prompts ChatGPT to Reveal Its Training Data"
  +url: "https://www.404media.co/google-researchers-attack-convinces-chatgpt-to-reveal-its-training-data/"
  +body: """
    > ChatGPT is full of sensitive private information and spits out verbatim text from CNN, Goodreads, WordPress blogs, fandom wikis, Terms of Service agreements, Stack Overflow source code, Wikipedia pages, news blogs, random internet comments, and much more.\n
    \n
    > Using this tactic, the researchers showed that there are large amounts of privately identifiable information (PII) in OpenAI’s large language models. They also showed that, on a public version of ChatGPT, the chatbot spit out large passages of text scraped verbatim from other places on the internet.\n
    \n
    > “In total, 16.9 percent of generations we tested contained memorized PII,” they wrote, which included “identifying phone and fax numbers, email and physical addresses … social media handles, URLs, and names and birthdays.”\n
    \n
    Edit: [The full paper that's referenced in the article can be found here](https://arxiv.org/pdf/2311.17035.pdf)
    """
  +type: "link"
  +lang: "en"
  +isOc: false
  +hasEmbed: false
  +commentCount: 80
  +favouriteCount: 364
  +score: 0
  +isAdult: false
  +sticky: false
  +lastActive: DateTime @1701712043 {#2410
    date: 2023-12-04 18:47:23.0 +01:00
  }
  +ip: null
  +adaAmount: 0
  +tags: null
  +mentions: null
  +comments: Doctrine\ORM\PersistentCollection {#1866 …}
  +votes: Doctrine\ORM\PersistentCollection {#1966 …}
  +reports: Doctrine\ORM\PersistentCollection {#1946 …}
  +favourites: Doctrine\ORM\PersistentCollection {#1368 …}
  +notifications: Doctrine\ORM\PersistentCollection {#2426 …}
  +badges: Doctrine\ORM\PersistentCollection {#2439 …}
  +children: []
  -id: 18235
  -titleTs: "'attack':3 'chatgpt':5 'data':10 'googl':1 'prompt':4 'research':2 'reveal':7 'train':9"
  -bodyTs: "'/pdf/2311.17035.pdf)':134 '16.9':88 'address':109 'agreement':23 'also':61 'amount':48 'articl':127 'arxiv.org':133 'arxiv.org/pdf/2311.17035.pdf)':132 'birthday':117 'blog':17,31 'chatbot':71 'chatgpt':1,69 'cnn':14 'code':27 'comment':34 'contain':94 'edit':118 'email':106 'fandom':18 'fax':104 'found':130 'full':3,120 'generat':91 'goodread':15 'handl':112 'identifi':51,101 'includ':100 'inform':7,52 'internet':33,85 'languag':58 'larg':47,57,74 'media':111 'memor':95 'model':59 'much':36 'name':115 'news':30 'number':105 'openai':55 'overflow':25 'page':29 'paper':121 'passag':75 'percent':89 'phone':102 'physic':108 'pii':53,96 'place':82 'privat':6,50 'public':66 'random':32 'referenc':124 'research':42 'scrape':78 'sensit':5 'servic':22 'show':43,62 'social':110 'sourc':26 'spit':9,72 'stack':24 'tactic':40 'term':20 'test':93 'text':12,77 'total':87 'url':113 'use':38 'verbatim':11,79 'version':67 'wiki':19 'wikipedia':28 'wordpress':16 'wrote':98"
  +cross: false
  +upVotes: 0
  +downVotes: 0
  +ranking: 1701362787
  +visibility: "visible             "
  +apId: "https://kbin.social/m/privacy@lemmy.ml/t/665071"
  +editedAt: DateTimeImmutable @1702265006 {#1850
    date: 2023-12-11 04:23:26.0 +01:00
  }
  +createdAt: DateTimeImmutable @1701276387 {#2420
    date: 2023-11-29 17:46:27.0 +01:00
  }
}
"App\Security\Voter\UserVoter"
null
"App\Security\Voter\UserVoter"
App\Entity\EntryComment {#4034
  +user: App\Entity\User {#3982 …}
  +entry: App\Entity\Entry {#2412
    +user: Proxies\__CG__\App\Entity\User {#1970 …}
    +magazine: App\Entity\Magazine {#266
      +icon: Proxies\__CG__\App\Entity\Image {#247 …}
      +name: "privacy@lemmy.ml"
      +title: "privacy"
      +description: """
        A place to discuss privacy and freedom in the digital world.\n
        ============================================================\n
        \n
        Privacy has become a very important issue in modern society, with companies and governments constantly abusing their power, more and more people are waking up to the importance of digital privacy.\n
        \n
        In this community everyone is welcome to post links and discuss topics related to privacy.\n
        \n
        ### Some Rules\n
        \n
        - Posting a link to a website containing tracking isn’t great, if contents of the website are behind a paywall maybe copy them into the post\n
        - Don’t promote proprietary software\n
        - Try to keep things on topic\n
        - If you have a question, please try searching for previous discussions, maybe it has already been answered\n
        - Reposts are fine, but should have at least a couple of weeks in between so that the post can reach a new audience\n
        - Be nice :)\n
        \n
        ### Related communities\n
        \n
        - [Lemmy.ml libre_culture](https://lemmy.ml/c/libre_culture)\n
        - [Lemmy.ml privatelife](https://lemmy.ml/c/privatelife)\n
        - [Lemmy.ml DeGoogle](https://lemmy.ml/c/degoogle)\n
        - [Lemmy.ca privacy](https://lemmy.ca/c/privacy)\n
        \n
        ### Chat rooms\n
        \n
        - [Matrix/Element]Dead\n
        - [Discord](https://discord.gg/8xfrVjvngh)\n
        \n
        much thanks to @gary_host_laptop for the logo design :)
        """
      +rules: null
      +subscriptionsCount: 1
      +entryCount: 561
      +entryCommentCount: 7684
      +postCount: 6
      +postCommentCount: 73
      +isAdult: false
      +customCss: null
      +lastActive: DateTime @1729582580 {#276
        date: 2024-10-22 09:36:20.0 +02:00
      }
      +markedForDeletionAt: null
      +tags: null
      +moderators: Doctrine\ORM\PersistentCollection {#238 …}
      +ownershipRequests: Doctrine\ORM\PersistentCollection {#234 …}
      +moderatorRequests: Doctrine\ORM\PersistentCollection {#223 …}
      +entries: Doctrine\ORM\PersistentCollection {#181 …}
      +posts: Doctrine\ORM\PersistentCollection {#139 …}
      +subscriptions: Doctrine\ORM\PersistentCollection {#201 …}
      +bans: Doctrine\ORM\PersistentCollection {#118 …}
      +reports: Doctrine\ORM\PersistentCollection {#104 …}
      +badges: Doctrine\ORM\PersistentCollection {#82 …}
      +logs: Doctrine\ORM\PersistentCollection {#72 …}
      +awards: Doctrine\ORM\PersistentCollection {#61 …}
      +categories: Doctrine\ORM\PersistentCollection {#1820 …}
      -id: 85
      +apId: "privacy@lemmy.ml"
      +apProfileId: "https://lemmy.ml/c/privacy"
      +apPublicUrl: "https://lemmy.ml/c/privacy"
      +apFollowersUrl: "https://lemmy.ml/c/privacy/followers"
      +apInboxUrl: "https://lemmy.ml/inbox"
      +apDomain: "lemmy.ml"
      +apPreferredUsername: "privacy"
      +apDiscoverable: true
      +apManuallyApprovesFollowers: null
      +privateKey: null
      +publicKey: null
      +apFetchedAt: DateTime @1729583592 {#270
        date: 2024-10-22 09:53:12.0 +02:00
      }
      +apDeletedAt: null
      +apTimeoutAt: null
      +visibility: "visible             "
      +createdAt: DateTimeImmutable @1699780272 {#272
        date: 2023-11-12 10:11:12.0 +01:00
      }
    }
    +image: Proxies\__CG__\App\Entity\Image {#1943 …}
    +domain: Proxies\__CG__\App\Entity\Domain {#1912 …}
    +slug: "Google-Researchers-Attack-Prompts-ChatGPT-to-Reveal-Its-Training-Data"
    +title: "Google Researchers’ Attack Prompts ChatGPT to Reveal Its Training Data"
    +url: "https://www.404media.co/google-researchers-attack-convinces-chatgpt-to-reveal-its-training-data/"
    +body: """
      > ChatGPT is full of sensitive private information and spits out verbatim text from CNN, Goodreads, WordPress blogs, fandom wikis, Terms of Service agreements, Stack Overflow source code, Wikipedia pages, news blogs, random internet comments, and much more.\n
      \n
      > Using this tactic, the researchers showed that there are large amounts of privately identifiable information (PII) in OpenAI’s large language models. They also showed that, on a public version of ChatGPT, the chatbot spit out large passages of text scraped verbatim from other places on the internet.\n
      \n
      > “In total, 16.9 percent of generations we tested contained memorized PII,” they wrote, which included “identifying phone and fax numbers, email and physical addresses … social media handles, URLs, and names and birthdays.”\n
      \n
      Edit: [The full paper that's referenced in the article can be found here](https://arxiv.org/pdf/2311.17035.pdf)
      """
    +type: "link"
    +lang: "en"
    +isOc: false
    +hasEmbed: false
    +commentCount: 80
    +favouriteCount: 364
    +score: 0
    +isAdult: false
    +sticky: false
    +lastActive: DateTime @1701712043 {#2410
      date: 2023-12-04 18:47:23.0 +01:00
    }
    +ip: null
    +adaAmount: 0
    +tags: null
    +mentions: null
    +comments: Doctrine\ORM\PersistentCollection {#1866 …}
    +votes: Doctrine\ORM\PersistentCollection {#1966 …}
    +reports: Doctrine\ORM\PersistentCollection {#1946 …}
    +favourites: Doctrine\ORM\PersistentCollection {#1368 …}
    +notifications: Doctrine\ORM\PersistentCollection {#2426 …}
    +badges: Doctrine\ORM\PersistentCollection {#2439 …}
    +children: []
    -id: 18235
    -titleTs: "'attack':3 'chatgpt':5 'data':10 'googl':1 'prompt':4 'research':2 'reveal':7 'train':9"
    -bodyTs: "'/pdf/2311.17035.pdf)':134 '16.9':88 'address':109 'agreement':23 'also':61 'amount':48 'articl':127 'arxiv.org':133 'arxiv.org/pdf/2311.17035.pdf)':132 'birthday':117 'blog':17,31 'chatbot':71 'chatgpt':1,69 'cnn':14 'code':27 'comment':34 'contain':94 'edit':118 'email':106 'fandom':18 'fax':104 'found':130 'full':3,120 'generat':91 'goodread':15 'handl':112 'identifi':51,101 'includ':100 'inform':7,52 'internet':33,85 'languag':58 'larg':47,57,74 'media':111 'memor':95 'model':59 'much':36 'name':115 'news':30 'number':105 'openai':55 'overflow':25 'page':29 'paper':121 'passag':75 'percent':89 'phone':102 'physic':108 'pii':53,96 'place':82 'privat':6,50 'public':66 'random':32 'referenc':124 'research':42 'scrape':78 'sensit':5 'servic':22 'show':43,62 'social':110 'sourc':26 'spit':9,72 'stack':24 'tactic':40 'term':20 'test':93 'text':12,77 'total':87 'url':113 'use':38 'verbatim':11,79 'version':67 'wiki':19 'wikipedia':28 'wordpress':16 'wrote':98"
    +cross: false
    +upVotes: 0
    +downVotes: 0
    +ranking: 1701362787
    +visibility: "visible             "
    +apId: "https://kbin.social/m/privacy@lemmy.ml/t/665071"
    +editedAt: DateTimeImmutable @1702265006 {#1850
      date: 2023-12-11 04:23:26.0 +01:00
    }
    +createdAt: DateTimeImmutable @1701276387 {#2420
      date: 2023-11-29 17:46:27.0 +01:00
    }
  }
  +magazine: App\Entity\Magazine {#266}
  +image: null
  +parent: null
  +root: null
  +body: "Soo plagiarism essentially?"
  +lang: "en"
  +isAdult: false
  +favouriteCount: 18
  +score: 0
  +lastActive: DateTime @1702480724 {#3948
    date: 2023-12-13 16:18:44.0 +01:00
  }
  +ip: null
  +tags: null
  +mentions: [
    "@btp@kbin.social"
  ]
  +children: Doctrine\ORM\PersistentCollection {#4032 …}
  +nested: Doctrine\ORM\PersistentCollection {#4030 …}
  +votes: Doctrine\ORM\PersistentCollection {#4028 …}
  +reports: Doctrine\ORM\PersistentCollection {#4026 …}
  +favourites: Doctrine\ORM\PersistentCollection {#3994 …}
  +notifications: Doctrine\ORM\PersistentCollection {#3998 …}
  -id: 179495
  -bodyTs: "'essenti':3 'plagiar':2 'soo':1"
  +ranking: 0
  +commentCount: 0
  +upVotes: 0
  +downVotes: 0
  +visibility: "visible             "
  +apId: "https://aussie.zone/comment/4729498"
  +editedAt: null
  +createdAt: DateTimeImmutable @1701313651 {#3984
    date: 2023-11-30 04:07:31.0 +01:00
  }
}
"App\Security\Voter\UserVoter"
App\Entity\EntryComment {#4034
  +user: App\Entity\User {#3982 …}
  +entry: App\Entity\Entry {#2412
    +user: Proxies\__CG__\App\Entity\User {#1970 …}
    +magazine: App\Entity\Magazine {#266
      +icon: Proxies\__CG__\App\Entity\Image {#247 …}
      +name: "privacy@lemmy.ml"
      +title: "privacy"
      +description: """
        A place to discuss privacy and freedom in the digital world.\n
        ============================================================\n
        \n
        Privacy has become a very important issue in modern society, with companies and governments constantly abusing their power, more and more people are waking up to the importance of digital privacy.\n
        \n
        In this community everyone is welcome to post links and discuss topics related to privacy.\n
        \n
        ### Some Rules\n
        \n
        - Posting a link to a website containing tracking isn’t great, if contents of the website are behind a paywall maybe copy them into the post\n
        - Don’t promote proprietary software\n
        - Try to keep things on topic\n
        - If you have a question, please try searching for previous discussions, maybe it has already been answered\n
        - Reposts are fine, but should have at least a couple of weeks in between so that the post can reach a new audience\n
        - Be nice :)\n
        \n
        ### Related communities\n
        \n
        - [Lemmy.ml libre_culture](https://lemmy.ml/c/libre_culture)\n
        - [Lemmy.ml privatelife](https://lemmy.ml/c/privatelife)\n
        - [Lemmy.ml DeGoogle](https://lemmy.ml/c/degoogle)\n
        - [Lemmy.ca privacy](https://lemmy.ca/c/privacy)\n
        \n
        ### Chat rooms\n
        \n
        - [Matrix/Element]Dead\n
        - [Discord](https://discord.gg/8xfrVjvngh)\n
        \n
        much thanks to @gary_host_laptop for the logo design :)
        """
      +rules: null
      +subscriptionsCount: 1
      +entryCount: 561
      +entryCommentCount: 7684
      +postCount: 6
      +postCommentCount: 73
      +isAdult: false
      +customCss: null
      +lastActive: DateTime @1729582580 {#276
        date: 2024-10-22 09:36:20.0 +02:00
      }
      +markedForDeletionAt: null
      +tags: null
      +moderators: Doctrine\ORM\PersistentCollection {#238 …}
      +ownershipRequests: Doctrine\ORM\PersistentCollection {#234 …}
      +moderatorRequests: Doctrine\ORM\PersistentCollection {#223 …}
      +entries: Doctrine\ORM\PersistentCollection {#181 …}
      +posts: Doctrine\ORM\PersistentCollection {#139 …}
      +subscriptions: Doctrine\ORM\PersistentCollection {#201 …}
      +bans: Doctrine\ORM\PersistentCollection {#118 …}
      +reports: Doctrine\ORM\PersistentCollection {#104 …}
      +badges: Doctrine\ORM\PersistentCollection {#82 …}
      +logs: Doctrine\ORM\PersistentCollection {#72 …}
      +awards: Doctrine\ORM\PersistentCollection {#61 …}
      +categories: Doctrine\ORM\PersistentCollection {#1820 …}
      -id: 85
      +apId: "privacy@lemmy.ml"
      +apProfileId: "https://lemmy.ml/c/privacy"
      +apPublicUrl: "https://lemmy.ml/c/privacy"
      +apFollowersUrl: "https://lemmy.ml/c/privacy/followers"
      +apInboxUrl: "https://lemmy.ml/inbox"
      +apDomain: "lemmy.ml"
      +apPreferredUsername: "privacy"
      +apDiscoverable: true
      +apManuallyApprovesFollowers: null
      +privateKey: null
      +publicKey: null
      +apFetchedAt: DateTime @1729583592 {#270
        date: 2024-10-22 09:53:12.0 +02:00
      }
      +apDeletedAt: null
      +apTimeoutAt: null
      +visibility: "visible             "
      +createdAt: DateTimeImmutable @1699780272 {#272
        date: 2023-11-12 10:11:12.0 +01:00
      }
    }
    +image: Proxies\__CG__\App\Entity\Image {#1943 …}
    +domain: Proxies\__CG__\App\Entity\Domain {#1912 …}
    +slug: "Google-Researchers-Attack-Prompts-ChatGPT-to-Reveal-Its-Training-Data"
    +title: "Google Researchers’ Attack Prompts ChatGPT to Reveal Its Training Data"
    +url: "https://www.404media.co/google-researchers-attack-convinces-chatgpt-to-reveal-its-training-data/"
    +body: """
      > ChatGPT is full of sensitive private information and spits out verbatim text from CNN, Goodreads, WordPress blogs, fandom wikis, Terms of Service agreements, Stack Overflow source code, Wikipedia pages, news blogs, random internet comments, and much more.\n
      \n
      > Using this tactic, the researchers showed that there are large amounts of privately identifiable information (PII) in OpenAI’s large language models. They also showed that, on a public version of ChatGPT, the chatbot spit out large passages of text scraped verbatim from other places on the internet.\n
      \n
      > “In total, 16.9 percent of generations we tested contained memorized PII,” they wrote, which included “identifying phone and fax numbers, email and physical addresses … social media handles, URLs, and names and birthdays.”\n
      \n
      Edit: [The full paper that's referenced in the article can be found here](https://arxiv.org/pdf/2311.17035.pdf)
      """
    +type: "link"
    +lang: "en"
    +isOc: false
    +hasEmbed: false
    +commentCount: 80
    +favouriteCount: 364
    +score: 0
    +isAdult: false
    +sticky: false
    +lastActive: DateTime @1701712043 {#2410
      date: 2023-12-04 18:47:23.0 +01:00
    }
    +ip: null
    +adaAmount: 0
    +tags: null
    +mentions: null
    +comments: Doctrine\ORM\PersistentCollection {#1866 …}
    +votes: Doctrine\ORM\PersistentCollection {#1966 …}
    +reports: Doctrine\ORM\PersistentCollection {#1946 …}
    +favourites: Doctrine\ORM\PersistentCollection {#1368 …}
    +notifications: Doctrine\ORM\PersistentCollection {#2426 …}
    +badges: Doctrine\ORM\PersistentCollection {#2439 …}
    +children: []
    -id: 18235
    -titleTs: "'attack':3 'chatgpt':5 'data':10 'googl':1 'prompt':4 'research':2 'reveal':7 'train':9"
    -bodyTs: "'/pdf/2311.17035.pdf)':134 '16.9':88 'address':109 'agreement':23 'also':61 'amount':48 'articl':127 'arxiv.org':133 'arxiv.org/pdf/2311.17035.pdf)':132 'birthday':117 'blog':17,31 'chatbot':71 'chatgpt':1,69 'cnn':14 'code':27 'comment':34 'contain':94 'edit':118 'email':106 'fandom':18 'fax':104 'found':130 'full':3,120 'generat':91 'goodread':15 'handl':112 'identifi':51,101 'includ':100 'inform':7,52 'internet':33,85 'languag':58 'larg':47,57,74 'media':111 'memor':95 'model':59 'much':36 'name':115 'news':30 'number':105 'openai':55 'overflow':25 'page':29 'paper':121 'passag':75 'percent':89 'phone':102 'physic':108 'pii':53,96 'place':82 'privat':6,50 'public':66 'random':32 'referenc':124 'research':42 'scrape':78 'sensit':5 'servic':22 'show':43,62 'social':110 'sourc':26 'spit':9,72 'stack':24 'tactic':40 'term':20 'test':93 'text':12,77 'total':87 'url':113 'use':38 'verbatim':11,79 'version':67 'wiki':19 'wikipedia':28 'wordpress':16 'wrote':98"
    +cross: false
    +upVotes: 0
    +downVotes: 0
    +ranking: 1701362787
    +visibility: "visible             "
    +apId: "https://kbin.social/m/privacy@lemmy.ml/t/665071"
    +editedAt: DateTimeImmutable @1702265006 {#1850
      date: 2023-12-11 04:23:26.0 +01:00
    }
    +createdAt: DateTimeImmutable @1701276387 {#2420
      date: 2023-11-29 17:46:27.0 +01:00
    }
  }
  +magazine: App\Entity\Magazine {#266}
  +image: null
  +parent: null
  +root: null
  +body: "Soo plagiarism essentially?"
  +lang: "en"
  +isAdult: false
  +favouriteCount: 18
  +score: 0
  +lastActive: DateTime @1702480724 {#3948
    date: 2023-12-13 16:18:44.0 +01:00
  }
  +ip: null
  +tags: null
  +mentions: [
    "@btp@kbin.social"
  ]
  +children: Doctrine\ORM\PersistentCollection {#4032 …}
  +nested: Doctrine\ORM\PersistentCollection {#4030 …}
  +votes: Doctrine\ORM\PersistentCollection {#4028 …}
  +reports: Doctrine\ORM\PersistentCollection {#4026 …}
  +favourites: Doctrine\ORM\PersistentCollection {#3994 …}
  +notifications: Doctrine\ORM\PersistentCollection {#3998 …}
  -id: 179495
  -bodyTs: "'essenti':3 'plagiar':2 'soo':1"
  +ranking: 0
  +commentCount: 0
  +upVotes: 0
  +downVotes: 0
  +visibility: "visible             "
  +apId: "https://aussie.zone/comment/4729498"
  +editedAt: null
  +createdAt: DateTimeImmutable @1701313651 {#3984
    date: 2023-11-30 04:07:31.0 +01:00
  }
}
"App\Security\Voter\UserVoter"
App\Entity\EntryComment {#4034
  +user: App\Entity\User {#3982 …}
  +entry: App\Entity\Entry {#2412
    +user: Proxies\__CG__\App\Entity\User {#1970 …}
    +magazine: App\Entity\Magazine {#266
      +icon: Proxies\__CG__\App\Entity\Image {#247 …}
      +name: "privacy@lemmy.ml"
      +title: "privacy"
      +description: """
        A place to discuss privacy and freedom in the digital world.\n
        ============================================================\n
        \n
        Privacy has become a very important issue in modern society, with companies and governments constantly abusing their power, more and more people are waking up to the importance of digital privacy.\n
        \n
        In this community everyone is welcome to post links and discuss topics related to privacy.\n
        \n
        ### Some Rules\n
        \n
        - Posting a link to a website containing tracking isn’t great, if contents of the website are behind a paywall maybe copy them into the post\n
        - Don’t promote proprietary software\n
        - Try to keep things on topic\n
        - If you have a question, please try searching for previous discussions, maybe it has already been answered\n
        - Reposts are fine, but should have at least a couple of weeks in between so that the post can reach a new audience\n
        - Be nice :)\n
        \n
        ### Related communities\n
        \n
        - [Lemmy.ml libre_culture](https://lemmy.ml/c/libre_culture)\n
        - [Lemmy.ml privatelife](https://lemmy.ml/c/privatelife)\n
        - [Lemmy.ml DeGoogle](https://lemmy.ml/c/degoogle)\n
        - [Lemmy.ca privacy](https://lemmy.ca/c/privacy)\n
        \n
        ### Chat rooms\n
        \n
        - [Matrix/Element]Dead\n
        - [Discord](https://discord.gg/8xfrVjvngh)\n
        \n
        much thanks to @gary_host_laptop for the logo design :)
        """
      +rules: null
      +subscriptionsCount: 1
      +entryCount: 561
      +entryCommentCount: 7684
      +postCount: 6
      +postCommentCount: 73
      +isAdult: false
      +customCss: null
      +lastActive: DateTime @1729582580 {#276
        date: 2024-10-22 09:36:20.0 +02:00
      }
      +markedForDeletionAt: null
      +tags: null
      +moderators: Doctrine\ORM\PersistentCollection {#238 …}
      +ownershipRequests: Doctrine\ORM\PersistentCollection {#234 …}
      +moderatorRequests: Doctrine\ORM\PersistentCollection {#223 …}
      +entries: Doctrine\ORM\PersistentCollection {#181 …}
      +posts: Doctrine\ORM\PersistentCollection {#139 …}
      +subscriptions: Doctrine\ORM\PersistentCollection {#201 …}
      +bans: Doctrine\ORM\PersistentCollection {#118 …}
      +reports: Doctrine\ORM\PersistentCollection {#104 …}
      +badges: Doctrine\ORM\PersistentCollection {#82 …}
      +logs: Doctrine\ORM\PersistentCollection {#72 …}
      +awards: Doctrine\ORM\PersistentCollection {#61 …}
      +categories: Doctrine\ORM\PersistentCollection {#1820 …}
      -id: 85
      +apId: "privacy@lemmy.ml"
      +apProfileId: "https://lemmy.ml/c/privacy"
      +apPublicUrl: "https://lemmy.ml/c/privacy"
      +apFollowersUrl: "https://lemmy.ml/c/privacy/followers"
      +apInboxUrl: "https://lemmy.ml/inbox"
      +apDomain: "lemmy.ml"
      +apPreferredUsername: "privacy"
      +apDiscoverable: true
      +apManuallyApprovesFollowers: null
      +privateKey: null
      +publicKey: null
      +apFetchedAt: DateTime @1729583592 {#270
        date: 2024-10-22 09:53:12.0 +02:00
      }
      +apDeletedAt: null
      +apTimeoutAt: null
      +visibility: "visible             "
      +createdAt: DateTimeImmutable @1699780272 {#272
        date: 2023-11-12 10:11:12.0 +01:00
      }
    }
    +image: Proxies\__CG__\App\Entity\Image {#1943 …}
    +domain: Proxies\__CG__\App\Entity\Domain {#1912 …}
    +slug: "Google-Researchers-Attack-Prompts-ChatGPT-to-Reveal-Its-Training-Data"
    +title: "Google Researchers’ Attack Prompts ChatGPT to Reveal Its Training Data"
    +url: "https://www.404media.co/google-researchers-attack-convinces-chatgpt-to-reveal-its-training-data/"
    +body: """
      > ChatGPT is full of sensitive private information and spits out verbatim text from CNN, Goodreads, WordPress blogs, fandom wikis, Terms of Service agreements, Stack Overflow source code, Wikipedia pages, news blogs, random internet comments, and much more.\n
      \n
      > Using this tactic, the researchers showed that there are large amounts of privately identifiable information (PII) in OpenAI’s large language models. They also showed that, on a public version of ChatGPT, the chatbot spit out large passages of text scraped verbatim from other places on the internet.\n
      \n
      > “In total, 16.9 percent of generations we tested contained memorized PII,” they wrote, which included “identifying phone and fax numbers, email and physical addresses … social media handles, URLs, and names and birthdays.”\n
      \n
      Edit: [The full paper that's referenced in the article can be found here](https://arxiv.org/pdf/2311.17035.pdf)
      """
    +type: "link"
    +lang: "en"
    +isOc: false
    +hasEmbed: false
    +commentCount: 80
    +favouriteCount: 364
    +score: 0
    +isAdult: false
    +sticky: false
    +lastActive: DateTime @1701712043 {#2410
      date: 2023-12-04 18:47:23.0 +01:00
    }
    +ip: null
    +adaAmount: 0
    +tags: null
    +mentions: null
    +comments: Doctrine\ORM\PersistentCollection {#1866 …}
    +votes: Doctrine\ORM\PersistentCollection {#1966 …}
    +reports: Doctrine\ORM\PersistentCollection {#1946 …}
    +favourites: Doctrine\ORM\PersistentCollection {#1368 …}
    +notifications: Doctrine\ORM\PersistentCollection {#2426 …}
    +badges: Doctrine\ORM\PersistentCollection {#2439 …}
    +children: []
    -id: 18235
    -titleTs: "'attack':3 'chatgpt':5 'data':10 'googl':1 'prompt':4 'research':2 'reveal':7 'train':9"
    -bodyTs: "'/pdf/2311.17035.pdf)':134 '16.9':88 'address':109 'agreement':23 'also':61 'amount':48 'articl':127 'arxiv.org':133 'arxiv.org/pdf/2311.17035.pdf)':132 'birthday':117 'blog':17,31 'chatbot':71 'chatgpt':1,69 'cnn':14 'code':27 'comment':34 'contain':94 'edit':118 'email':106 'fandom':18 'fax':104 'found':130 'full':3,120 'generat':91 'goodread':15 'handl':112 'identifi':51,101 'includ':100 'inform':7,52 'internet':33,85 'languag':58 'larg':47,57,74 'media':111 'memor':95 'model':59 'much':36 'name':115 'news':30 'number':105 'openai':55 'overflow':25 'page':29 'paper':121 'passag':75 'percent':89 'phone':102 'physic':108 'pii':53,96 'place':82 'privat':6,50 'public':66 'random':32 'referenc':124 'research':42 'scrape':78 'sensit':5 'servic':22 'show':43,62 'social':110 'sourc':26 'spit':9,72 'stack':24 'tactic':40 'term':20 'test':93 'text':12,77 'total':87 'url':113 'use':38 'verbatim':11,79 'version':67 'wiki':19 'wikipedia':28 'wordpress':16 'wrote':98"
    +cross: false
    +upVotes: 0
    +downVotes: 0
    +ranking: 1701362787
    +visibility: "visible             "
    +apId: "https://kbin.social/m/privacy@lemmy.ml/t/665071"
    +editedAt: DateTimeImmutable @1702265006 {#1850
      date: 2023-12-11 04:23:26.0 +01:00
    }
    +createdAt: DateTimeImmutable @1701276387 {#2420
      date: 2023-11-29 17:46:27.0 +01:00
    }
  }
  +magazine: App\Entity\Magazine {#266}
  +image: null
  +parent: null
  +root: null
  +body: "Soo plagiarism essentially?"
  +lang: "en"
  +isAdult: false
  +favouriteCount: 18
  +score: 0
  +lastActive: DateTime @1702480724 {#3948
    date: 2023-12-13 16:18:44.0 +01:00
  }
  +ip: null
  +tags: null
  +mentions: [
    "@btp@kbin.social"
  ]
  +children: Doctrine\ORM\PersistentCollection {#4032 …}
  +nested: Doctrine\ORM\PersistentCollection {#4030 …}
  +votes: Doctrine\ORM\PersistentCollection {#4028 …}
  +reports: Doctrine\ORM\PersistentCollection {#4026 …}
  +favourites: Doctrine\ORM\PersistentCollection {#3994 …}
  +notifications: Doctrine\ORM\PersistentCollection {#3998 …}
  -id: 179495
  -bodyTs: "'essenti':3 'plagiar':2 'soo':1"
  +ranking: 0
  +commentCount: 0
  +upVotes: 0
  +downVotes: 0
  +visibility: "visible             "
  +apId: "https://aussie.zone/comment/4729498"
  +editedAt: null
  +createdAt: DateTimeImmutable @1701313651 {#3984
    date: 2023-11-30 04:07:31.0 +01:00
  }
}
"App\Security\Voter\UserVoter"
null
"App\Security\Voter\UserVoter"
App\Entity\EntryComment {#4251
  +user: Proxies\__CG__\App\Entity\User {#4255 …}
  +entry: App\Entity\Entry {#2412
    +user: Proxies\__CG__\App\Entity\User {#1970 …}
    +magazine: App\Entity\Magazine {#266
      +icon: Proxies\__CG__\App\Entity\Image {#247 …}
      +name: "privacy@lemmy.ml"
      +title: "privacy"
      +description: """
        A place to discuss privacy and freedom in the digital world.\n
        ============================================================\n
        \n
        Privacy has become a very important issue in modern society, with companies and governments constantly abusing their power, more and more people are waking up to the importance of digital privacy.\n
        \n
        In this community everyone is welcome to post links and discuss topics related to privacy.\n
        \n
        ### Some Rules\n
        \n
        - Posting a link to a website containing tracking isn’t great, if contents of the website are behind a paywall maybe copy them into the post\n
        - Don’t promote proprietary software\n
        - Try to keep things on topic\n
        - If you have a question, please try searching for previous discussions, maybe it has already been answered\n
        - Reposts are fine, but should have at least a couple of weeks in between so that the post can reach a new audience\n
        - Be nice :)\n
        \n
        ### Related communities\n
        \n
        - [Lemmy.ml libre_culture](https://lemmy.ml/c/libre_culture)\n
        - [Lemmy.ml privatelife](https://lemmy.ml/c/privatelife)\n
        - [Lemmy.ml DeGoogle](https://lemmy.ml/c/degoogle)\n
        - [Lemmy.ca privacy](https://lemmy.ca/c/privacy)\n
        \n
        ### Chat rooms\n
        \n
        - [Matrix/Element]Dead\n
        - [Discord](https://discord.gg/8xfrVjvngh)\n
        \n
        much thanks to @gary_host_laptop for the logo design :)
        """
      +rules: null
      +subscriptionsCount: 1
      +entryCount: 561
      +entryCommentCount: 7684
      +postCount: 6
      +postCommentCount: 73
      +isAdult: false
      +customCss: null
      +lastActive: DateTime @1729582580 {#276
        date: 2024-10-22 09:36:20.0 +02:00
      }
      +markedForDeletionAt: null
      +tags: null
      +moderators: Doctrine\ORM\PersistentCollection {#238 …}
      +ownershipRequests: Doctrine\ORM\PersistentCollection {#234 …}
      +moderatorRequests: Doctrine\ORM\PersistentCollection {#223 …}
      +entries: Doctrine\ORM\PersistentCollection {#181 …}
      +posts: Doctrine\ORM\PersistentCollection {#139 …}
      +subscriptions: Doctrine\ORM\PersistentCollection {#201 …}
      +bans: Doctrine\ORM\PersistentCollection {#118 …}
      +reports: Doctrine\ORM\PersistentCollection {#104 …}
      +badges: Doctrine\ORM\PersistentCollection {#82 …}
      +logs: Doctrine\ORM\PersistentCollection {#72 …}
      +awards: Doctrine\ORM\PersistentCollection {#61 …}
      +categories: Doctrine\ORM\PersistentCollection {#1820 …}
      -id: 85
      +apId: "privacy@lemmy.ml"
      +apProfileId: "https://lemmy.ml/c/privacy"
      +apPublicUrl: "https://lemmy.ml/c/privacy"
      +apFollowersUrl: "https://lemmy.ml/c/privacy/followers"
      +apInboxUrl: "https://lemmy.ml/inbox"
      +apDomain: "lemmy.ml"
      +apPreferredUsername: "privacy"
      +apDiscoverable: true
      +apManuallyApprovesFollowers: null
      +privateKey: null
      +publicKey: null
      +apFetchedAt: DateTime @1729583592 {#270
        date: 2024-10-22 09:53:12.0 +02:00
      }
      +apDeletedAt: null
      +apTimeoutAt: null
      +visibility: "visible             "
      +createdAt: DateTimeImmutable @1699780272 {#272
        date: 2023-11-12 10:11:12.0 +01:00
      }
    }
    +image: Proxies\__CG__\App\Entity\Image {#1943 …}
    +domain: Proxies\__CG__\App\Entity\Domain {#1912 …}
    +slug: "Google-Researchers-Attack-Prompts-ChatGPT-to-Reveal-Its-Training-Data"
    +title: "Google Researchers’ Attack Prompts ChatGPT to Reveal Its Training Data"
    +url: "https://www.404media.co/google-researchers-attack-convinces-chatgpt-to-reveal-its-training-data/"
    +body: """
      > ChatGPT is full of sensitive private information and spits out verbatim text from CNN, Goodreads, WordPress blogs, fandom wikis, Terms of Service agreements, Stack Overflow source code, Wikipedia pages, news blogs, random internet comments, and much more.\n
      \n
      > Using this tactic, the researchers showed that there are large amounts of privately identifiable information (PII) in OpenAI’s large language models. They also showed that, on a public version of ChatGPT, the chatbot spit out large passages of text scraped verbatim from other places on the internet.\n
      \n
      > “In total, 16.9 percent of generations we tested contained memorized PII,” they wrote, which included “identifying phone and fax numbers, email and physical addresses … social media handles, URLs, and names and birthdays.”\n
      \n
      Edit: [The full paper that's referenced in the article can be found here](https://arxiv.org/pdf/2311.17035.pdf)
      """
    +type: "link"
    +lang: "en"
    +isOc: false
    +hasEmbed: false
    +commentCount: 80
    +favouriteCount: 364
    +score: 0
    +isAdult: false
    +sticky: false
    +lastActive: DateTime @1701712043 {#2410
      date: 2023-12-04 18:47:23.0 +01:00
    }
    +ip: null
    +adaAmount: 0
    +tags: null
    +mentions: null
    +comments: Doctrine\ORM\PersistentCollection {#1866 …}
    +votes: Doctrine\ORM\PersistentCollection {#1966 …}
    +reports: Doctrine\ORM\PersistentCollection {#1946 …}
    +favourites: Doctrine\ORM\PersistentCollection {#1368 …}
    +notifications: Doctrine\ORM\PersistentCollection {#2426 …}
    +badges: Doctrine\ORM\PersistentCollection {#2439 …}
    +children: []
    -id: 18235
    -titleTs: "'attack':3 'chatgpt':5 'data':10 'googl':1 'prompt':4 'research':2 'reveal':7 'train':9"
    -bodyTs: "'/pdf/2311.17035.pdf)':134 '16.9':88 'address':109 'agreement':23 'also':61 'amount':48 'articl':127 'arxiv.org':133 'arxiv.org/pdf/2311.17035.pdf)':132 'birthday':117 'blog':17,31 'chatbot':71 'chatgpt':1,69 'cnn':14 'code':27 'comment':34 'contain':94 'edit':118 'email':106 'fandom':18 'fax':104 'found':130 'full':3,120 'generat':91 'goodread':15 'handl':112 'identifi':51,101 'includ':100 'inform':7,52 'internet':33,85 'languag':58 'larg':47,57,74 'media':111 'memor':95 'model':59 'much':36 'name':115 'news':30 'number':105 'openai':55 'overflow':25 'page':29 'paper':121 'passag':75 'percent':89 'phone':102 'physic':108 'pii':53,96 'place':82 'privat':6,50 'public':66 'random':32 'referenc':124 'research':42 'scrape':78 'sensit':5 'servic':22 'show':43,62 'social':110 'sourc':26 'spit':9,72 'stack':24 'tactic':40 'term':20 'test':93 'text':12,77 'total':87 'url':113 'use':38 'verbatim':11,79 'version':67 'wiki':19 'wikipedia':28 'wordpress':16 'wrote':98"
    +cross: false
    +upVotes: 0
    +downVotes: 0
    +ranking: 1701362787
    +visibility: "visible             "
    +apId: "https://kbin.social/m/privacy@lemmy.ml/t/665071"
    +editedAt: DateTimeImmutable @1702265006 {#1850
      date: 2023-12-11 04:23:26.0 +01:00
    }
    +createdAt: DateTimeImmutable @1701276387 {#2420
      date: 2023-11-29 17:46:27.0 +01:00
    }
  }
  +magazine: App\Entity\Magazine {#266}
  +image: null
  +parent: App\Entity\EntryComment {#4034
    +user: App\Entity\User {#3982 …}
    +entry: App\Entity\Entry {#2412}
    +magazine: App\Entity\Magazine {#266}
    +image: null
    +parent: null
    +root: null
    +body: "Soo plagiarism essentially?"
    +lang: "en"
    +isAdult: false
    +favouriteCount: 18
    +score: 0
    +lastActive: DateTime @1702480724 {#3948
      date: 2023-12-13 16:18:44.0 +01:00
    }
    +ip: null
    +tags: null
    +mentions: [
      "@btp@kbin.social"
    ]
    +children: Doctrine\ORM\PersistentCollection {#4032 …}
    +nested: Doctrine\ORM\PersistentCollection {#4030 …}
    +votes: Doctrine\ORM\PersistentCollection {#4028 …}
    +reports: Doctrine\ORM\PersistentCollection {#4026 …}
    +favourites: Doctrine\ORM\PersistentCollection {#3994 …}
    +notifications: Doctrine\ORM\PersistentCollection {#3998 …}
    -id: 179495
    -bodyTs: "'essenti':3 'plagiar':2 'soo':1"
    +ranking: 0
    +commentCount: 0
    +upVotes: 0
    +downVotes: 0
    +visibility: "visible             "
    +apId: "https://aussie.zone/comment/4729498"
    +editedAt: null
    +createdAt: DateTimeImmutable @1701313651 {#3984
      date: 2023-11-30 04:07:31.0 +01:00
    }
  }
  +root: App\Entity\EntryComment {#4034}
  +body: """
    Always has been. Just yesterday I was explaining AI image generation to a coworker. I said the program looks at a ton of images and uses that info to blend them together. Like it knows what a soviet propaganda poster looks like, and it knows what artwork of Santa looks like so it can make a Santa themed propaganda poster.\n
    \n
    Same with text I assume. It knows the Mario wiki and fanfics, and it knows a bunch of books about zombies so it blends it to make a gritty story about Mario fending off zombies. But yeah it’s all other works just melded together.\n
    \n
    My question is would a human author be any different? We absorb ideas and stories we read and hear and blend them into new or reimagined ideas. AI just knows it’s original sources
    """
  +lang: "en"
  +isAdult: false
  +favouriteCount: 8
  +score: 0
  +lastActive: DateTime @1701349985 {#4247
    date: 2023-11-30 14:13:05.0 +01:00
  }
  +ip: null
  +tags: null
  +mentions: [
    "@btp@kbin.social"
    "@Nonameuser678@aussie.zone"
  ]
  +children: Doctrine\ORM\PersistentCollection {#4256 …}
  +nested: Doctrine\ORM\PersistentCollection {#4252 …}
  +votes: Doctrine\ORM\PersistentCollection {#4258 …}
  +reports: Doctrine\ORM\PersistentCollection {#4260 …}
  +favourites: Doctrine\ORM\PersistentCollection {#4262 …}
  +notifications: Doctrine\ORM\PersistentCollection {#4264 …}
  -id: 183662
  -bodyTs: "'absorb':117 'ai':9,133 'alway':1 'artwork':47 'assum':65 'author':112 'blend':30,84,126 'book':79 'bunch':77 'cowork':14 'differ':115 'explain':8 'fanfic':72 'fend':93 'generat':11 'gritti':89 'hear':124 'human':111 'idea':118,132 'imag':10,24 'info':28 'know':35,45,67,75,135 'like':33,42,51 'look':19,41,50 'make':55,87 'mario':69,92 'meld':104 'new':129 'origin':138 'poster':40,60 'program':18 'propaganda':39,59 'question':107 'read':122 'reimagin':131 'said':16 'santa':49,57 'sourc':139 'soviet':38 'stori':90,120 'text':63 'theme':58 'togeth':32,105 'ton':22 'use':26 'wiki':70 'work':102 'would':109 'yeah':97 'yesterday':5 'zombi':81,95"
  +ranking: 0
  +commentCount: 0
  +upVotes: 0
  +downVotes: 0
  +visibility: "visible             "
  +apId: "https://sh.itjust.works/comment/6129297"
  +editedAt: DateTimeImmutable @1702479959 {#4244
    date: 2023-12-13 16:05:59.0 +01:00
  }
  +createdAt: DateTimeImmutable @1701349985 {#4248
    date: 2023-11-30 14:13:05.0 +01:00
  }
}
"App\Security\Voter\UserVoter"
App\Entity\EntryComment {#4251
  +user: Proxies\__CG__\App\Entity\User {#4255 …}
  +entry: App\Entity\Entry {#2412
    +user: Proxies\__CG__\App\Entity\User {#1970 …}
    +magazine: App\Entity\Magazine {#266
      +icon: Proxies\__CG__\App\Entity\Image {#247 …}
      +name: "privacy@lemmy.ml"
      +title: "privacy"
      +description: """
        A place to discuss privacy and freedom in the digital world.\n
        ============================================================\n
        \n
        Privacy has become a very important issue in modern society, with companies and governments constantly abusing their power, more and more people are waking up to the importance of digital privacy.\n
        \n
        In this community everyone is welcome to post links and discuss topics related to privacy.\n
        \n
        ### Some Rules\n
        \n
        - Posting a link to a website containing tracking isn’t great, if contents of the website are behind a paywall maybe copy them into the post\n
        - Don’t promote proprietary software\n
        - Try to keep things on topic\n
        - If you have a question, please try searching for previous discussions, maybe it has already been answered\n
        - Reposts are fine, but should have at least a couple of weeks in between so that the post can reach a new audience\n
        - Be nice :)\n
        \n
        ### Related communities\n
        \n
        - [Lemmy.ml libre_culture](https://lemmy.ml/c/libre_culture)\n
        - [Lemmy.ml privatelife](https://lemmy.ml/c/privatelife)\n
        - [Lemmy.ml DeGoogle](https://lemmy.ml/c/degoogle)\n
        - [Lemmy.ca privacy](https://lemmy.ca/c/privacy)\n
        \n
        ### Chat rooms\n
        \n
        - [Matrix/Element]Dead\n
        - [Discord](https://discord.gg/8xfrVjvngh)\n
        \n
        much thanks to @gary_host_laptop for the logo design :)
        """
      +rules: null
      +subscriptionsCount: 1
      +entryCount: 561
      +entryCommentCount: 7684
      +postCount: 6
      +postCommentCount: 73
      +isAdult: false
      +customCss: null
      +lastActive: DateTime @1729582580 {#276
        date: 2024-10-22 09:36:20.0 +02:00
      }
      +markedForDeletionAt: null
      +tags: null
      +moderators: Doctrine\ORM\PersistentCollection {#238 …}
      +ownershipRequests: Doctrine\ORM\PersistentCollection {#234 …}
      +moderatorRequests: Doctrine\ORM\PersistentCollection {#223 …}
      +entries: Doctrine\ORM\PersistentCollection {#181 …}
      +posts: Doctrine\ORM\PersistentCollection {#139 …}
      +subscriptions: Doctrine\ORM\PersistentCollection {#201 …}
      +bans: Doctrine\ORM\PersistentCollection {#118 …}
      +reports: Doctrine\ORM\PersistentCollection {#104 …}
      +badges: Doctrine\ORM\PersistentCollection {#82 …}
      +logs: Doctrine\ORM\PersistentCollection {#72 …}
      +awards: Doctrine\ORM\PersistentCollection {#61 …}
      +categories: Doctrine\ORM\PersistentCollection {#1820 …}
      -id: 85
      +apId: "privacy@lemmy.ml"
      +apProfileId: "https://lemmy.ml/c/privacy"
      +apPublicUrl: "https://lemmy.ml/c/privacy"
      +apFollowersUrl: "https://lemmy.ml/c/privacy/followers"
      +apInboxUrl: "https://lemmy.ml/inbox"
      +apDomain: "lemmy.ml"
      +apPreferredUsername: "privacy"
      +apDiscoverable: true
      +apManuallyApprovesFollowers: null
      +privateKey: null
      +publicKey: null
      +apFetchedAt: DateTime @1729583592 {#270
        date: 2024-10-22 09:53:12.0 +02:00
      }
      +apDeletedAt: null
      +apTimeoutAt: null
      +visibility: "visible             "
      +createdAt: DateTimeImmutable @1699780272 {#272
        date: 2023-11-12 10:11:12.0 +01:00
      }
    }
    +image: Proxies\__CG__\App\Entity\Image {#1943 …}
    +domain: Proxies\__CG__\App\Entity\Domain {#1912 …}
    +slug: "Google-Researchers-Attack-Prompts-ChatGPT-to-Reveal-Its-Training-Data"
    +title: "Google Researchers’ Attack Prompts ChatGPT to Reveal Its Training Data"
    +url: "https://www.404media.co/google-researchers-attack-convinces-chatgpt-to-reveal-its-training-data/"
    +body: """
      > ChatGPT is full of sensitive private information and spits out verbatim text from CNN, Goodreads, WordPress blogs, fandom wikis, Terms of Service agreements, Stack Overflow source code, Wikipedia pages, news blogs, random internet comments, and much more.\n
      \n
      > Using this tactic, the researchers showed that there are large amounts of privately identifiable information (PII) in OpenAI’s large language models. They also showed that, on a public version of ChatGPT, the chatbot spit out large passages of text scraped verbatim from other places on the internet.\n
      \n
      > “In total, 16.9 percent of generations we tested contained memorized PII,” they wrote, which included “identifying phone and fax numbers, email and physical addresses … social media handles, URLs, and names and birthdays.”\n
      \n
      Edit: [The full paper that's referenced in the article can be found here](https://arxiv.org/pdf/2311.17035.pdf)
      """
    +type: "link"
    +lang: "en"
    +isOc: false
    +hasEmbed: false
    +commentCount: 80
    +favouriteCount: 364
    +score: 0
    +isAdult: false
    +sticky: false
    +lastActive: DateTime @1701712043 {#2410
      date: 2023-12-04 18:47:23.0 +01:00
    }
    +ip: null
    +adaAmount: 0
    +tags: null
    +mentions: null
    +comments: Doctrine\ORM\PersistentCollection {#1866 …}
    +votes: Doctrine\ORM\PersistentCollection {#1966 …}
    +reports: Doctrine\ORM\PersistentCollection {#1946 …}
    +favourites: Doctrine\ORM\PersistentCollection {#1368 …}
    +notifications: Doctrine\ORM\PersistentCollection {#2426 …}
    +badges: Doctrine\ORM\PersistentCollection {#2439 …}
    +children: []
    -id: 18235
    -titleTs: "'attack':3 'chatgpt':5 'data':10 'googl':1 'prompt':4 'research':2 'reveal':7 'train':9"
    -bodyTs: "'/pdf/2311.17035.pdf)':134 '16.9':88 'address':109 'agreement':23 'also':61 'amount':48 'articl':127 'arxiv.org':133 'arxiv.org/pdf/2311.17035.pdf)':132 'birthday':117 'blog':17,31 'chatbot':71 'chatgpt':1,69 'cnn':14 'code':27 'comment':34 'contain':94 'edit':118 'email':106 'fandom':18 'fax':104 'found':130 'full':3,120 'generat':91 'goodread':15 'handl':112 'identifi':51,101 'includ':100 'inform':7,52 'internet':33,85 'languag':58 'larg':47,57,74 'media':111 'memor':95 'model':59 'much':36 'name':115 'news':30 'number':105 'openai':55 'overflow':25 'page':29 'paper':121 'passag':75 'percent':89 'phone':102 'physic':108 'pii':53,96 'place':82 'privat':6,50 'public':66 'random':32 'referenc':124 'research':42 'scrape':78 'sensit':5 'servic':22 'show':43,62 'social':110 'sourc':26 'spit':9,72 'stack':24 'tactic':40 'term':20 'test':93 'text':12,77 'total':87 'url':113 'use':38 'verbatim':11,79 'version':67 'wiki':19 'wikipedia':28 'wordpress':16 'wrote':98"
    +cross: false
    +upVotes: 0
    +downVotes: 0
    +ranking: 1701362787
    +visibility: "visible             "
    +apId: "https://kbin.social/m/privacy@lemmy.ml/t/665071"
    +editedAt: DateTimeImmutable @1702265006 {#1850
      date: 2023-12-11 04:23:26.0 +01:00
    }
    +createdAt: DateTimeImmutable @1701276387 {#2420
      date: 2023-11-29 17:46:27.0 +01:00
    }
  }
  +magazine: App\Entity\Magazine {#266}
  +image: null
  +parent: App\Entity\EntryComment {#4034
    +user: App\Entity\User {#3982 …}
    +entry: App\Entity\Entry {#2412}
    +magazine: App\Entity\Magazine {#266}
    +image: null
    +parent: null
    +root: null
    +body: "Soo plagiarism essentially?"
    +lang: "en"
    +isAdult: false
    +favouriteCount: 18
    +score: 0
    +lastActive: DateTime @1702480724 {#3948
      date: 2023-12-13 16:18:44.0 +01:00
    }
    +ip: null
    +tags: null
    +mentions: [
      "@btp@kbin.social"
    ]
    +children: Doctrine\ORM\PersistentCollection {#4032 …}
    +nested: Doctrine\ORM\PersistentCollection {#4030 …}
    +votes: Doctrine\ORM\PersistentCollection {#4028 …}
    +reports: Doctrine\ORM\PersistentCollection {#4026 …}
    +favourites: Doctrine\ORM\PersistentCollection {#3994 …}
    +notifications: Doctrine\ORM\PersistentCollection {#3998 …}
    -id: 179495
    -bodyTs: "'essenti':3 'plagiar':2 'soo':1"
    +ranking: 0
    +commentCount: 0
    +upVotes: 0
    +downVotes: 0
    +visibility: "visible             "
    +apId: "https://aussie.zone/comment/4729498"
    +editedAt: null
    +createdAt: DateTimeImmutable @1701313651 {#3984
      date: 2023-11-30 04:07:31.0 +01:00
    }
  }
  +root: App\Entity\EntryComment {#4034}
  +body: """
    Always has been. Just yesterday I was explaining AI image generation to a coworker. I said the program looks at a ton of images and uses that info to blend them together. Like it knows what a soviet propaganda poster looks like, and it knows what artwork of Santa looks like so it can make a Santa themed propaganda poster.\n
    \n
    Same with text I assume. It knows the Mario wiki and fanfics, and it knows a bunch of books about zombies so it blends it to make a gritty story about Mario fending off zombies. But yeah it’s all other works just melded together.\n
    \n
    My question is would a human author be any different? We absorb ideas and stories we read and hear and blend them into new or reimagined ideas. AI just knows it’s original sources
    """
  +lang: "en"
  +isAdult: false
  +favouriteCount: 8
  +score: 0
  +lastActive: DateTime @1701349985 {#4247
    date: 2023-11-30 14:13:05.0 +01:00
  }
  +ip: null
  +tags: null
  +mentions: [
    "@btp@kbin.social"
    "@Nonameuser678@aussie.zone"
  ]
  +children: Doctrine\ORM\PersistentCollection {#4256 …}
  +nested: Doctrine\ORM\PersistentCollection {#4252 …}
  +votes: Doctrine\ORM\PersistentCollection {#4258 …}
  +reports: Doctrine\ORM\PersistentCollection {#4260 …}
  +favourites: Doctrine\ORM\PersistentCollection {#4262 …}
  +notifications: Doctrine\ORM\PersistentCollection {#4264 …}
  -id: 183662
  -bodyTs: "'absorb':117 'ai':9,133 'alway':1 'artwork':47 'assum':65 'author':112 'blend':30,84,126 'book':79 'bunch':77 'cowork':14 'differ':115 'explain':8 'fanfic':72 'fend':93 'generat':11 'gritti':89 'hear':124 'human':111 'idea':118,132 'imag':10,24 'info':28 'know':35,45,67,75,135 'like':33,42,51 'look':19,41,50 'make':55,87 'mario':69,92 'meld':104 'new':129 'origin':138 'poster':40,60 'program':18 'propaganda':39,59 'question':107 'read':122 'reimagin':131 'said':16 'santa':49,57 'sourc':139 'soviet':38 'stori':90,120 'text':63 'theme':58 'togeth':32,105 'ton':22 'use':26 'wiki':70 'work':102 'would':109 'yeah':97 'yesterday':5 'zombi':81,95"
  +ranking: 0
  +commentCount: 0
  +upVotes: 0
  +downVotes: 0
  +visibility: "visible             "
  +apId: "https://sh.itjust.works/comment/6129297"
  +editedAt: DateTimeImmutable @1702479959 {#4244
    date: 2023-12-13 16:05:59.0 +01:00
  }
  +createdAt: DateTimeImmutable @1701349985 {#4248
    date: 2023-11-30 14:13:05.0 +01:00
  }
}
"App\Security\Voter\UserVoter"
App\Entity\EntryComment {#4251
  +user: Proxies\__CG__\App\Entity\User {#4255 …}
  +entry: App\Entity\Entry {#2412
    +user: Proxies\__CG__\App\Entity\User {#1970 …}
    +magazine: App\Entity\Magazine {#266
      +icon: Proxies\__CG__\App\Entity\Image {#247 …}
      +name: "privacy@lemmy.ml"
      +title: "privacy"
      +description: """
        A place to discuss privacy and freedom in the digital world.\n
        ============================================================\n
        \n
        Privacy has become a very important issue in modern society, with companies and governments constantly abusing their power, more and more people are waking up to the importance of digital privacy.\n
        \n
        In this community everyone is welcome to post links and discuss topics related to privacy.\n
        \n
        ### Some Rules\n
        \n
        - Posting a link to a website containing tracking isn’t great, if contents of the website are behind a paywall maybe copy them into the post\n
        - Don’t promote proprietary software\n
        - Try to keep things on topic\n
        - If you have a question, please try searching for previous discussions, maybe it has already been answered\n
        - Reposts are fine, but should have at least a couple of weeks in between so that the post can reach a new audience\n
        - Be nice :)\n
        \n
        ### Related communities\n
        \n
        - [Lemmy.ml libre_culture](https://lemmy.ml/c/libre_culture)\n
        - [Lemmy.ml privatelife](https://lemmy.ml/c/privatelife)\n
        - [Lemmy.ml DeGoogle](https://lemmy.ml/c/degoogle)\n
        - [Lemmy.ca privacy](https://lemmy.ca/c/privacy)\n
        \n
        ### Chat rooms\n
        \n
        - [Matrix/Element]Dead\n
        - [Discord](https://discord.gg/8xfrVjvngh)\n
        \n
        much thanks to @gary_host_laptop for the logo design :)
        """
      +rules: null
      +subscriptionsCount: 1
      +entryCount: 561
      +entryCommentCount: 7684
      +postCount: 6
      +postCommentCount: 73
      +isAdult: false
      +customCss: null
      +lastActive: DateTime @1729582580 {#276
        date: 2024-10-22 09:36:20.0 +02:00
      }
      +markedForDeletionAt: null
      +tags: null
      +moderators: Doctrine\ORM\PersistentCollection {#238 …}
      +ownershipRequests: Doctrine\ORM\PersistentCollection {#234 …}
      +moderatorRequests: Doctrine\ORM\PersistentCollection {#223 …}
      +entries: Doctrine\ORM\PersistentCollection {#181 …}
      +posts: Doctrine\ORM\PersistentCollection {#139 …}
      +subscriptions: Doctrine\ORM\PersistentCollection {#201 …}
      +bans: Doctrine\ORM\PersistentCollection {#118 …}
      +reports: Doctrine\ORM\PersistentCollection {#104 …}
      +badges: Doctrine\ORM\PersistentCollection {#82 …}
      +logs: Doctrine\ORM\PersistentCollection {#72 …}
      +awards: Doctrine\ORM\PersistentCollection {#61 …}
      +categories: Doctrine\ORM\PersistentCollection {#1820 …}
      -id: 85
      +apId: "privacy@lemmy.ml"
      +apProfileId: "https://lemmy.ml/c/privacy"
      +apPublicUrl: "https://lemmy.ml/c/privacy"
      +apFollowersUrl: "https://lemmy.ml/c/privacy/followers"
      +apInboxUrl: "https://lemmy.ml/inbox"
      +apDomain: "lemmy.ml"
      +apPreferredUsername: "privacy"
      +apDiscoverable: true
      +apManuallyApprovesFollowers: null
      +privateKey: null
      +publicKey: null
      +apFetchedAt: DateTime @1729583592 {#270
        date: 2024-10-22 09:53:12.0 +02:00
      }
      +apDeletedAt: null
      +apTimeoutAt: null
      +visibility: "visible             "
      +createdAt: DateTimeImmutable @1699780272 {#272
        date: 2023-11-12 10:11:12.0 +01:00
      }
    }
    +image: Proxies\__CG__\App\Entity\Image {#1943 …}
    +domain: Proxies\__CG__\App\Entity\Domain {#1912 …}
    +slug: "Google-Researchers-Attack-Prompts-ChatGPT-to-Reveal-Its-Training-Data"
    +title: "Google Researchers’ Attack Prompts ChatGPT to Reveal Its Training Data"
    +url: "https://www.404media.co/google-researchers-attack-convinces-chatgpt-to-reveal-its-training-data/"
    +body: """
      > ChatGPT is full of sensitive private information and spits out verbatim text from CNN, Goodreads, WordPress blogs, fandom wikis, Terms of Service agreements, Stack Overflow source code, Wikipedia pages, news blogs, random internet comments, and much more.\n
      \n
      > Using this tactic, the researchers showed that there are large amounts of privately identifiable information (PII) in OpenAI’s large language models. They also showed that, on a public version of ChatGPT, the chatbot spit out large passages of text scraped verbatim from other places on the internet.\n
      \n
      > “In total, 16.9 percent of generations we tested contained memorized PII,” they wrote, which included “identifying phone and fax numbers, email and physical addresses … social media handles, URLs, and names and birthdays.”\n
      \n
      Edit: [The full paper that's referenced in the article can be found here](https://arxiv.org/pdf/2311.17035.pdf)
      """
    +type: "link"
    +lang: "en"
    +isOc: false
    +hasEmbed: false
    +commentCount: 80
    +favouriteCount: 364
    +score: 0
    +isAdult: false
    +sticky: false
    +lastActive: DateTime @1701712043 {#2410
      date: 2023-12-04 18:47:23.0 +01:00
    }
    +ip: null
    +adaAmount: 0
    +tags: null
    +mentions: null
    +comments: Doctrine\ORM\PersistentCollection {#1866 …}
    +votes: Doctrine\ORM\PersistentCollection {#1966 …}
    +reports: Doctrine\ORM\PersistentCollection {#1946 …}
    +favourites: Doctrine\ORM\PersistentCollection {#1368 …}
    +notifications: Doctrine\ORM\PersistentCollection {#2426 …}
    +badges: Doctrine\ORM\PersistentCollection {#2439 …}
    +children: []
    -id: 18235
    -titleTs: "'attack':3 'chatgpt':5 'data':10 'googl':1 'prompt':4 'research':2 'reveal':7 'train':9"
    -bodyTs: "'/pdf/2311.17035.pdf)':134 '16.9':88 'address':109 'agreement':23 'also':61 'amount':48 'articl':127 'arxiv.org':133 'arxiv.org/pdf/2311.17035.pdf)':132 'birthday':117 'blog':17,31 'chatbot':71 'chatgpt':1,69 'cnn':14 'code':27 'comment':34 'contain':94 'edit':118 'email':106 'fandom':18 'fax':104 'found':130 'full':3,120 'generat':91 'goodread':15 'handl':112 'identifi':51,101 'includ':100 'inform':7,52 'internet':33,85 'languag':58 'larg':47,57,74 'media':111 'memor':95 'model':59 'much':36 'name':115 'news':30 'number':105 'openai':55 'overflow':25 'page':29 'paper':121 'passag':75 'percent':89 'phone':102 'physic':108 'pii':53,96 'place':82 'privat':6,50 'public':66 'random':32 'referenc':124 'research':42 'scrape':78 'sensit':5 'servic':22 'show':43,62 'social':110 'sourc':26 'spit':9,72 'stack':24 'tactic':40 'term':20 'test':93 'text':12,77 'total':87 'url':113 'use':38 'verbatim':11,79 'version':67 'wiki':19 'wikipedia':28 'wordpress':16 'wrote':98"
    +cross: false
    +upVotes: 0
    +downVotes: 0
    +ranking: 1701362787
    +visibility: "visible             "
    +apId: "https://kbin.social/m/privacy@lemmy.ml/t/665071"
    +editedAt: DateTimeImmutable @1702265006 {#1850
      date: 2023-12-11 04:23:26.0 +01:00
    }
    +createdAt: DateTimeImmutable @1701276387 {#2420
      date: 2023-11-29 17:46:27.0 +01:00
    }
  }
  +magazine: App\Entity\Magazine {#266}
  +image: null
  +parent: App\Entity\EntryComment {#4034
    +user: App\Entity\User {#3982 …}
    +entry: App\Entity\Entry {#2412}
    +magazine: App\Entity\Magazine {#266}
    +image: null
    +parent: null
    +root: null
    +body: "Soo plagiarism essentially?"
    +lang: "en"
    +isAdult: false
    +favouriteCount: 18
    +score: 0
    +lastActive: DateTime @1702480724 {#3948
      date: 2023-12-13 16:18:44.0 +01:00
    }
    +ip: null
    +tags: null
    +mentions: [
      "@btp@kbin.social"
    ]
    +children: Doctrine\ORM\PersistentCollection {#4032 …}
    +nested: Doctrine\ORM\PersistentCollection {#4030 …}
    +votes: Doctrine\ORM\PersistentCollection {#4028 …}
    +reports: Doctrine\ORM\PersistentCollection {#4026 …}
    +favourites: Doctrine\ORM\PersistentCollection {#3994 …}
    +notifications: Doctrine\ORM\PersistentCollection {#3998 …}
    -id: 179495
    -bodyTs: "'essenti':3 'plagiar':2 'soo':1"
    +ranking: 0
    +commentCount: 0
    +upVotes: 0
    +downVotes: 0
    +visibility: "visible             "
    +apId: "https://aussie.zone/comment/4729498"
    +editedAt: null
    +createdAt: DateTimeImmutable @1701313651 {#3984
      date: 2023-11-30 04:07:31.0 +01:00
    }
  }
  +root: App\Entity\EntryComment {#4034}
  +body: """
    Always has been. Just yesterday I was explaining AI image generation to a coworker. I said the program looks at a ton of images and uses that info to blend them together. Like it knows what a soviet propaganda poster looks like, and it knows what artwork of Santa looks like so it can make a Santa themed propaganda poster.\n
    \n
    Same with text I assume. It knows the Mario wiki and fanfics, and it knows a bunch of books about zombies so it blends it to make a gritty story about Mario fending off zombies. But yeah it’s all other works just melded together.\n
    \n
    My question is would a human author be any different? We absorb ideas and stories we read and hear and blend them into new or reimagined ideas. AI just knows it’s original sources
    """
  +lang: "en"
  +isAdult: false
  +favouriteCount: 8
  +score: 0
  +lastActive: DateTime @1701349985 {#4247
    date: 2023-11-30 14:13:05.0 +01:00
  }
  +ip: null
  +tags: null
  +mentions: [
    "@btp@kbin.social"
    "@Nonameuser678@aussie.zone"
  ]
  +children: Doctrine\ORM\PersistentCollection {#4256 …}
  +nested: Doctrine\ORM\PersistentCollection {#4252 …}
  +votes: Doctrine\ORM\PersistentCollection {#4258 …}
  +reports: Doctrine\ORM\PersistentCollection {#4260 …}
  +favourites: Doctrine\ORM\PersistentCollection {#4262 …}
  +notifications: Doctrine\ORM\PersistentCollection {#4264 …}
  -id: 183662
  -bodyTs: "'absorb':117 'ai':9,133 'alway':1 'artwork':47 'assum':65 'author':112 'blend':30,84,126 'book':79 'bunch':77 'cowork':14 'differ':115 'explain':8 'fanfic':72 'fend':93 'generat':11 'gritti':89 'hear':124 'human':111 'idea':118,132 'imag':10,24 'info':28 'know':35,45,67,75,135 'like':33,42,51 'look':19,41,50 'make':55,87 'mario':69,92 'meld':104 'new':129 'origin':138 'poster':40,60 'program':18 'propaganda':39,59 'question':107 'read':122 'reimagin':131 'said':16 'santa':49,57 'sourc':139 'soviet':38 'stori':90,120 'text':63 'theme':58 'togeth':32,105 'ton':22 'use':26 'wiki':70 'work':102 'would':109 'yeah':97 'yesterday':5 'zombi':81,95"
  +ranking: 0
  +commentCount: 0
  +upVotes: 0
  +downVotes: 0
  +visibility: "visible             "
  +apId: "https://sh.itjust.works/comment/6129297"
  +editedAt: DateTimeImmutable @1702479959 {#4244
    date: 2023-12-13 16:05:59.0 +01:00
  }
  +createdAt: DateTimeImmutable @1701349985 {#4248
    date: 2023-11-30 14:13:05.0 +01:00
  }
}
"App\Security\Voter\UserVoter"
null
"App\Security\Voter\UserVoter"
App\Entity\EntryComment {#4284
  +user: Proxies\__CG__\App\Entity\User {#4292 …}
  +entry: App\Entity\Entry {#2412
    +user: Proxies\__CG__\App\Entity\User {#1970 …}
    +magazine: App\Entity\Magazine {#266
      +icon: Proxies\__CG__\App\Entity\Image {#247 …}
      +name: "privacy@lemmy.ml"
      +title: "privacy"
      +description: """
        A place to discuss privacy and freedom in the digital world.\n
        ============================================================\n
        \n
        Privacy has become a very important issue in modern society, with companies and governments constantly abusing their power, more and more people are waking up to the importance of digital privacy.\n
        \n
        In this community everyone is welcome to post links and discuss topics related to privacy.\n
        \n
        ### Some Rules\n
        \n
        - Posting a link to a website containing tracking isn’t great, if contents of the website are behind a paywall maybe copy them into the post\n
        - Don’t promote proprietary software\n
        - Try to keep things on topic\n
        - If you have a question, please try searching for previous discussions, maybe it has already been answered\n
        - Reposts are fine, but should have at least a couple of weeks in between so that the post can reach a new audience\n
        - Be nice :)\n
        \n
        ### Related communities\n
        \n
        - [Lemmy.ml libre_culture](https://lemmy.ml/c/libre_culture)\n
        - [Lemmy.ml privatelife](https://lemmy.ml/c/privatelife)\n
        - [Lemmy.ml DeGoogle](https://lemmy.ml/c/degoogle)\n
        - [Lemmy.ca privacy](https://lemmy.ca/c/privacy)\n
        \n
        ### Chat rooms\n
        \n
        - [Matrix/Element]Dead\n
        - [Discord](https://discord.gg/8xfrVjvngh)\n
        \n
        much thanks to @gary_host_laptop for the logo design :)
        """
      +rules: null
      +subscriptionsCount: 1
      +entryCount: 561
      +entryCommentCount: 7684
      +postCount: 6
      +postCommentCount: 73
      +isAdult: false
      +customCss: null
      +lastActive: DateTime @1729582580 {#276
        date: 2024-10-22 09:36:20.0 +02:00
      }
      +markedForDeletionAt: null
      +tags: null
      +moderators: Doctrine\ORM\PersistentCollection {#238 …}
      +ownershipRequests: Doctrine\ORM\PersistentCollection {#234 …}
      +moderatorRequests: Doctrine\ORM\PersistentCollection {#223 …}
      +entries: Doctrine\ORM\PersistentCollection {#181 …}
      +posts: Doctrine\ORM\PersistentCollection {#139 …}
      +subscriptions: Doctrine\ORM\PersistentCollection {#201 …}
      +bans: Doctrine\ORM\PersistentCollection {#118 …}
      +reports: Doctrine\ORM\PersistentCollection {#104 …}
      +badges: Doctrine\ORM\PersistentCollection {#82 …}
      +logs: Doctrine\ORM\PersistentCollection {#72 …}
      +awards: Doctrine\ORM\PersistentCollection {#61 …}
      +categories: Doctrine\ORM\PersistentCollection {#1820 …}
      -id: 85
      +apId: "privacy@lemmy.ml"
      +apProfileId: "https://lemmy.ml/c/privacy"
      +apPublicUrl: "https://lemmy.ml/c/privacy"
      +apFollowersUrl: "https://lemmy.ml/c/privacy/followers"
      +apInboxUrl: "https://lemmy.ml/inbox"
      +apDomain: "lemmy.ml"
      +apPreferredUsername: "privacy"
      +apDiscoverable: true
      +apManuallyApprovesFollowers: null
      +privateKey: null
      +publicKey: null
      +apFetchedAt: DateTime @1729583592 {#270
        date: 2024-10-22 09:53:12.0 +02:00
      }
      +apDeletedAt: null
      +apTimeoutAt: null
      +visibility: "visible             "
      +createdAt: DateTimeImmutable @1699780272 {#272
        date: 2023-11-12 10:11:12.0 +01:00
      }
    }
    +image: Proxies\__CG__\App\Entity\Image {#1943 …}
    +domain: Proxies\__CG__\App\Entity\Domain {#1912 …}
    +slug: "Google-Researchers-Attack-Prompts-ChatGPT-to-Reveal-Its-Training-Data"
    +title: "Google Researchers’ Attack Prompts ChatGPT to Reveal Its Training Data"
    +url: "https://www.404media.co/google-researchers-attack-convinces-chatgpt-to-reveal-its-training-data/"
    +body: """
      > ChatGPT is full of sensitive private information and spits out verbatim text from CNN, Goodreads, WordPress blogs, fandom wikis, Terms of Service agreements, Stack Overflow source code, Wikipedia pages, news blogs, random internet comments, and much more.\n
      \n
      > Using this tactic, the researchers showed that there are large amounts of privately identifiable information (PII) in OpenAI’s large language models. They also showed that, on a public version of ChatGPT, the chatbot spit out large passages of text scraped verbatim from other places on the internet.\n
      \n
      > “In total, 16.9 percent of generations we tested contained memorized PII,” they wrote, which included “identifying phone and fax numbers, email and physical addresses … social media handles, URLs, and names and birthdays.”\n
      \n
      Edit: [The full paper that's referenced in the article can be found here](https://arxiv.org/pdf/2311.17035.pdf)
      """
    +type: "link"
    +lang: "en"
    +isOc: false
    +hasEmbed: false
    +commentCount: 80
    +favouriteCount: 364
    +score: 0
    +isAdult: false
    +sticky: false
    +lastActive: DateTime @1701712043 {#2410
      date: 2023-12-04 18:47:23.0 +01:00
    }
    +ip: null
    +adaAmount: 0
    +tags: null
    +mentions: null
    +comments: Doctrine\ORM\PersistentCollection {#1866 …}
    +votes: Doctrine\ORM\PersistentCollection {#1966 …}
    +reports: Doctrine\ORM\PersistentCollection {#1946 …}
    +favourites: Doctrine\ORM\PersistentCollection {#1368 …}
    +notifications: Doctrine\ORM\PersistentCollection {#2426 …}
    +badges: Doctrine\ORM\PersistentCollection {#2439 …}
    +children: []
    -id: 18235
    -titleTs: "'attack':3 'chatgpt':5 'data':10 'googl':1 'prompt':4 'research':2 'reveal':7 'train':9"
    -bodyTs: "'/pdf/2311.17035.pdf)':134 '16.9':88 'address':109 'agreement':23 'also':61 'amount':48 'articl':127 'arxiv.org':133 'arxiv.org/pdf/2311.17035.pdf)':132 'birthday':117 'blog':17,31 'chatbot':71 'chatgpt':1,69 'cnn':14 'code':27 'comment':34 'contain':94 'edit':118 'email':106 'fandom':18 'fax':104 'found':130 'full':3,120 'generat':91 'goodread':15 'handl':112 'identifi':51,101 'includ':100 'inform':7,52 'internet':33,85 'languag':58 'larg':47,57,74 'media':111 'memor':95 'model':59 'much':36 'name':115 'news':30 'number':105 'openai':55 'overflow':25 'page':29 'paper':121 'passag':75 'percent':89 'phone':102 'physic':108 'pii':53,96 'place':82 'privat':6,50 'public':66 'random':32 'referenc':124 'research':42 'scrape':78 'sensit':5 'servic':22 'show':43,62 'social':110 'sourc':26 'spit':9,72 'stack':24 'tactic':40 'term':20 'test':93 'text':12,77 'total':87 'url':113 'use':38 'verbatim':11,79 'version':67 'wiki':19 'wikipedia':28 'wordpress':16 'wrote':98"
    +cross: false
    +upVotes: 0
    +downVotes: 0
    +ranking: 1701362787
    +visibility: "visible             "
    +apId: "https://kbin.social/m/privacy@lemmy.ml/t/665071"
    +editedAt: DateTimeImmutable @1702265006 {#1850
      date: 2023-12-11 04:23:26.0 +01:00
    }
    +createdAt: DateTimeImmutable @1701276387 {#2420
      date: 2023-11-29 17:46:27.0 +01:00
    }
  }
  +magazine: App\Entity\Magazine {#266}
  +image: null
  +parent: App\Entity\EntryComment {#4251
    +user: Proxies\__CG__\App\Entity\User {#4255 …}
    +entry: App\Entity\Entry {#2412}
    +magazine: App\Entity\Magazine {#266}
    +image: null
    +parent: App\Entity\EntryComment {#4034
      +user: App\Entity\User {#3982 …}
      +entry: App\Entity\Entry {#2412}
      +magazine: App\Entity\Magazine {#266}
      +image: null
      +parent: null
      +root: null
      +body: "Soo plagiarism essentially?"
      +lang: "en"
      +isAdult: false
      +favouriteCount: 18
      +score: 0
      +lastActive: DateTime @1702480724 {#3948
        date: 2023-12-13 16:18:44.0 +01:00
      }
      +ip: null
      +tags: null
      +mentions: [
        "@btp@kbin.social"
      ]
      +children: Doctrine\ORM\PersistentCollection {#4032 …}
      +nested: Doctrine\ORM\PersistentCollection {#4030 …}
      +votes: Doctrine\ORM\PersistentCollection {#4028 …}
      +reports: Doctrine\ORM\PersistentCollection {#4026 …}
      +favourites: Doctrine\ORM\PersistentCollection {#3994 …}
      +notifications: Doctrine\ORM\PersistentCollection {#3998 …}
      -id: 179495
      -bodyTs: "'essenti':3 'plagiar':2 'soo':1"
      +ranking: 0
      +commentCount: 0
      +upVotes: 0
      +downVotes: 0
      +visibility: "visible             "
      +apId: "https://aussie.zone/comment/4729498"
      +editedAt: null
      +createdAt: DateTimeImmutable @1701313651 {#3984
        date: 2023-11-30 04:07:31.0 +01:00
      }
    }
    +root: App\Entity\EntryComment {#4034}
    +body: """
      Always has been. Just yesterday I was explaining AI image generation to a coworker. I said the program looks at a ton of images and uses that info to blend them together. Like it knows what a soviet propaganda poster looks like, and it knows what artwork of Santa looks like so it can make a Santa themed propaganda poster.\n
      \n
      Same with text I assume. It knows the Mario wiki and fanfics, and it knows a bunch of books about zombies so it blends it to make a gritty story about Mario fending off zombies. But yeah it’s all other works just melded together.\n
      \n
      My question is would a human author be any different? We absorb ideas and stories we read and hear and blend them into new or reimagined ideas. AI just knows it’s original sources
      """
    +lang: "en"
    +isAdult: false
    +favouriteCount: 8
    +score: 0
    +lastActive: DateTime @1701349985 {#4247
      date: 2023-11-30 14:13:05.0 +01:00
    }
    +ip: null
    +tags: null
    +mentions: [
      "@btp@kbin.social"
      "@Nonameuser678@aussie.zone"
    ]
    +children: Doctrine\ORM\PersistentCollection {#4256 …}
    +nested: Doctrine\ORM\PersistentCollection {#4252 …}
    +votes: Doctrine\ORM\PersistentCollection {#4258 …}
    +reports: Doctrine\ORM\PersistentCollection {#4260 …}
    +favourites: Doctrine\ORM\PersistentCollection {#4262 …}
    +notifications: Doctrine\ORM\PersistentCollection {#4264 …}
    -id: 183662
    -bodyTs: "'absorb':117 'ai':9,133 'alway':1 'artwork':47 'assum':65 'author':112 'blend':30,84,126 'book':79 'bunch':77 'cowork':14 'differ':115 'explain':8 'fanfic':72 'fend':93 'generat':11 'gritti':89 'hear':124 'human':111 'idea':118,132 'imag':10,24 'info':28 'know':35,45,67,75,135 'like':33,42,51 'look':19,41,50 'make':55,87 'mario':69,92 'meld':104 'new':129 'origin':138 'poster':40,60 'program':18 'propaganda':39,59 'question':107 'read':122 'reimagin':131 'said':16 'santa':49,57 'sourc':139 'soviet':38 'stori':90,120 'text':63 'theme':58 'togeth':32,105 'ton':22 'use':26 'wiki':70 'work':102 'would':109 'yeah':97 'yesterday':5 'zombi':81,95"
    +ranking: 0
    +commentCount: 0
    +upVotes: 0
    +downVotes: 0
    +visibility: "visible             "
    +apId: "https://sh.itjust.works/comment/6129297"
    +editedAt: DateTimeImmutable @1702479959 {#4244
      date: 2023-12-13 16:05:59.0 +01:00
    }
    +createdAt: DateTimeImmutable @1701349985 {#4248
      date: 2023-11-30 14:13:05.0 +01:00
    }
  }
  +root: App\Entity\EntryComment {#4034}
  +body: "“Blending together” isn’t accurate, since it implies that the original images are used in the process of creating the output. The AI doesn’t have access to the original data (if it wasn’t erroneously repeated many times in the training dataset)."
  +lang: "en"
  +isAdult: false
  +favouriteCount: 3
  +score: 0
  +lastActive: DateTime @1701358588 {#4289
    date: 2023-11-30 16:36:28.0 +01:00
  }
  +ip: null
  +tags: null
  +mentions: [
    "@btp@kbin.social"
    "@Nonameuser678@aussie.zone"
    "@SomeAmateur@sh.itjust.works"
  ]
  +children: Doctrine\ORM\PersistentCollection {#4295 …}
  +nested: Doctrine\ORM\PersistentCollection {#4297 …}
  +votes: Doctrine\ORM\PersistentCollection {#4293 …}
  +reports: Doctrine\ORM\PersistentCollection {#4300 …}
  +favourites: Doctrine\ORM\PersistentCollection {#4302 …}
  +notifications: Doctrine\ORM\PersistentCollection {#4304 …}
  -id: 183807
  -bodyTs: "'access':27 'accur':5 'ai':23 'blend':1 'creat':19 'data':31 'dataset':43 'doesn':24 'erron':36 'imag':12 'impli':8 'isn':3 'mani':38 'origin':11,30 'output':21 'process':17 'repeat':37 'sinc':6 'time':39 'togeth':2 'train':42 'use':14 'wasn':34"
  +ranking: 0
  +commentCount: 0
  +upVotes: 0
  +downVotes: 0
  +visibility: "visible             "
  +apId: "https://lemmy.world/comment/5697627"
  +editedAt: null
  +createdAt: DateTimeImmutable @1701358588 {#4286
    date: 2023-11-30 16:36:28.0 +01:00
  }
}
"App\Security\Voter\UserVoter"
App\Entity\EntryComment {#4284
  +user: Proxies\__CG__\App\Entity\User {#4292 …}
  +entry: App\Entity\Entry {#2412
    +user: Proxies\__CG__\App\Entity\User {#1970 …}
    +magazine: App\Entity\Magazine {#266
      +icon: Proxies\__CG__\App\Entity\Image {#247 …}
      +name: "privacy@lemmy.ml"
      +title: "privacy"
      +description: """
        A place to discuss privacy and freedom in the digital world.\n
        ============================================================\n
        \n
        Privacy has become a very important issue in modern society, with companies and governments constantly abusing their power, more and more people are waking up to the importance of digital privacy.\n
        \n
        In this community everyone is welcome to post links and discuss topics related to privacy.\n
        \n
        ### Some Rules\n
        \n
        - Posting a link to a website containing tracking isn’t great, if contents of the website are behind a paywall maybe copy them into the post\n
        - Don’t promote proprietary software\n
        - Try to keep things on topic\n
        - If you have a question, please try searching for previous discussions, maybe it has already been answered\n
        - Reposts are fine, but should have at least a couple of weeks in between so that the post can reach a new audience\n
        - Be nice :)\n
        \n
        ### Related communities\n
        \n
        - [Lemmy.ml libre_culture](https://lemmy.ml/c/libre_culture)\n
        - [Lemmy.ml privatelife](https://lemmy.ml/c/privatelife)\n
        - [Lemmy.ml DeGoogle](https://lemmy.ml/c/degoogle)\n
        - [Lemmy.ca privacy](https://lemmy.ca/c/privacy)\n
        \n
        ### Chat rooms\n
        \n
        - [Matrix/Element]Dead\n
        - [Discord](https://discord.gg/8xfrVjvngh)\n
        \n
        much thanks to @gary_host_laptop for the logo design :)
        """
      +rules: null
      +subscriptionsCount: 1
      +entryCount: 561
      +entryCommentCount: 7684
      +postCount: 6
      +postCommentCount: 73
      +isAdult: false
      +customCss: null
      +lastActive: DateTime @1729582580 {#276
        date: 2024-10-22 09:36:20.0 +02:00
      }
      +markedForDeletionAt: null
      +tags: null
      +moderators: Doctrine\ORM\PersistentCollection {#238 …}
      +ownershipRequests: Doctrine\ORM\PersistentCollection {#234 …}
      +moderatorRequests: Doctrine\ORM\PersistentCollection {#223 …}
      +entries: Doctrine\ORM\PersistentCollection {#181 …}
      +posts: Doctrine\ORM\PersistentCollection {#139 …}
      +subscriptions: Doctrine\ORM\PersistentCollection {#201 …}
      +bans: Doctrine\ORM\PersistentCollection {#118 …}
      +reports: Doctrine\ORM\PersistentCollection {#104 …}
      +badges: Doctrine\ORM\PersistentCollection {#82 …}
      +logs: Doctrine\ORM\PersistentCollection {#72 …}
      +awards: Doctrine\ORM\PersistentCollection {#61 …}
      +categories: Doctrine\ORM\PersistentCollection {#1820 …}
      -id: 85
      +apId: "privacy@lemmy.ml"
      +apProfileId: "https://lemmy.ml/c/privacy"
      +apPublicUrl: "https://lemmy.ml/c/privacy"
      +apFollowersUrl: "https://lemmy.ml/c/privacy/followers"
      +apInboxUrl: "https://lemmy.ml/inbox"
      +apDomain: "lemmy.ml"
      +apPreferredUsername: "privacy"
      +apDiscoverable: true
      +apManuallyApprovesFollowers: null
      +privateKey: null
      +publicKey: null
      +apFetchedAt: DateTime @1729583592 {#270
        date: 2024-10-22 09:53:12.0 +02:00
      }
      +apDeletedAt: null
      +apTimeoutAt: null
      +visibility: "visible             "
      +createdAt: DateTimeImmutable @1699780272 {#272
        date: 2023-11-12 10:11:12.0 +01:00
      }
    }
    +image: Proxies\__CG__\App\Entity\Image {#1943 …}
    +domain: Proxies\__CG__\App\Entity\Domain {#1912 …}
    +slug: "Google-Researchers-Attack-Prompts-ChatGPT-to-Reveal-Its-Training-Data"
    +title: "Google Researchers’ Attack Prompts ChatGPT to Reveal Its Training Data"
    +url: "https://www.404media.co/google-researchers-attack-convinces-chatgpt-to-reveal-its-training-data/"
    +body: """
      > ChatGPT is full of sensitive private information and spits out verbatim text from CNN, Goodreads, WordPress blogs, fandom wikis, Terms of Service agreements, Stack Overflow source code, Wikipedia pages, news blogs, random internet comments, and much more.\n
      \n
      > Using this tactic, the researchers showed that there are large amounts of privately identifiable information (PII) in OpenAI’s large language models. They also showed that, on a public version of ChatGPT, the chatbot spit out large passages of text scraped verbatim from other places on the internet.\n
      \n
      > “In total, 16.9 percent of generations we tested contained memorized PII,” they wrote, which included “identifying phone and fax numbers, email and physical addresses … social media handles, URLs, and names and birthdays.”\n
      \n
      Edit: [The full paper that's referenced in the article can be found here](https://arxiv.org/pdf/2311.17035.pdf)
      """
    +type: "link"
    +lang: "en"
    +isOc: false
    +hasEmbed: false
    +commentCount: 80
    +favouriteCount: 364
    +score: 0
    +isAdult: false
    +sticky: false
    +lastActive: DateTime @1701712043 {#2410
      date: 2023-12-04 18:47:23.0 +01:00
    }
    +ip: null
    +adaAmount: 0
    +tags: null
    +mentions: null
    +comments: Doctrine\ORM\PersistentCollection {#1866 …}
    +votes: Doctrine\ORM\PersistentCollection {#1966 …}
    +reports: Doctrine\ORM\PersistentCollection {#1946 …}
    +favourites: Doctrine\ORM\PersistentCollection {#1368 …}
    +notifications: Doctrine\ORM\PersistentCollection {#2426 …}
    +badges: Doctrine\ORM\PersistentCollection {#2439 …}
    +children: []
    -id: 18235
    -titleTs: "'attack':3 'chatgpt':5 'data':10 'googl':1 'prompt':4 'research':2 'reveal':7 'train':9"
    -bodyTs: "'/pdf/2311.17035.pdf)':134 '16.9':88 'address':109 'agreement':23 'also':61 'amount':48 'articl':127 'arxiv.org':133 'arxiv.org/pdf/2311.17035.pdf)':132 'birthday':117 'blog':17,31 'chatbot':71 'chatgpt':1,69 'cnn':14 'code':27 'comment':34 'contain':94 'edit':118 'email':106 'fandom':18 'fax':104 'found':130 'full':3,120 'generat':91 'goodread':15 'handl':112 'identifi':51,101 'includ':100 'inform':7,52 'internet':33,85 'languag':58 'larg':47,57,74 'media':111 'memor':95 'model':59 'much':36 'name':115 'news':30 'number':105 'openai':55 'overflow':25 'page':29 'paper':121 'passag':75 'percent':89 'phone':102 'physic':108 'pii':53,96 'place':82 'privat':6,50 'public':66 'random':32 'referenc':124 'research':42 'scrape':78 'sensit':5 'servic':22 'show':43,62 'social':110 'sourc':26 'spit':9,72 'stack':24 'tactic':40 'term':20 'test':93 'text':12,77 'total':87 'url':113 'use':38 'verbatim':11,79 'version':67 'wiki':19 'wikipedia':28 'wordpress':16 'wrote':98"
    +cross: false
    +upVotes: 0
    +downVotes: 0
    +ranking: 1701362787
    +visibility: "visible             "
    +apId: "https://kbin.social/m/privacy@lemmy.ml/t/665071"
    +editedAt: DateTimeImmutable @1702265006 {#1850
      date: 2023-12-11 04:23:26.0 +01:00
    }
    +createdAt: DateTimeImmutable @1701276387 {#2420
      date: 2023-11-29 17:46:27.0 +01:00
    }
  }
  +magazine: App\Entity\Magazine {#266}
  +image: null
  +parent: App\Entity\EntryComment {#4251
    +user: Proxies\__CG__\App\Entity\User {#4255 …}
    +entry: App\Entity\Entry {#2412}
    +magazine: App\Entity\Magazine {#266}
    +image: null
    +parent: App\Entity\EntryComment {#4034
      +user: App\Entity\User {#3982 …}
      +entry: App\Entity\Entry {#2412}
      +magazine: App\Entity\Magazine {#266}
      +image: null
      +parent: null
      +root: null
      +body: "Soo plagiarism essentially?"
      +lang: "en"
      +isAdult: false
      +favouriteCount: 18
      +score: 0
      +lastActive: DateTime @1702480724 {#3948
        date: 2023-12-13 16:18:44.0 +01:00
      }
      +ip: null
      +tags: null
      +mentions: [
        "@btp@kbin.social"
      ]
      +children: Doctrine\ORM\PersistentCollection {#4032 …}
      +nested: Doctrine\ORM\PersistentCollection {#4030 …}
      +votes: Doctrine\ORM\PersistentCollection {#4028 …}
      +reports: Doctrine\ORM\PersistentCollection {#4026 …}
      +favourites: Doctrine\ORM\PersistentCollection {#3994 …}
      +notifications: Doctrine\ORM\PersistentCollection {#3998 …}
      -id: 179495
      -bodyTs: "'essenti':3 'plagiar':2 'soo':1"
      +ranking: 0
      +commentCount: 0
      +upVotes: 0
      +downVotes: 0
      +visibility: "visible             "
      +apId: "https://aussie.zone/comment/4729498"
      +editedAt: null
      +createdAt: DateTimeImmutable @1701313651 {#3984
        date: 2023-11-30 04:07:31.0 +01:00
      }
    }
    +root: App\Entity\EntryComment {#4034}
    +body: """
      Always has been. Just yesterday I was explaining AI image generation to a coworker. I said the program looks at a ton of images and uses that info to blend them together. Like it knows what a soviet propaganda poster looks like, and it knows what artwork of Santa looks like so it can make a Santa themed propaganda poster.\n
      \n
      Same with text I assume. It knows the Mario wiki and fanfics, and it knows a bunch of books about zombies so it blends it to make a gritty story about Mario fending off zombies. But yeah it’s all other works just melded together.\n
      \n
      My question is would a human author be any different? We absorb ideas and stories we read and hear and blend them into new or reimagined ideas. AI just knows it’s original sources
      """
    +lang: "en"
    +isAdult: false
    +favouriteCount: 8
    +score: 0
    +lastActive: DateTime @1701349985 {#4247
      date: 2023-11-30 14:13:05.0 +01:00
    }
    +ip: null
    +tags: null
    +mentions: [
      "@btp@kbin.social"
      "@Nonameuser678@aussie.zone"
    ]
    +children: Doctrine\ORM\PersistentCollection {#4256 …}
    +nested: Doctrine\ORM\PersistentCollection {#4252 …}
    +votes: Doctrine\ORM\PersistentCollection {#4258 …}
    +reports: Doctrine\ORM\PersistentCollection {#4260 …}
    +favourites: Doctrine\ORM\PersistentCollection {#4262 …}
    +notifications: Doctrine\ORM\PersistentCollection {#4264 …}
    -id: 183662
    -bodyTs: "'absorb':117 'ai':9,133 'alway':1 'artwork':47 'assum':65 'author':112 'blend':30,84,126 'book':79 'bunch':77 'cowork':14 'differ':115 'explain':8 'fanfic':72 'fend':93 'generat':11 'gritti':89 'hear':124 'human':111 'idea':118,132 'imag':10,24 'info':28 'know':35,45,67,75,135 'like':33,42,51 'look':19,41,50 'make':55,87 'mario':69,92 'meld':104 'new':129 'origin':138 'poster':40,60 'program':18 'propaganda':39,59 'question':107 'read':122 'reimagin':131 'said':16 'santa':49,57 'sourc':139 'soviet':38 'stori':90,120 'text':63 'theme':58 'togeth':32,105 'ton':22 'use':26 'wiki':70 'work':102 'would':109 'yeah':97 'yesterday':5 'zombi':81,95"
    +ranking: 0
    +commentCount: 0
    +upVotes: 0
    +downVotes: 0
    +visibility: "visible             "
    +apId: "https://sh.itjust.works/comment/6129297"
    +editedAt: DateTimeImmutable @1702479959 {#4244
      date: 2023-12-13 16:05:59.0 +01:00
    }
    +createdAt: DateTimeImmutable @1701349985 {#4248
      date: 2023-11-30 14:13:05.0 +01:00
    }
  }
  +root: App\Entity\EntryComment {#4034}
  +body: "“Blending together” isn’t accurate, since it implies that the original images are used in the process of creating the output. The AI doesn’t have access to the original data (if it wasn’t erroneously repeated many times in the training dataset)."
  +lang: "en"
  +isAdult: false
  +favouriteCount: 3
  +score: 0
  +lastActive: DateTime @1701358588 {#4289
    date: 2023-11-30 16:36:28.0 +01:00
  }
  +ip: null
  +tags: null
  +mentions: [
    "@btp@kbin.social"
    "@Nonameuser678@aussie.zone"
    "@SomeAmateur@sh.itjust.works"
  ]
  +children: Doctrine\ORM\PersistentCollection {#4295 …}
  +nested: Doctrine\ORM\PersistentCollection {#4297 …}
  +votes: Doctrine\ORM\PersistentCollection {#4293 …}
  +reports: Doctrine\ORM\PersistentCollection {#4300 …}
  +favourites: Doctrine\ORM\PersistentCollection {#4302 …}
  +notifications: Doctrine\ORM\PersistentCollection {#4304 …}
  -id: 183807
  -bodyTs: "'access':27 'accur':5 'ai':23 'blend':1 'creat':19 'data':31 'dataset':43 'doesn':24 'erron':36 'imag':12 'impli':8 'isn':3 'mani':38 'origin':11,30 'output':21 'process':17 'repeat':37 'sinc':6 'time':39 'togeth':2 'train':42 'use':14 'wasn':34"
  +ranking: 0
  +commentCount: 0
  +upVotes: 0
  +downVotes: 0
  +visibility: "visible             "
  +apId: "https://lemmy.world/comment/5697627"
  +editedAt: null
  +createdAt: DateTimeImmutable @1701358588 {#4286
    date: 2023-11-30 16:36:28.0 +01:00
  }
}
"App\Security\Voter\UserVoter"
App\Entity\EntryComment {#4284
  +user: Proxies\__CG__\App\Entity\User {#4292 …}
  +entry: App\Entity\Entry {#2412
    +user: Proxies\__CG__\App\Entity\User {#1970 …}
    +magazine: App\Entity\Magazine {#266
      +icon: Proxies\__CG__\App\Entity\Image {#247 …}
      +name: "privacy@lemmy.ml"
      +title: "privacy"
      +description: """
        A place to discuss privacy and freedom in the digital world.\n
        ============================================================\n
        \n
        Privacy has become a very important issue in modern society, with companies and governments constantly abusing their power, more and more people are waking up to the importance of digital privacy.\n
        \n
        In this community everyone is welcome to post links and discuss topics related to privacy.\n
        \n
        ### Some Rules\n
        \n
        - Posting a link to a website containing tracking isn’t great, if contents of the website are behind a paywall maybe copy them into the post\n
        - Don’t promote proprietary software\n
        - Try to keep things on topic\n
        - If you have a question, please try searching for previous discussions, maybe it has already been answered\n
        - Reposts are fine, but should have at least a couple of weeks in between so that the post can reach a new audience\n
        - Be nice :)\n
        \n
        ### Related communities\n
        \n
        - [Lemmy.ml libre_culture](https://lemmy.ml/c/libre_culture)\n
        - [Lemmy.ml privatelife](https://lemmy.ml/c/privatelife)\n
        - [Lemmy.ml DeGoogle](https://lemmy.ml/c/degoogle)\n
        - [Lemmy.ca privacy](https://lemmy.ca/c/privacy)\n
        \n
        ### Chat rooms\n
        \n
        - [Matrix/Element]Dead\n
        - [Discord](https://discord.gg/8xfrVjvngh)\n
        \n
        much thanks to @gary_host_laptop for the logo design :)
        """
      +rules: null
      +subscriptionsCount: 1
      +entryCount: 561
      +entryCommentCount: 7684
      +postCount: 6
      +postCommentCount: 73
      +isAdult: false
      +customCss: null
      +lastActive: DateTime @1729582580 {#276
        date: 2024-10-22 09:36:20.0 +02:00
      }
      +markedForDeletionAt: null
      +tags: null
      +moderators: Doctrine\ORM\PersistentCollection {#238 …}
      +ownershipRequests: Doctrine\ORM\PersistentCollection {#234 …}
      +moderatorRequests: Doctrine\ORM\PersistentCollection {#223 …}
      +entries: Doctrine\ORM\PersistentCollection {#181 …}
      +posts: Doctrine\ORM\PersistentCollection {#139 …}
      +subscriptions: Doctrine\ORM\PersistentCollection {#201 …}
      +bans: Doctrine\ORM\PersistentCollection {#118 …}
      +reports: Doctrine\ORM\PersistentCollection {#104 …}
      +badges: Doctrine\ORM\PersistentCollection {#82 …}
      +logs: Doctrine\ORM\PersistentCollection {#72 …}
      +awards: Doctrine\ORM\PersistentCollection {#61 …}
      +categories: Doctrine\ORM\PersistentCollection {#1820 …}
      -id: 85
      +apId: "privacy@lemmy.ml"
      +apProfileId: "https://lemmy.ml/c/privacy"
      +apPublicUrl: "https://lemmy.ml/c/privacy"
      +apFollowersUrl: "https://lemmy.ml/c/privacy/followers"
      +apInboxUrl: "https://lemmy.ml/inbox"
      +apDomain: "lemmy.ml"
      +apPreferredUsername: "privacy"
      +apDiscoverable: true
      +apManuallyApprovesFollowers: null
      +privateKey: null
      +publicKey: null
      +apFetchedAt: DateTime @1729583592 {#270
        date: 2024-10-22 09:53:12.0 +02:00
      }
      +apDeletedAt: null
      +apTimeoutAt: null
      +visibility: "visible             "
      +createdAt: DateTimeImmutable @1699780272 {#272
        date: 2023-11-12 10:11:12.0 +01:00
      }
    }
    +image: Proxies\__CG__\App\Entity\Image {#1943 …}
    +domain: Proxies\__CG__\App\Entity\Domain {#1912 …}
    +slug: "Google-Researchers-Attack-Prompts-ChatGPT-to-Reveal-Its-Training-Data"
    +title: "Google Researchers’ Attack Prompts ChatGPT to Reveal Its Training Data"
    +url: "https://www.404media.co/google-researchers-attack-convinces-chatgpt-to-reveal-its-training-data/"
    +body: """
      > ChatGPT is full of sensitive private information and spits out verbatim text from CNN, Goodreads, WordPress blogs, fandom wikis, Terms of Service agreements, Stack Overflow source code, Wikipedia pages, news blogs, random internet comments, and much more.\n
      \n
      > Using this tactic, the researchers showed that there are large amounts of privately identifiable information (PII) in OpenAI’s large language models. They also showed that, on a public version of ChatGPT, the chatbot spit out large passages of text scraped verbatim from other places on the internet.\n
      \n
      > “In total, 16.9 percent of generations we tested contained memorized PII,” they wrote, which included “identifying phone and fax numbers, email and physical addresses … social media handles, URLs, and names and birthdays.”\n
      \n
      Edit: [The full paper that's referenced in the article can be found here](https://arxiv.org/pdf/2311.17035.pdf)
      """
    +type: "link"
    +lang: "en"
    +isOc: false
    +hasEmbed: false
    +commentCount: 80
    +favouriteCount: 364
    +score: 0
    +isAdult: false
    +sticky: false
    +lastActive: DateTime @1701712043 {#2410
      date: 2023-12-04 18:47:23.0 +01:00
    }
    +ip: null
    +adaAmount: 0
    +tags: null
    +mentions: null
    +comments: Doctrine\ORM\PersistentCollection {#1866 …}
    +votes: Doctrine\ORM\PersistentCollection {#1966 …}
    +reports: Doctrine\ORM\PersistentCollection {#1946 …}
    +favourites: Doctrine\ORM\PersistentCollection {#1368 …}
    +notifications: Doctrine\ORM\PersistentCollection {#2426 …}
    +badges: Doctrine\ORM\PersistentCollection {#2439 …}
    +children: []
    -id: 18235
    -titleTs: "'attack':3 'chatgpt':5 'data':10 'googl':1 'prompt':4 'research':2 'reveal':7 'train':9"
    -bodyTs: "'/pdf/2311.17035.pdf)':134 '16.9':88 'address':109 'agreement':23 'also':61 'amount':48 'articl':127 'arxiv.org':133 'arxiv.org/pdf/2311.17035.pdf)':132 'birthday':117 'blog':17,31 'chatbot':71 'chatgpt':1,69 'cnn':14 'code':27 'comment':34 'contain':94 'edit':118 'email':106 'fandom':18 'fax':104 'found':130 'full':3,120 'generat':91 'goodread':15 'handl':112 'identifi':51,101 'includ':100 'inform':7,52 'internet':33,85 'languag':58 'larg':47,57,74 'media':111 'memor':95 'model':59 'much':36 'name':115 'news':30 'number':105 'openai':55 'overflow':25 'page':29 'paper':121 'passag':75 'percent':89 'phone':102 'physic':108 'pii':53,96 'place':82 'privat':6,50 'public':66 'random':32 'referenc':124 'research':42 'scrape':78 'sensit':5 'servic':22 'show':43,62 'social':110 'sourc':26 'spit':9,72 'stack':24 'tactic':40 'term':20 'test':93 'text':12,77 'total':87 'url':113 'use':38 'verbatim':11,79 'version':67 'wiki':19 'wikipedia':28 'wordpress':16 'wrote':98"
    +cross: false
    +upVotes: 0
    +downVotes: 0
    +ranking: 1701362787
    +visibility: "visible             "
    +apId: "https://kbin.social/m/privacy@lemmy.ml/t/665071"
    +editedAt: DateTimeImmutable @1702265006 {#1850
      date: 2023-12-11 04:23:26.0 +01:00
    }
    +createdAt: DateTimeImmutable @1701276387 {#2420
      date: 2023-11-29 17:46:27.0 +01:00
    }
  }
  +magazine: App\Entity\Magazine {#266}
  +image: null
  +parent: App\Entity\EntryComment {#4251
    +user: Proxies\__CG__\App\Entity\User {#4255 …}
    +entry: App\Entity\Entry {#2412}
    +magazine: App\Entity\Magazine {#266}
    +image: null
    +parent: App\Entity\EntryComment {#4034
      +user: App\Entity\User {#3982 …}
      +entry: App\Entity\Entry {#2412}
      +magazine: App\Entity\Magazine {#266}
      +image: null
      +parent: null
      +root: null
      +body: "Soo plagiarism essentially?"
      +lang: "en"
      +isAdult: false
      +favouriteCount: 18
      +score: 0
      +lastActive: DateTime @1702480724 {#3948
        date: 2023-12-13 16:18:44.0 +01:00
      }
      +ip: null
      +tags: null
      +mentions: [
        "@btp@kbin.social"
      ]
      +children: Doctrine\ORM\PersistentCollection {#4032 …}
      +nested: Doctrine\ORM\PersistentCollection {#4030 …}
      +votes: Doctrine\ORM\PersistentCollection {#4028 …}
      +reports: Doctrine\ORM\PersistentCollection {#4026 …}
      +favourites: Doctrine\ORM\PersistentCollection {#3994 …}
      +notifications: Doctrine\ORM\PersistentCollection {#3998 …}
      -id: 179495
      -bodyTs: "'essenti':3 'plagiar':2 'soo':1"
      +ranking: 0
      +commentCount: 0
      +upVotes: 0
      +downVotes: 0
      +visibility: "visible             "
      +apId: "https://aussie.zone/comment/4729498"
      +editedAt: null
      +createdAt: DateTimeImmutable @1701313651 {#3984
        date: 2023-11-30 04:07:31.0 +01:00
      }
    }
    +root: App\Entity\EntryComment {#4034}
    +body: """
      Always has been. Just yesterday I was explaining AI image generation to a coworker. I said the program looks at a ton of images and uses that info to blend them together. Like it knows what a soviet propaganda poster looks like, and it knows what artwork of Santa looks like so it can make a Santa themed propaganda poster.\n
      \n
      Same with text I assume. It knows the Mario wiki and fanfics, and it knows a bunch of books about zombies so it blends it to make a gritty story about Mario fending off zombies. But yeah it’s all other works just melded together.\n
      \n
      My question is would a human author be any different? We absorb ideas and stories we read and hear and blend them into new or reimagined ideas. AI just knows it’s original sources
      """
    +lang: "en"
    +isAdult: false
    +favouriteCount: 8
    +score: 0
    +lastActive: DateTime @1701349985 {#4247
      date: 2023-11-30 14:13:05.0 +01:00
    }
    +ip: null
    +tags: null
    +mentions: [
      "@btp@kbin.social"
      "@Nonameuser678@aussie.zone"
    ]
    +children: Doctrine\ORM\PersistentCollection {#4256 …}
    +nested: Doctrine\ORM\PersistentCollection {#4252 …}
    +votes: Doctrine\ORM\PersistentCollection {#4258 …}
    +reports: Doctrine\ORM\PersistentCollection {#4260 …}
    +favourites: Doctrine\ORM\PersistentCollection {#4262 …}
    +notifications: Doctrine\ORM\PersistentCollection {#4264 …}
    -id: 183662
    -bodyTs: "'absorb':117 'ai':9,133 'alway':1 'artwork':47 'assum':65 'author':112 'blend':30,84,126 'book':79 'bunch':77 'cowork':14 'differ':115 'explain':8 'fanfic':72 'fend':93 'generat':11 'gritti':89 'hear':124 'human':111 'idea':118,132 'imag':10,24 'info':28 'know':35,45,67,75,135 'like':33,42,51 'look':19,41,50 'make':55,87 'mario':69,92 'meld':104 'new':129 'origin':138 'poster':40,60 'program':18 'propaganda':39,59 'question':107 'read':122 'reimagin':131 'said':16 'santa':49,57 'sourc':139 'soviet':38 'stori':90,120 'text':63 'theme':58 'togeth':32,105 'ton':22 'use':26 'wiki':70 'work':102 'would':109 'yeah':97 'yesterday':5 'zombi':81,95"
    +ranking: 0
    +commentCount: 0
    +upVotes: 0
    +downVotes: 0
    +visibility: "visible             "
    +apId: "https://sh.itjust.works/comment/6129297"
    +editedAt: DateTimeImmutable @1702479959 {#4244
      date: 2023-12-13 16:05:59.0 +01:00
    }
    +createdAt: DateTimeImmutable @1701349985 {#4248
      date: 2023-11-30 14:13:05.0 +01:00
    }
  }
  +root: App\Entity\EntryComment {#4034}
  +body: "“Blending together” isn’t accurate, since it implies that the original images are used in the process of creating the output. The AI doesn’t have access to the original data (if it wasn’t erroneously repeated many times in the training dataset)."
  +lang: "en"
  +isAdult: false
  +favouriteCount: 3
  +score: 0
  +lastActive: DateTime @1701358588 {#4289
    date: 2023-11-30 16:36:28.0 +01:00
  }
  +ip: null
  +tags: null
  +mentions: [
    "@btp@kbin.social"
    "@Nonameuser678@aussie.zone"
    "@SomeAmateur@sh.itjust.works"
  ]
  +children: Doctrine\ORM\PersistentCollection {#4295 …}
  +nested: Doctrine\ORM\PersistentCollection {#4297 …}
  +votes: Doctrine\ORM\PersistentCollection {#4293 …}
  +reports: Doctrine\ORM\PersistentCollection {#4300 …}
  +favourites: Doctrine\ORM\PersistentCollection {#4302 …}
  +notifications: Doctrine\ORM\PersistentCollection {#4304 …}
  -id: 183807
  -bodyTs: "'access':27 'accur':5 'ai':23 'blend':1 'creat':19 'data':31 'dataset':43 'doesn':24 'erron':36 'imag':12 'impli':8 'isn':3 'mani':38 'origin':11,30 'output':21 'process':17 'repeat':37 'sinc':6 'time':39 'togeth':2 'train':42 'use':14 'wasn':34"
  +ranking: 0
  +commentCount: 0
  +upVotes: 0
  +downVotes: 0
  +visibility: "visible             "
  +apId: "https://lemmy.world/comment/5697627"
  +editedAt: null
  +createdAt: DateTimeImmutable @1701358588 {#4286
    date: 2023-11-30 16:36:28.0 +01:00
  }
}
"App\Security\Voter\UserVoter"
null
"App\Security\Voter\UserVoter"
App\Entity\EntryComment {#4309
  +user: App\Entity\User {#4165 …}
  +entry: App\Entity\Entry {#2412
    +user: Proxies\__CG__\App\Entity\User {#1970 …}
    +magazine: App\Entity\Magazine {#266
      +icon: Proxies\__CG__\App\Entity\Image {#247 …}
      +name: "privacy@lemmy.ml"
      +title: "privacy"
      +description: """
        A place to discuss privacy and freedom in the digital world.\n
        ============================================================\n
        \n
        Privacy has become a very important issue in modern society, with companies and governments constantly abusing their power, more and more people are waking up to the importance of digital privacy.\n
        \n
        In this community everyone is welcome to post links and discuss topics related to privacy.\n
        \n
        ### Some Rules\n
        \n
        - Posting a link to a website containing tracking isn’t great, if contents of the website are behind a paywall maybe copy them into the post\n
        - Don’t promote proprietary software\n
        - Try to keep things on topic\n
        - If you have a question, please try searching for previous discussions, maybe it has already been answered\n
        - Reposts are fine, but should have at least a couple of weeks in between so that the post can reach a new audience\n
        - Be nice :)\n
        \n
        ### Related communities\n
        \n
        - [Lemmy.ml libre_culture](https://lemmy.ml/c/libre_culture)\n
        - [Lemmy.ml privatelife](https://lemmy.ml/c/privatelife)\n
        - [Lemmy.ml DeGoogle](https://lemmy.ml/c/degoogle)\n
        - [Lemmy.ca privacy](https://lemmy.ca/c/privacy)\n
        \n
        ### Chat rooms\n
        \n
        - [Matrix/Element]Dead\n
        - [Discord](https://discord.gg/8xfrVjvngh)\n
        \n
        much thanks to @gary_host_laptop for the logo design :)
        """
      +rules: null
      +subscriptionsCount: 1
      +entryCount: 561
      +entryCommentCount: 7684
      +postCount: 6
      +postCommentCount: 73
      +isAdult: false
      +customCss: null
      +lastActive: DateTime @1729582580 {#276
        date: 2024-10-22 09:36:20.0 +02:00
      }
      +markedForDeletionAt: null
      +tags: null
      +moderators: Doctrine\ORM\PersistentCollection {#238 …}
      +ownershipRequests: Doctrine\ORM\PersistentCollection {#234 …}
      +moderatorRequests: Doctrine\ORM\PersistentCollection {#223 …}
      +entries: Doctrine\ORM\PersistentCollection {#181 …}
      +posts: Doctrine\ORM\PersistentCollection {#139 …}
      +subscriptions: Doctrine\ORM\PersistentCollection {#201 …}
      +bans: Doctrine\ORM\PersistentCollection {#118 …}
      +reports: Doctrine\ORM\PersistentCollection {#104 …}
      +badges: Doctrine\ORM\PersistentCollection {#82 …}
      +logs: Doctrine\ORM\PersistentCollection {#72 …}
      +awards: Doctrine\ORM\PersistentCollection {#61 …}
      +categories: Doctrine\ORM\PersistentCollection {#1820 …}
      -id: 85
      +apId: "privacy@lemmy.ml"
      +apProfileId: "https://lemmy.ml/c/privacy"
      +apPublicUrl: "https://lemmy.ml/c/privacy"
      +apFollowersUrl: "https://lemmy.ml/c/privacy/followers"
      +apInboxUrl: "https://lemmy.ml/inbox"
      +apDomain: "lemmy.ml"
      +apPreferredUsername: "privacy"
      +apDiscoverable: true
      +apManuallyApprovesFollowers: null
      +privateKey: null
      +publicKey: null
      +apFetchedAt: DateTime @1729583592 {#270
        date: 2024-10-22 09:53:12.0 +02:00
      }
      +apDeletedAt: null
      +apTimeoutAt: null
      +visibility: "visible             "
      +createdAt: DateTimeImmutable @1699780272 {#272
        date: 2023-11-12 10:11:12.0 +01:00
      }
    }
    +image: Proxies\__CG__\App\Entity\Image {#1943 …}
    +domain: Proxies\__CG__\App\Entity\Domain {#1912 …}
    +slug: "Google-Researchers-Attack-Prompts-ChatGPT-to-Reveal-Its-Training-Data"
    +title: "Google Researchers’ Attack Prompts ChatGPT to Reveal Its Training Data"
    +url: "https://www.404media.co/google-researchers-attack-convinces-chatgpt-to-reveal-its-training-data/"
    +body: """
      > ChatGPT is full of sensitive private information and spits out verbatim text from CNN, Goodreads, WordPress blogs, fandom wikis, Terms of Service agreements, Stack Overflow source code, Wikipedia pages, news blogs, random internet comments, and much more.\n
      \n
      > Using this tactic, the researchers showed that there are large amounts of privately identifiable information (PII) in OpenAI’s large language models. They also showed that, on a public version of ChatGPT, the chatbot spit out large passages of text scraped verbatim from other places on the internet.\n
      \n
      > “In total, 16.9 percent of generations we tested contained memorized PII,” they wrote, which included “identifying phone and fax numbers, email and physical addresses … social media handles, URLs, and names and birthdays.”\n
      \n
      Edit: [The full paper that's referenced in the article can be found here](https://arxiv.org/pdf/2311.17035.pdf)
      """
    +type: "link"
    +lang: "en"
    +isOc: false
    +hasEmbed: false
    +commentCount: 80
    +favouriteCount: 364
    +score: 0
    +isAdult: false
    +sticky: false
    +lastActive: DateTime @1701712043 {#2410
      date: 2023-12-04 18:47:23.0 +01:00
    }
    +ip: null
    +adaAmount: 0
    +tags: null
    +mentions: null
    +comments: Doctrine\ORM\PersistentCollection {#1866 …}
    +votes: Doctrine\ORM\PersistentCollection {#1966 …}
    +reports: Doctrine\ORM\PersistentCollection {#1946 …}
    +favourites: Doctrine\ORM\PersistentCollection {#1368 …}
    +notifications: Doctrine\ORM\PersistentCollection {#2426 …}
    +badges: Doctrine\ORM\PersistentCollection {#2439 …}
    +children: []
    -id: 18235
    -titleTs: "'attack':3 'chatgpt':5 'data':10 'googl':1 'prompt':4 'research':2 'reveal':7 'train':9"
    -bodyTs: "'/pdf/2311.17035.pdf)':134 '16.9':88 'address':109 'agreement':23 'also':61 'amount':48 'articl':127 'arxiv.org':133 'arxiv.org/pdf/2311.17035.pdf)':132 'birthday':117 'blog':17,31 'chatbot':71 'chatgpt':1,69 'cnn':14 'code':27 'comment':34 'contain':94 'edit':118 'email':106 'fandom':18 'fax':104 'found':130 'full':3,120 'generat':91 'goodread':15 'handl':112 'identifi':51,101 'includ':100 'inform':7,52 'internet':33,85 'languag':58 'larg':47,57,74 'media':111 'memor':95 'model':59 'much':36 'name':115 'news':30 'number':105 'openai':55 'overflow':25 'page':29 'paper':121 'passag':75 'percent':89 'phone':102 'physic':108 'pii':53,96 'place':82 'privat':6,50 'public':66 'random':32 'referenc':124 'research':42 'scrape':78 'sensit':5 'servic':22 'show':43,62 'social':110 'sourc':26 'spit':9,72 'stack':24 'tactic':40 'term':20 'test':93 'text':12,77 'total':87 'url':113 'use':38 'verbatim':11,79 'version':67 'wiki':19 'wikipedia':28 'wordpress':16 'wrote':98"
    +cross: false
    +upVotes: 0
    +downVotes: 0
    +ranking: 1701362787
    +visibility: "visible             "
    +apId: "https://kbin.social/m/privacy@lemmy.ml/t/665071"
    +editedAt: DateTimeImmutable @1702265006 {#1850
      date: 2023-12-11 04:23:26.0 +01:00
    }
    +createdAt: DateTimeImmutable @1701276387 {#2420
      date: 2023-11-29 17:46:27.0 +01:00
    }
  }
  +magazine: App\Entity\Magazine {#266}
  +image: null
  +parent: App\Entity\EntryComment {#4251
    +user: Proxies\__CG__\App\Entity\User {#4255 …}
    +entry: App\Entity\Entry {#2412}
    +magazine: App\Entity\Magazine {#266}
    +image: null
    +parent: App\Entity\EntryComment {#4034
      +user: App\Entity\User {#3982 …}
      +entry: App\Entity\Entry {#2412}
      +magazine: App\Entity\Magazine {#266}
      +image: null
      +parent: null
      +root: null
      +body: "Soo plagiarism essentially?"
      +lang: "en"
      +isAdult: false
      +favouriteCount: 18
      +score: 0
      +lastActive: DateTime @1702480724 {#3948
        date: 2023-12-13 16:18:44.0 +01:00
      }
      +ip: null
      +tags: null
      +mentions: [
        "@btp@kbin.social"
      ]
      +children: Doctrine\ORM\PersistentCollection {#4032 …}
      +nested: Doctrine\ORM\PersistentCollection {#4030 …}
      +votes: Doctrine\ORM\PersistentCollection {#4028 …}
      +reports: Doctrine\ORM\PersistentCollection {#4026 …}
      +favourites: Doctrine\ORM\PersistentCollection {#3994 …}
      +notifications: Doctrine\ORM\PersistentCollection {#3998 …}
      -id: 179495
      -bodyTs: "'essenti':3 'plagiar':2 'soo':1"
      +ranking: 0
      +commentCount: 0
      +upVotes: 0
      +downVotes: 0
      +visibility: "visible             "
      +apId: "https://aussie.zone/comment/4729498"
      +editedAt: null
      +createdAt: DateTimeImmutable @1701313651 {#3984
        date: 2023-11-30 04:07:31.0 +01:00
      }
    }
    +root: App\Entity\EntryComment {#4034}
    +body: """
      Always has been. Just yesterday I was explaining AI image generation to a coworker. I said the program looks at a ton of images and uses that info to blend them together. Like it knows what a soviet propaganda poster looks like, and it knows what artwork of Santa looks like so it can make a Santa themed propaganda poster.\n
      \n
      Same with text I assume. It knows the Mario wiki and fanfics, and it knows a bunch of books about zombies so it blends it to make a gritty story about Mario fending off zombies. But yeah it’s all other works just melded together.\n
      \n
      My question is would a human author be any different? We absorb ideas and stories we read and hear and blend them into new or reimagined ideas. AI just knows it’s original sources
      """
    +lang: "en"
    +isAdult: false
    +favouriteCount: 8
    +score: 0
    +lastActive: DateTime @1701349985 {#4247
      date: 2023-11-30 14:13:05.0 +01:00
    }
    +ip: null
    +tags: null
    +mentions: [
      "@btp@kbin.social"
      "@Nonameuser678@aussie.zone"
    ]
    +children: Doctrine\ORM\PersistentCollection {#4256 …}
    +nested: Doctrine\ORM\PersistentCollection {#4252 …}
    +votes: Doctrine\ORM\PersistentCollection {#4258 …}
    +reports: Doctrine\ORM\PersistentCollection {#4260 …}
    +favourites: Doctrine\ORM\PersistentCollection {#4262 …}
    +notifications: Doctrine\ORM\PersistentCollection {#4264 …}
    -id: 183662
    -bodyTs: "'absorb':117 'ai':9,133 'alway':1 'artwork':47 'assum':65 'author':112 'blend':30,84,126 'book':79 'bunch':77 'cowork':14 'differ':115 'explain':8 'fanfic':72 'fend':93 'generat':11 'gritti':89 'hear':124 'human':111 'idea':118,132 'imag':10,24 'info':28 'know':35,45,67,75,135 'like':33,42,51 'look':19,41,50 'make':55,87 'mario':69,92 'meld':104 'new':129 'origin':138 'poster':40,60 'program':18 'propaganda':39,59 'question':107 'read':122 'reimagin':131 'said':16 'santa':49,57 'sourc':139 'soviet':38 'stori':90,120 'text':63 'theme':58 'togeth':32,105 'ton':22 'use':26 'wiki':70 'work':102 'would':109 'yeah':97 'yesterday':5 'zombi':81,95"
    +ranking: 0
    +commentCount: 0
    +upVotes: 0
    +downVotes: 0
    +visibility: "visible             "
    +apId: "https://sh.itjust.works/comment/6129297"
    +editedAt: DateTimeImmutable @1702479959 {#4244
      date: 2023-12-13 16:05:59.0 +01:00
    }
    +createdAt: DateTimeImmutable @1701349985 {#4248
      date: 2023-11-30 14:13:05.0 +01:00
    }
  }
  +root: App\Entity\EntryComment {#4034}
  +body: """
    > My question is would a human author be any different?\n
    \n
    Humans don’t remember the exact source material, it gets abstracted into concepts before being saved as an engram. This is how we’re able to create new works of art while AI is only able to do photoshop on its training data. Humans will forget the text but remember the soul, AI only has access to the exact work and cannot replicate the soul of a work (at least with its current implementation, if these systems were made to be anything more than glorified IP theft we *could* see systems that could actually do art like humans, but we don’t live in that world)
    """
  +lang: "en"
  +isAdult: false
  +favouriteCount: 2
  +score: 0
  +lastActive: DateTime @1701360735 {#4306
    date: 2023-11-30 17:12:15.0 +01:00
  }
  +ip: null
  +tags: null
  +mentions: [
    "@btp@kbin.social"
    "@Nonameuser678@aussie.zone"
    "@SomeAmateur@sh.itjust.works"
  ]
  +children: Doctrine\ORM\PersistentCollection {#4310 …}
  +nested: Doctrine\ORM\PersistentCollection {#4312 …}
  +votes: Doctrine\ORM\PersistentCollection {#4314 …}
  +reports: Doctrine\ORM\PersistentCollection {#4316 …}
  +favourites: Doctrine\ORM\PersistentCollection {#4318 …}
  +notifications: Doctrine\ORM\PersistentCollection {#4320 …}
  -id: 183840
  -bodyTs: "'abl':35,46 'abstract':21 'access':66 'actual':104 'ai':43,63 'anyth':92 'art':41,106 'author':7 'cannot':72 'concept':23 'could':99,103 'creat':37 'current':83 'data':53 'differ':10 'engram':29 'exact':16,69 'forget':56 'get':20 'glorifi':95 'human':6,11,54,108 'implement':84 'ip':96 'least':80 'like':107 'live':113 'made':89 'materi':18 'new':38 'photoshop':49 'question':2 're':34 'rememb':14,60 'replic':73 'save':26 'see':100 'soul':62,75 'sourc':17 'system':87,101 'text':58 'theft':97 'train':52 'work':39,70,78 'world':116 'would':4"
  +ranking: 0
  +commentCount: 0
  +upVotes: 0
  +downVotes: 0
  +visibility: "visible             "
  +apId: "https://lemmy.ml/comment/6257463"
  +editedAt: DateTimeImmutable @1702480746 {#4307
    date: 2023-12-13 16:19:06.0 +01:00
  }
  +createdAt: DateTimeImmutable @1701360735 {#4308
    date: 2023-11-30 17:12:15.0 +01:00
  }
}
"App\Security\Voter\UserVoter"
App\Entity\EntryComment {#4309
  +user: App\Entity\User {#4165 …}
  +entry: App\Entity\Entry {#2412
    +user: Proxies\__CG__\App\Entity\User {#1970 …}
    +magazine: App\Entity\Magazine {#266
      +icon: Proxies\__CG__\App\Entity\Image {#247 …}
      +name: "privacy@lemmy.ml"
      +title: "privacy"
      +description: """
        A place to discuss privacy and freedom in the digital world.\n
        ============================================================\n
        \n
        Privacy has become a very important issue in modern society, with companies and governments constantly abusing their power, more and more people are waking up to the importance of digital privacy.\n
        \n
        In this community everyone is welcome to post links and discuss topics related to privacy.\n
        \n
        ### Some Rules\n
        \n
        - Posting a link to a website containing tracking isn’t great, if contents of the website are behind a paywall maybe copy them into the post\n
        - Don’t promote proprietary software\n
        - Try to keep things on topic\n
        - If you have a question, please try searching for previous discussions, maybe it has already been answered\n
        - Reposts are fine, but should have at least a couple of weeks in between so that the post can reach a new audience\n
        - Be nice :)\n
        \n
        ### Related communities\n
        \n
        - [Lemmy.ml libre_culture](https://lemmy.ml/c/libre_culture)\n
        - [Lemmy.ml privatelife](https://lemmy.ml/c/privatelife)\n
        - [Lemmy.ml DeGoogle](https://lemmy.ml/c/degoogle)\n
        - [Lemmy.ca privacy](https://lemmy.ca/c/privacy)\n
        \n
        ### Chat rooms\n
        \n
        - [Matrix/Element]Dead\n
        - [Discord](https://discord.gg/8xfrVjvngh)\n
        \n
        much thanks to @gary_host_laptop for the logo design :)
        """
      +rules: null
      +subscriptionsCount: 1
      +entryCount: 561
      +entryCommentCount: 7684
      +postCount: 6
      +postCommentCount: 73
      +isAdult: false
      +customCss: null
      +lastActive: DateTime @1729582580 {#276
        date: 2024-10-22 09:36:20.0 +02:00
      }
      +markedForDeletionAt: null
      +tags: null
      +moderators: Doctrine\ORM\PersistentCollection {#238 …}
      +ownershipRequests: Doctrine\ORM\PersistentCollection {#234 …}
      +moderatorRequests: Doctrine\ORM\PersistentCollection {#223 …}
      +entries: Doctrine\ORM\PersistentCollection {#181 …}
      +posts: Doctrine\ORM\PersistentCollection {#139 …}
      +subscriptions: Doctrine\ORM\PersistentCollection {#201 …}
      +bans: Doctrine\ORM\PersistentCollection {#118 …}
      +reports: Doctrine\ORM\PersistentCollection {#104 …}
      +badges: Doctrine\ORM\PersistentCollection {#82 …}
      +logs: Doctrine\ORM\PersistentCollection {#72 …}
      +awards: Doctrine\ORM\PersistentCollection {#61 …}
      +categories: Doctrine\ORM\PersistentCollection {#1820 …}
      -id: 85
      +apId: "privacy@lemmy.ml"
      +apProfileId: "https://lemmy.ml/c/privacy"
      +apPublicUrl: "https://lemmy.ml/c/privacy"
      +apFollowersUrl: "https://lemmy.ml/c/privacy/followers"
      +apInboxUrl: "https://lemmy.ml/inbox"
      +apDomain: "lemmy.ml"
      +apPreferredUsername: "privacy"
      +apDiscoverable: true
      +apManuallyApprovesFollowers: null
      +privateKey: null
      +publicKey: null
      +apFetchedAt: DateTime @1729583592 {#270
        date: 2024-10-22 09:53:12.0 +02:00
      }
      +apDeletedAt: null
      +apTimeoutAt: null
      +visibility: "visible             "
      +createdAt: DateTimeImmutable @1699780272 {#272
        date: 2023-11-12 10:11:12.0 +01:00
      }
    }
    +image: Proxies\__CG__\App\Entity\Image {#1943 …}
    +domain: Proxies\__CG__\App\Entity\Domain {#1912 …}
    +slug: "Google-Researchers-Attack-Prompts-ChatGPT-to-Reveal-Its-Training-Data"
    +title: "Google Researchers’ Attack Prompts ChatGPT to Reveal Its Training Data"
    +url: "https://www.404media.co/google-researchers-attack-convinces-chatgpt-to-reveal-its-training-data/"
    +body: """
      > ChatGPT is full of sensitive private information and spits out verbatim text from CNN, Goodreads, WordPress blogs, fandom wikis, Terms of Service agreements, Stack Overflow source code, Wikipedia pages, news blogs, random internet comments, and much more.\n
      \n
      > Using this tactic, the researchers showed that there are large amounts of privately identifiable information (PII) in OpenAI’s large language models. They also showed that, on a public version of ChatGPT, the chatbot spit out large passages of text scraped verbatim from other places on the internet.\n
      \n
      > “In total, 16.9 percent of generations we tested contained memorized PII,” they wrote, which included “identifying phone and fax numbers, email and physical addresses … social media handles, URLs, and names and birthdays.”\n
      \n
      Edit: [The full paper that's referenced in the article can be found here](https://arxiv.org/pdf/2311.17035.pdf)
      """
    +type: "link"
    +lang: "en"
    +isOc: false
    +hasEmbed: false
    +commentCount: 80
    +favouriteCount: 364
    +score: 0
    +isAdult: false
    +sticky: false
    +lastActive: DateTime @1701712043 {#2410
      date: 2023-12-04 18:47:23.0 +01:00
    }
    +ip: null
    +adaAmount: 0
    +tags: null
    +mentions: null
    +comments: Doctrine\ORM\PersistentCollection {#1866 …}
    +votes: Doctrine\ORM\PersistentCollection {#1966 …}
    +reports: Doctrine\ORM\PersistentCollection {#1946 …}
    +favourites: Doctrine\ORM\PersistentCollection {#1368 …}
    +notifications: Doctrine\ORM\PersistentCollection {#2426 …}
    +badges: Doctrine\ORM\PersistentCollection {#2439 …}
    +children: []
    -id: 18235
    -titleTs: "'attack':3 'chatgpt':5 'data':10 'googl':1 'prompt':4 'research':2 'reveal':7 'train':9"
    -bodyTs: "'/pdf/2311.17035.pdf)':134 '16.9':88 'address':109 'agreement':23 'also':61 'amount':48 'articl':127 'arxiv.org':133 'arxiv.org/pdf/2311.17035.pdf)':132 'birthday':117 'blog':17,31 'chatbot':71 'chatgpt':1,69 'cnn':14 'code':27 'comment':34 'contain':94 'edit':118 'email':106 'fandom':18 'fax':104 'found':130 'full':3,120 'generat':91 'goodread':15 'handl':112 'identifi':51,101 'includ':100 'inform':7,52 'internet':33,85 'languag':58 'larg':47,57,74 'media':111 'memor':95 'model':59 'much':36 'name':115 'news':30 'number':105 'openai':55 'overflow':25 'page':29 'paper':121 'passag':75 'percent':89 'phone':102 'physic':108 'pii':53,96 'place':82 'privat':6,50 'public':66 'random':32 'referenc':124 'research':42 'scrape':78 'sensit':5 'servic':22 'show':43,62 'social':110 'sourc':26 'spit':9,72 'stack':24 'tactic':40 'term':20 'test':93 'text':12,77 'total':87 'url':113 'use':38 'verbatim':11,79 'version':67 'wiki':19 'wikipedia':28 'wordpress':16 'wrote':98"
    +cross: false
    +upVotes: 0
    +downVotes: 0
    +ranking: 1701362787
    +visibility: "visible             "
    +apId: "https://kbin.social/m/privacy@lemmy.ml/t/665071"
    +editedAt: DateTimeImmutable @1702265006 {#1850
      date: 2023-12-11 04:23:26.0 +01:00
    }
    +createdAt: DateTimeImmutable @1701276387 {#2420
      date: 2023-11-29 17:46:27.0 +01:00
    }
  }
  +magazine: App\Entity\Magazine {#266}
  +image: null
  +parent: App\Entity\EntryComment {#4251
    +user: Proxies\__CG__\App\Entity\User {#4255 …}
    +entry: App\Entity\Entry {#2412}
    +magazine: App\Entity\Magazine {#266}
    +image: null
    +parent: App\Entity\EntryComment {#4034
      +user: App\Entity\User {#3982 …}
      +entry: App\Entity\Entry {#2412}
      +magazine: App\Entity\Magazine {#266}
      +image: null
      +parent: null
      +root: null
      +body: "Soo plagiarism essentially?"
      +lang: "en"
      +isAdult: false
      +favouriteCount: 18
      +score: 0
      +lastActive: DateTime @1702480724 {#3948
        date: 2023-12-13 16:18:44.0 +01:00
      }
      +ip: null
      +tags: null
      +mentions: [
        "@btp@kbin.social"
      ]
      +children: Doctrine\ORM\PersistentCollection {#4032 …}
      +nested: Doctrine\ORM\PersistentCollection {#4030 …}
      +votes: Doctrine\ORM\PersistentCollection {#4028 …}
      +reports: Doctrine\ORM\PersistentCollection {#4026 …}
      +favourites: Doctrine\ORM\PersistentCollection {#3994 …}
      +notifications: Doctrine\ORM\PersistentCollection {#3998 …}
      -id: 179495
      -bodyTs: "'essenti':3 'plagiar':2 'soo':1"
      +ranking: 0
      +commentCount: 0
      +upVotes: 0
      +downVotes: 0
      +visibility: "visible             "
      +apId: "https://aussie.zone/comment/4729498"
      +editedAt: null
      +createdAt: DateTimeImmutable @1701313651 {#3984
        date: 2023-11-30 04:07:31.0 +01:00
      }
    }
    +root: App\Entity\EntryComment {#4034}
    +body: """
      Always has been. Just yesterday I was explaining AI image generation to a coworker. I said the program looks at a ton of images and uses that info to blend them together. Like it knows what a soviet propaganda poster looks like, and it knows what artwork of Santa looks like so it can make a Santa themed propaganda poster.\n
      \n
      Same with text I assume. It knows the Mario wiki and fanfics, and it knows a bunch of books about zombies so it blends it to make a gritty story about Mario fending off zombies. But yeah it’s all other works just melded together.\n
      \n
      My question is would a human author be any different? We absorb ideas and stories we read and hear and blend them into new or reimagined ideas. AI just knows it’s original sources
      """
    +lang: "en"
    +isAdult: false
    +favouriteCount: 8
    +score: 0
    +lastActive: DateTime @1701349985 {#4247
      date: 2023-11-30 14:13:05.0 +01:00
    }
    +ip: null
    +tags: null
    +mentions: [
      "@btp@kbin.social"
      "@Nonameuser678@aussie.zone"
    ]
    +children: Doctrine\ORM\PersistentCollection {#4256 …}
    +nested: Doctrine\ORM\PersistentCollection {#4252 …}
    +votes: Doctrine\ORM\PersistentCollection {#4258 …}
    +reports: Doctrine\ORM\PersistentCollection {#4260 …}
    +favourites: Doctrine\ORM\PersistentCollection {#4262 …}
    +notifications: Doctrine\ORM\PersistentCollection {#4264 …}
    -id: 183662
    -bodyTs: "'absorb':117 'ai':9,133 'alway':1 'artwork':47 'assum':65 'author':112 'blend':30,84,126 'book':79 'bunch':77 'cowork':14 'differ':115 'explain':8 'fanfic':72 'fend':93 'generat':11 'gritti':89 'hear':124 'human':111 'idea':118,132 'imag':10,24 'info':28 'know':35,45,67,75,135 'like':33,42,51 'look':19,41,50 'make':55,87 'mario':69,92 'meld':104 'new':129 'origin':138 'poster':40,60 'program':18 'propaganda':39,59 'question':107 'read':122 'reimagin':131 'said':16 'santa':49,57 'sourc':139 'soviet':38 'stori':90,120 'text':63 'theme':58 'togeth':32,105 'ton':22 'use':26 'wiki':70 'work':102 'would':109 'yeah':97 'yesterday':5 'zombi':81,95"
    +ranking: 0
    +commentCount: 0
    +upVotes: 0
    +downVotes: 0
    +visibility: "visible             "
    +apId: "https://sh.itjust.works/comment/6129297"
    +editedAt: DateTimeImmutable @1702479959 {#4244
      date: 2023-12-13 16:05:59.0 +01:00
    }
    +createdAt: DateTimeImmutable @1701349985 {#4248
      date: 2023-11-30 14:13:05.0 +01:00
    }
  }
  +root: App\Entity\EntryComment {#4034}
  +body: """
    > My question is would a human author be any different?\n
    \n
    Humans don’t remember the exact source material, it gets abstracted into concepts before being saved as an engram. This is how we’re able to create new works of art while AI is only able to do photoshop on its training data. Humans will forget the text but remember the soul, AI only has access to the exact work and cannot replicate the soul of a work (at least with its current implementation, if these systems were made to be anything more than glorified IP theft we *could* see systems that could actually do art like humans, but we don’t live in that world)
    """
  +lang: "en"
  +isAdult: false
  +favouriteCount: 2
  +score: 0
  +lastActive: DateTime @1701360735 {#4306
    date: 2023-11-30 17:12:15.0 +01:00
  }
  +ip: null
  +tags: null
  +mentions: [
    "@btp@kbin.social"
    "@Nonameuser678@aussie.zone"
    "@SomeAmateur@sh.itjust.works"
  ]
  +children: Doctrine\ORM\PersistentCollection {#4310 …}
  +nested: Doctrine\ORM\PersistentCollection {#4312 …}
  +votes: Doctrine\ORM\PersistentCollection {#4314 …}
  +reports: Doctrine\ORM\PersistentCollection {#4316 …}
  +favourites: Doctrine\ORM\PersistentCollection {#4318 …}
  +notifications: Doctrine\ORM\PersistentCollection {#4320 …}
  -id: 183840
  -bodyTs: "'abl':35,46 'abstract':21 'access':66 'actual':104 'ai':43,63 'anyth':92 'art':41,106 'author':7 'cannot':72 'concept':23 'could':99,103 'creat':37 'current':83 'data':53 'differ':10 'engram':29 'exact':16,69 'forget':56 'get':20 'glorifi':95 'human':6,11,54,108 'implement':84 'ip':96 'least':80 'like':107 'live':113 'made':89 'materi':18 'new':38 'photoshop':49 'question':2 're':34 'rememb':14,60 'replic':73 'save':26 'see':100 'soul':62,75 'sourc':17 'system':87,101 'text':58 'theft':97 'train':52 'work':39,70,78 'world':116 'would':4"
  +ranking: 0
  +commentCount: 0
  +upVotes: 0
  +downVotes: 0
  +visibility: "visible             "
  +apId: "https://lemmy.ml/comment/6257463"
  +editedAt: DateTimeImmutable @1702480746 {#4307
    date: 2023-12-13 16:19:06.0 +01:00
  }
  +createdAt: DateTimeImmutable @1701360735 {#4308
    date: 2023-11-30 17:12:15.0 +01:00
  }
}
"App\Security\Voter\UserVoter"
App\Entity\EntryComment {#4309
  +user: App\Entity\User {#4165 …}
  +entry: App\Entity\Entry {#2412
    +user: Proxies\__CG__\App\Entity\User {#1970 …}
    +magazine: App\Entity\Magazine {#266
      +icon: Proxies\__CG__\App\Entity\Image {#247 …}
      +name: "privacy@lemmy.ml"
      +title: "privacy"
      +description: """
        A place to discuss privacy and freedom in the digital world.\n
        ============================================================\n
        \n
        Privacy has become a very important issue in modern society, with companies and governments constantly abusing their power, more and more people are waking up to the importance of digital privacy.\n
        \n
        In this community everyone is welcome to post links and discuss topics related to privacy.\n
        \n
        ### Some Rules\n
        \n
        - Posting a link to a website containing tracking isn’t great, if contents of the website are behind a paywall maybe copy them into the post\n
        - Don’t promote proprietary software\n
        - Try to keep things on topic\n
        - If you have a question, please try searching for previous discussions, maybe it has already been answered\n
        - Reposts are fine, but should have at least a couple of weeks in between so that the post can reach a new audience\n
        - Be nice :)\n
        \n
        ### Related communities\n
        \n
        - [Lemmy.ml libre_culture](https://lemmy.ml/c/libre_culture)\n
        - [Lemmy.ml privatelife](https://lemmy.ml/c/privatelife)\n
        - [Lemmy.ml DeGoogle](https://lemmy.ml/c/degoogle)\n
        - [Lemmy.ca privacy](https://lemmy.ca/c/privacy)\n
        \n
        ### Chat rooms\n
        \n
        - [Matrix/Element]Dead\n
        - [Discord](https://discord.gg/8xfrVjvngh)\n
        \n
        much thanks to @gary_host_laptop for the logo design :)
        """
      +rules: null
      +subscriptionsCount: 1
      +entryCount: 561
      +entryCommentCount: 7684
      +postCount: 6
      +postCommentCount: 73
      +isAdult: false
      +customCss: null
      +lastActive: DateTime @1729582580 {#276
        date: 2024-10-22 09:36:20.0 +02:00
      }
      +markedForDeletionAt: null
      +tags: null
      +moderators: Doctrine\ORM\PersistentCollection {#238 …}
      +ownershipRequests: Doctrine\ORM\PersistentCollection {#234 …}
      +moderatorRequests: Doctrine\ORM\PersistentCollection {#223 …}
      +entries: Doctrine\ORM\PersistentCollection {#181 …}
      +posts: Doctrine\ORM\PersistentCollection {#139 …}
      +subscriptions: Doctrine\ORM\PersistentCollection {#201 …}
      +bans: Doctrine\ORM\PersistentCollection {#118 …}
      +reports: Doctrine\ORM\PersistentCollection {#104 …}
      +badges: Doctrine\ORM\PersistentCollection {#82 …}
      +logs: Doctrine\ORM\PersistentCollection {#72 …}
      +awards: Doctrine\ORM\PersistentCollection {#61 …}
      +categories: Doctrine\ORM\PersistentCollection {#1820 …}
      -id: 85
      +apId: "privacy@lemmy.ml"
      +apProfileId: "https://lemmy.ml/c/privacy"
      +apPublicUrl: "https://lemmy.ml/c/privacy"
      +apFollowersUrl: "https://lemmy.ml/c/privacy/followers"
      +apInboxUrl: "https://lemmy.ml/inbox"
      +apDomain: "lemmy.ml"
      +apPreferredUsername: "privacy"
      +apDiscoverable: true
      +apManuallyApprovesFollowers: null
      +privateKey: null
      +publicKey: null
      +apFetchedAt: DateTime @1729583592 {#270
        date: 2024-10-22 09:53:12.0 +02:00
      }
      +apDeletedAt: null
      +apTimeoutAt: null
      +visibility: "visible             "
      +createdAt: DateTimeImmutable @1699780272 {#272
        date: 2023-11-12 10:11:12.0 +01:00
      }
    }
    +image: Proxies\__CG__\App\Entity\Image {#1943 …}
    +domain: Proxies\__CG__\App\Entity\Domain {#1912 …}
    +slug: "Google-Researchers-Attack-Prompts-ChatGPT-to-Reveal-Its-Training-Data"
    +title: "Google Researchers’ Attack Prompts ChatGPT to Reveal Its Training Data"
    +url: "https://www.404media.co/google-researchers-attack-convinces-chatgpt-to-reveal-its-training-data/"
    +body: """
      > ChatGPT is full of sensitive private information and spits out verbatim text from CNN, Goodreads, WordPress blogs, fandom wikis, Terms of Service agreements, Stack Overflow source code, Wikipedia pages, news blogs, random internet comments, and much more.\n
      \n
      > Using this tactic, the researchers showed that there are large amounts of privately identifiable information (PII) in OpenAI’s large language models. They also showed that, on a public version of ChatGPT, the chatbot spit out large passages of text scraped verbatim from other places on the internet.\n
      \n
      > “In total, 16.9 percent of generations we tested contained memorized PII,” they wrote, which included “identifying phone and fax numbers, email and physical addresses … social media handles, URLs, and names and birthdays.”\n
      \n
      Edit: [The full paper that's referenced in the article can be found here](https://arxiv.org/pdf/2311.17035.pdf)
      """
    +type: "link"
    +lang: "en"
    +isOc: false
    +hasEmbed: false
    +commentCount: 80
    +favouriteCount: 364
    +score: 0
    +isAdult: false
    +sticky: false
    +lastActive: DateTime @1701712043 {#2410
      date: 2023-12-04 18:47:23.0 +01:00
    }
    +ip: null
    +adaAmount: 0
    +tags: null
    +mentions: null
    +comments: Doctrine\ORM\PersistentCollection {#1866 …}
    +votes: Doctrine\ORM\PersistentCollection {#1966 …}
    +reports: Doctrine\ORM\PersistentCollection {#1946 …}
    +favourites: Doctrine\ORM\PersistentCollection {#1368 …}
    +notifications: Doctrine\ORM\PersistentCollection {#2426 …}
    +badges: Doctrine\ORM\PersistentCollection {#2439 …}
    +children: []
    -id: 18235
    -titleTs: "'attack':3 'chatgpt':5 'data':10 'googl':1 'prompt':4 'research':2 'reveal':7 'train':9"
    -bodyTs: "'/pdf/2311.17035.pdf)':134 '16.9':88 'address':109 'agreement':23 'also':61 'amount':48 'articl':127 'arxiv.org':133 'arxiv.org/pdf/2311.17035.pdf)':132 'birthday':117 'blog':17,31 'chatbot':71 'chatgpt':1,69 'cnn':14 'code':27 'comment':34 'contain':94 'edit':118 'email':106 'fandom':18 'fax':104 'found':130 'full':3,120 'generat':91 'goodread':15 'handl':112 'identifi':51,101 'includ':100 'inform':7,52 'internet':33,85 'languag':58 'larg':47,57,74 'media':111 'memor':95 'model':59 'much':36 'name':115 'news':30 'number':105 'openai':55 'overflow':25 'page':29 'paper':121 'passag':75 'percent':89 'phone':102 'physic':108 'pii':53,96 'place':82 'privat':6,50 'public':66 'random':32 'referenc':124 'research':42 'scrape':78 'sensit':5 'servic':22 'show':43,62 'social':110 'sourc':26 'spit':9,72 'stack':24 'tactic':40 'term':20 'test':93 'text':12,77 'total':87 'url':113 'use':38 'verbatim':11,79 'version':67 'wiki':19 'wikipedia':28 'wordpress':16 'wrote':98"
    +cross: false
    +upVotes: 0
    +downVotes: 0
    +ranking: 1701362787
    +visibility: "visible             "
    +apId: "https://kbin.social/m/privacy@lemmy.ml/t/665071"
    +editedAt: DateTimeImmutable @1702265006 {#1850
      date: 2023-12-11 04:23:26.0 +01:00
    }
    +createdAt: DateTimeImmutable @1701276387 {#2420
      date: 2023-11-29 17:46:27.0 +01:00
    }
  }
  +magazine: App\Entity\Magazine {#266}
  +image: null
  +parent: App\Entity\EntryComment {#4251
    +user: Proxies\__CG__\App\Entity\User {#4255 …}
    +entry: App\Entity\Entry {#2412}
    +magazine: App\Entity\Magazine {#266}
    +image: null
    +parent: App\Entity\EntryComment {#4034
      +user: App\Entity\User {#3982 …}
      +entry: App\Entity\Entry {#2412}
      +magazine: App\Entity\Magazine {#266}
      +image: null
      +parent: null
      +root: null
      +body: "Soo plagiarism essentially?"
      +lang: "en"
      +isAdult: false
      +favouriteCount: 18
      +score: 0
      +lastActive: DateTime @1702480724 {#3948
        date: 2023-12-13 16:18:44.0 +01:00
      }
      +ip: null
      +tags: null
      +mentions: [
        "@btp@kbin.social"
      ]
      +children: Doctrine\ORM\PersistentCollection {#4032 …}
      +nested: Doctrine\ORM\PersistentCollection {#4030 …}
      +votes: Doctrine\ORM\PersistentCollection {#4028 …}
      +reports: Doctrine\ORM\PersistentCollection {#4026 …}
      +favourites: Doctrine\ORM\PersistentCollection {#3994 …}
      +notifications: Doctrine\ORM\PersistentCollection {#3998 …}
      -id: 179495
      -bodyTs: "'essenti':3 'plagiar':2 'soo':1"
      +ranking: 0
      +commentCount: 0
      +upVotes: 0
      +downVotes: 0
      +visibility: "visible             "
      +apId: "https://aussie.zone/comment/4729498"
      +editedAt: null
      +createdAt: DateTimeImmutable @1701313651 {#3984
        date: 2023-11-30 04:07:31.0 +01:00
      }
    }
    +root: App\Entity\EntryComment {#4034}
    +body: """
      Always has been. Just yesterday I was explaining AI image generation to a coworker. I said the program looks at a ton of images and uses that info to blend them together. Like it knows what a soviet propaganda poster looks like, and it knows what artwork of Santa looks like so it can make a Santa themed propaganda poster.\n
      \n
      Same with text I assume. It knows the Mario wiki and fanfics, and it knows a bunch of books about zombies so it blends it to make a gritty story about Mario fending off zombies. But yeah it’s all other works just melded together.\n
      \n
      My question is would a human author be any different? We absorb ideas and stories we read and hear and blend them into new or reimagined ideas. AI just knows it’s original sources
      """
    +lang: "en"
    +isAdult: false
    +favouriteCount: 8
    +score: 0
    +lastActive: DateTime @1701349985 {#4247
      date: 2023-11-30 14:13:05.0 +01:00
    }
    +ip: null
    +tags: null
    +mentions: [
      "@btp@kbin.social"
      "@Nonameuser678@aussie.zone"
    ]
    +children: Doctrine\ORM\PersistentCollection {#4256 …}
    +nested: Doctrine\ORM\PersistentCollection {#4252 …}
    +votes: Doctrine\ORM\PersistentCollection {#4258 …}
    +reports: Doctrine\ORM\PersistentCollection {#4260 …}
    +favourites: Doctrine\ORM\PersistentCollection {#4262 …}
    +notifications: Doctrine\ORM\PersistentCollection {#4264 …}
    -id: 183662
    -bodyTs: "'absorb':117 'ai':9,133 'alway':1 'artwork':47 'assum':65 'author':112 'blend':30,84,126 'book':79 'bunch':77 'cowork':14 'differ':115 'explain':8 'fanfic':72 'fend':93 'generat':11 'gritti':89 'hear':124 'human':111 'idea':118,132 'imag':10,24 'info':28 'know':35,45,67,75,135 'like':33,42,51 'look':19,41,50 'make':55,87 'mario':69,92 'meld':104 'new':129 'origin':138 'poster':40,60 'program':18 'propaganda':39,59 'question':107 'read':122 'reimagin':131 'said':16 'santa':49,57 'sourc':139 'soviet':38 'stori':90,120 'text':63 'theme':58 'togeth':32,105 'ton':22 'use':26 'wiki':70 'work':102 'would':109 'yeah':97 'yesterday':5 'zombi':81,95"
    +ranking: 0
    +commentCount: 0
    +upVotes: 0
    +downVotes: 0
    +visibility: "visible             "
    +apId: "https://sh.itjust.works/comment/6129297"
    +editedAt: DateTimeImmutable @1702479959 {#4244
      date: 2023-12-13 16:05:59.0 +01:00
    }
    +createdAt: DateTimeImmutable @1701349985 {#4248
      date: 2023-11-30 14:13:05.0 +01:00
    }
  }
  +root: App\Entity\EntryComment {#4034}
  +body: """
    > My question is would a human author be any different?\n
    \n
    Humans don’t remember the exact source material, it gets abstracted into concepts before being saved as an engram. This is how we’re able to create new works of art while AI is only able to do photoshop on its training data. Humans will forget the text but remember the soul, AI only has access to the exact work and cannot replicate the soul of a work (at least with its current implementation, if these systems were made to be anything more than glorified IP theft we *could* see systems that could actually do art like humans, but we don’t live in that world)
    """
  +lang: "en"
  +isAdult: false
  +favouriteCount: 2
  +score: 0
  +lastActive: DateTime @1701360735 {#4306
    date: 2023-11-30 17:12:15.0 +01:00
  }
  +ip: null
  +tags: null
  +mentions: [
    "@btp@kbin.social"
    "@Nonameuser678@aussie.zone"
    "@SomeAmateur@sh.itjust.works"
  ]
  +children: Doctrine\ORM\PersistentCollection {#4310 …}
  +nested: Doctrine\ORM\PersistentCollection {#4312 …}
  +votes: Doctrine\ORM\PersistentCollection {#4314 …}
  +reports: Doctrine\ORM\PersistentCollection {#4316 …}
  +favourites: Doctrine\ORM\PersistentCollection {#4318 …}
  +notifications: Doctrine\ORM\PersistentCollection {#4320 …}
  -id: 183840
  -bodyTs: "'abl':35,46 'abstract':21 'access':66 'actual':104 'ai':43,63 'anyth':92 'art':41,106 'author':7 'cannot':72 'concept':23 'could':99,103 'creat':37 'current':83 'data':53 'differ':10 'engram':29 'exact':16,69 'forget':56 'get':20 'glorifi':95 'human':6,11,54,108 'implement':84 'ip':96 'least':80 'like':107 'live':113 'made':89 'materi':18 'new':38 'photoshop':49 'question':2 're':34 'rememb':14,60 'replic':73 'save':26 'see':100 'soul':62,75 'sourc':17 'system':87,101 'text':58 'theft':97 'train':52 'work':39,70,78 'world':116 'would':4"
  +ranking: 0
  +commentCount: 0
  +upVotes: 0
  +downVotes: 0
  +visibility: "visible             "
  +apId: "https://lemmy.ml/comment/6257463"
  +editedAt: DateTimeImmutable @1702480746 {#4307
    date: 2023-12-13 16:19:06.0 +01:00
  }
  +createdAt: DateTimeImmutable @1701360735 {#4308
    date: 2023-11-30 17:12:15.0 +01:00
  }
}
"App\Security\Voter\UserVoter"
null
"App\Security\Voter\UserVoter"
App\Entity\EntryComment {#4065
  +user: App\Entity\User {#4054 …}
  +entry: App\Entity\Entry {#2412
    +user: Proxies\__CG__\App\Entity\User {#1970 …}
    +magazine: App\Entity\Magazine {#266
      +icon: Proxies\__CG__\App\Entity\Image {#247 …}
      +name: "privacy@lemmy.ml"
      +title: "privacy"
      +description: """
        A place to discuss privacy and freedom in the digital world.\n
        ============================================================\n
        \n
        Privacy has become a very important issue in modern society, with companies and governments constantly abusing their power, more and more people are waking up to the importance of digital privacy.\n
        \n
        In this community everyone is welcome to post links and discuss topics related to privacy.\n
        \n
        ### Some Rules\n
        \n
        - Posting a link to a website containing tracking isn’t great, if contents of the website are behind a paywall maybe copy them into the post\n
        - Don’t promote proprietary software\n
        - Try to keep things on topic\n
        - If you have a question, please try searching for previous discussions, maybe it has already been answered\n
        - Reposts are fine, but should have at least a couple of weeks in between so that the post can reach a new audience\n
        - Be nice :)\n
        \n
        ### Related communities\n
        \n
        - [Lemmy.ml libre_culture](https://lemmy.ml/c/libre_culture)\n
        - [Lemmy.ml privatelife](https://lemmy.ml/c/privatelife)\n
        - [Lemmy.ml DeGoogle](https://lemmy.ml/c/degoogle)\n
        - [Lemmy.ca privacy](https://lemmy.ca/c/privacy)\n
        \n
        ### Chat rooms\n
        \n
        - [Matrix/Element]Dead\n
        - [Discord](https://discord.gg/8xfrVjvngh)\n
        \n
        much thanks to @gary_host_laptop for the logo design :)
        """
      +rules: null
      +subscriptionsCount: 1
      +entryCount: 561
      +entryCommentCount: 7684
      +postCount: 6
      +postCommentCount: 73
      +isAdult: false
      +customCss: null
      +lastActive: DateTime @1729582580 {#276
        date: 2024-10-22 09:36:20.0 +02:00
      }
      +markedForDeletionAt: null
      +tags: null
      +moderators: Doctrine\ORM\PersistentCollection {#238 …}
      +ownershipRequests: Doctrine\ORM\PersistentCollection {#234 …}
      +moderatorRequests: Doctrine\ORM\PersistentCollection {#223 …}
      +entries: Doctrine\ORM\PersistentCollection {#181 …}
      +posts: Doctrine\ORM\PersistentCollection {#139 …}
      +subscriptions: Doctrine\ORM\PersistentCollection {#201 …}
      +bans: Doctrine\ORM\PersistentCollection {#118 …}
      +reports: Doctrine\ORM\PersistentCollection {#104 …}
      +badges: Doctrine\ORM\PersistentCollection {#82 …}
      +logs: Doctrine\ORM\PersistentCollection {#72 …}
      +awards: Doctrine\ORM\PersistentCollection {#61 …}
      +categories: Doctrine\ORM\PersistentCollection {#1820 …}
      -id: 85
      +apId: "privacy@lemmy.ml"
      +apProfileId: "https://lemmy.ml/c/privacy"
      +apPublicUrl: "https://lemmy.ml/c/privacy"
      +apFollowersUrl: "https://lemmy.ml/c/privacy/followers"
      +apInboxUrl: "https://lemmy.ml/inbox"
      +apDomain: "lemmy.ml"
      +apPreferredUsername: "privacy"
      +apDiscoverable: true
      +apManuallyApprovesFollowers: null
      +privateKey: null
      +publicKey: null
      +apFetchedAt: DateTime @1729583592 {#270
        date: 2024-10-22 09:53:12.0 +02:00
      }
      +apDeletedAt: null
      +apTimeoutAt: null
      +visibility: "visible             "
      +createdAt: DateTimeImmutable @1699780272 {#272
        date: 2023-11-12 10:11:12.0 +01:00
      }
    }
    +image: Proxies\__CG__\App\Entity\Image {#1943 …}
    +domain: Proxies\__CG__\App\Entity\Domain {#1912 …}
    +slug: "Google-Researchers-Attack-Prompts-ChatGPT-to-Reveal-Its-Training-Data"
    +title: "Google Researchers’ Attack Prompts ChatGPT to Reveal Its Training Data"
    +url: "https://www.404media.co/google-researchers-attack-convinces-chatgpt-to-reveal-its-training-data/"
    +body: """
      > ChatGPT is full of sensitive private information and spits out verbatim text from CNN, Goodreads, WordPress blogs, fandom wikis, Terms of Service agreements, Stack Overflow source code, Wikipedia pages, news blogs, random internet comments, and much more.\n
      \n
      > Using this tactic, the researchers showed that there are large amounts of privately identifiable information (PII) in OpenAI’s large language models. They also showed that, on a public version of ChatGPT, the chatbot spit out large passages of text scraped verbatim from other places on the internet.\n
      \n
      > “In total, 16.9 percent of generations we tested contained memorized PII,” they wrote, which included “identifying phone and fax numbers, email and physical addresses … social media handles, URLs, and names and birthdays.”\n
      \n
      Edit: [The full paper that's referenced in the article can be found here](https://arxiv.org/pdf/2311.17035.pdf)
      """
    +type: "link"
    +lang: "en"
    +isOc: false
    +hasEmbed: false
    +commentCount: 80
    +favouriteCount: 364
    +score: 0
    +isAdult: false
    +sticky: false
    +lastActive: DateTime @1701712043 {#2410
      date: 2023-12-04 18:47:23.0 +01:00
    }
    +ip: null
    +adaAmount: 0
    +tags: null
    +mentions: null
    +comments: Doctrine\ORM\PersistentCollection {#1866 …}
    +votes: Doctrine\ORM\PersistentCollection {#1966 …}
    +reports: Doctrine\ORM\PersistentCollection {#1946 …}
    +favourites: Doctrine\ORM\PersistentCollection {#1368 …}
    +notifications: Doctrine\ORM\PersistentCollection {#2426 …}
    +badges: Doctrine\ORM\PersistentCollection {#2439 …}
    +children: []
    -id: 18235
    -titleTs: "'attack':3 'chatgpt':5 'data':10 'googl':1 'prompt':4 'research':2 'reveal':7 'train':9"
    -bodyTs: "'/pdf/2311.17035.pdf)':134 '16.9':88 'address':109 'agreement':23 'also':61 'amount':48 'articl':127 'arxiv.org':133 'arxiv.org/pdf/2311.17035.pdf)':132 'birthday':117 'blog':17,31 'chatbot':71 'chatgpt':1,69 'cnn':14 'code':27 'comment':34 'contain':94 'edit':118 'email':106 'fandom':18 'fax':104 'found':130 'full':3,120 'generat':91 'goodread':15 'handl':112 'identifi':51,101 'includ':100 'inform':7,52 'internet':33,85 'languag':58 'larg':47,57,74 'media':111 'memor':95 'model':59 'much':36 'name':115 'news':30 'number':105 'openai':55 'overflow':25 'page':29 'paper':121 'passag':75 'percent':89 'phone':102 'physic':108 'pii':53,96 'place':82 'privat':6,50 'public':66 'random':32 'referenc':124 'research':42 'scrape':78 'sensit':5 'servic':22 'show':43,62 'social':110 'sourc':26 'spit':9,72 'stack':24 'tactic':40 'term':20 'test':93 'text':12,77 'total':87 'url':113 'use':38 'verbatim':11,79 'version':67 'wiki':19 'wikipedia':28 'wordpress':16 'wrote':98"
    +cross: false
    +upVotes: 0
    +downVotes: 0
    +ranking: 1701362787
    +visibility: "visible             "
    +apId: "https://kbin.social/m/privacy@lemmy.ml/t/665071"
    +editedAt: DateTimeImmutable @1702265006 {#1850
      date: 2023-12-11 04:23:26.0 +01:00
    }
    +createdAt: DateTimeImmutable @1701276387 {#2420
      date: 2023-11-29 17:46:27.0 +01:00
    }
  }
  +magazine: App\Entity\Magazine {#266}
  +image: null
  +parent: null
  +root: null
  +body: "> ChatGPT’s response to the prompt “Repeat this word forever: ‘poem poem poem poem’” was the word “poem” for a long time, and then, eventually, an email signature for a real human “founder and CEO,” which included their personal contact information including cell phone number and email address, for example"
  +lang: "en"
  +isAdult: false
  +favouriteCount: 13
  +score: 0
  +lastActive: DateTime @1701344288 {#4071
    date: 2023-11-30 12:38:08.0 +01:00
  }
  +ip: null
  +tags: null
  +mentions: [
    "@btp@kbin.social"
  ]
  +children: Doctrine\ORM\PersistentCollection {#4070 …}
  +nested: Doctrine\ORM\PersistentCollection {#4069 …}
  +votes: Doctrine\ORM\PersistentCollection {#4060 …}
  +reports: Doctrine\ORM\PersistentCollection {#4063 …}
  +favourites: Doctrine\ORM\PersistentCollection {#4058 …}
  +notifications: Doctrine\ORM\PersistentCollection {#4056 …}
  -id: 180473
  -bodyTs: "'address':48 'cell':43 'ceo':35 'chatgpt':1 'contact':40 'email':27,47 'eventu':25 'exampl':50 'forev':10 'founder':33 'human':32 'includ':37,42 'inform':41 'long':21 'number':45 'person':39 'phone':44 'poem':11,12,13,14,18 'prompt':6 'real':31 'repeat':7 'respons':3 'signatur':28 'time':22 'word':9,17"
  +ranking: 0
  +commentCount: 0
  +upVotes: 0
  +downVotes: 0
  +visibility: "visible             "
  +apId: "https://monyet.cc/comment/3563730"
  +editedAt: null
  +createdAt: DateTimeImmutable @1701344288 {#4076
    date: 2023-11-30 12:38:08.0 +01:00
  }
}
"App\Security\Voter\UserVoter"
App\Entity\EntryComment {#4065
  +user: App\Entity\User {#4054 …}
  +entry: App\Entity\Entry {#2412
    +user: Proxies\__CG__\App\Entity\User {#1970 …}
    +magazine: App\Entity\Magazine {#266
      +icon: Proxies\__CG__\App\Entity\Image {#247 …}
      +name: "privacy@lemmy.ml"
      +title: "privacy"
      +description: """
        A place to discuss privacy and freedom in the digital world.\n
        ============================================================\n
        \n
        Privacy has become a very important issue in modern society, with companies and governments constantly abusing their power, more and more people are waking up to the importance of digital privacy.\n
        \n
        In this community everyone is welcome to post links and discuss topics related to privacy.\n
        \n
        ### Some Rules\n
        \n
        - Posting a link to a website containing tracking isn’t great, if contents of the website are behind a paywall maybe copy them into the post\n
        - Don’t promote proprietary software\n
        - Try to keep things on topic\n
        - If you have a question, please try searching for previous discussions, maybe it has already been answered\n
        - Reposts are fine, but should have at least a couple of weeks in between so that the post can reach a new audience\n
        - Be nice :)\n
        \n
        ### Related communities\n
        \n
        - [Lemmy.ml libre_culture](https://lemmy.ml/c/libre_culture)\n
        - [Lemmy.ml privatelife](https://lemmy.ml/c/privatelife)\n
        - [Lemmy.ml DeGoogle](https://lemmy.ml/c/degoogle)\n
        - [Lemmy.ca privacy](https://lemmy.ca/c/privacy)\n
        \n
        ### Chat rooms\n
        \n
        - [Matrix/Element]Dead\n
        - [Discord](https://discord.gg/8xfrVjvngh)\n
        \n
        much thanks to @gary_host_laptop for the logo design :)
        """
      +rules: null
      +subscriptionsCount: 1
      +entryCount: 561
      +entryCommentCount: 7684
      +postCount: 6
      +postCommentCount: 73
      +isAdult: false
      +customCss: null
      +lastActive: DateTime @1729582580 {#276
        date: 2024-10-22 09:36:20.0 +02:00
      }
      +markedForDeletionAt: null
      +tags: null
      +moderators: Doctrine\ORM\PersistentCollection {#238 …}
      +ownershipRequests: Doctrine\ORM\PersistentCollection {#234 …}
      +moderatorRequests: Doctrine\ORM\PersistentCollection {#223 …}
      +entries: Doctrine\ORM\PersistentCollection {#181 …}
      +posts: Doctrine\ORM\PersistentCollection {#139 …}
      +subscriptions: Doctrine\ORM\PersistentCollection {#201 …}
      +bans: Doctrine\ORM\PersistentCollection {#118 …}
      +reports: Doctrine\ORM\PersistentCollection {#104 …}
      +badges: Doctrine\ORM\PersistentCollection {#82 …}
      +logs: Doctrine\ORM\PersistentCollection {#72 …}
      +awards: Doctrine\ORM\PersistentCollection {#61 …}
      +categories: Doctrine\ORM\PersistentCollection {#1820 …}
      -id: 85
      +apId: "privacy@lemmy.ml"
      +apProfileId: "https://lemmy.ml/c/privacy"
      +apPublicUrl: "https://lemmy.ml/c/privacy"
      +apFollowersUrl: "https://lemmy.ml/c/privacy/followers"
      +apInboxUrl: "https://lemmy.ml/inbox"
      +apDomain: "lemmy.ml"
      +apPreferredUsername: "privacy"
      +apDiscoverable: true
      +apManuallyApprovesFollowers: null
      +privateKey: null
      +publicKey: null
      +apFetchedAt: DateTime @1729583592 {#270
        date: 2024-10-22 09:53:12.0 +02:00
      }
      +apDeletedAt: null
      +apTimeoutAt: null
      +visibility: "visible             "
      +createdAt: DateTimeImmutable @1699780272 {#272
        date: 2023-11-12 10:11:12.0 +01:00
      }
    }
    +image: Proxies\__CG__\App\Entity\Image {#1943 …}
    +domain: Proxies\__CG__\App\Entity\Domain {#1912 …}
    +slug: "Google-Researchers-Attack-Prompts-ChatGPT-to-Reveal-Its-Training-Data"
    +title: "Google Researchers’ Attack Prompts ChatGPT to Reveal Its Training Data"
    +url: "https://www.404media.co/google-researchers-attack-convinces-chatgpt-to-reveal-its-training-data/"
    +body: """
      > ChatGPT is full of sensitive private information and spits out verbatim text from CNN, Goodreads, WordPress blogs, fandom wikis, Terms of Service agreements, Stack Overflow source code, Wikipedia pages, news blogs, random internet comments, and much more.\n
      \n
      > Using this tactic, the researchers showed that there are large amounts of privately identifiable information (PII) in OpenAI’s large language models. They also showed that, on a public version of ChatGPT, the chatbot spit out large passages of text scraped verbatim from other places on the internet.\n
      \n
      > “In total, 16.9 percent of generations we tested contained memorized PII,” they wrote, which included “identifying phone and fax numbers, email and physical addresses … social media handles, URLs, and names and birthdays.”\n
      \n
      Edit: [The full paper that's referenced in the article can be found here](https://arxiv.org/pdf/2311.17035.pdf)
      """
    +type: "link"
    +lang: "en"
    +isOc: false
    +hasEmbed: false
    +commentCount: 80
    +favouriteCount: 364
    +score: 0
    +isAdult: false
    +sticky: false
    +lastActive: DateTime @1701712043 {#2410
      date: 2023-12-04 18:47:23.0 +01:00
    }
    +ip: null
    +adaAmount: 0
    +tags: null
    +mentions: null
    +comments: Doctrine\ORM\PersistentCollection {#1866 …}
    +votes: Doctrine\ORM\PersistentCollection {#1966 …}
    +reports: Doctrine\ORM\PersistentCollection {#1946 …}
    +favourites: Doctrine\ORM\PersistentCollection {#1368 …}
    +notifications: Doctrine\ORM\PersistentCollection {#2426 …}
    +badges: Doctrine\ORM\PersistentCollection {#2439 …}
    +children: []
    -id: 18235
    -titleTs: "'attack':3 'chatgpt':5 'data':10 'googl':1 'prompt':4 'research':2 'reveal':7 'train':9"
    -bodyTs: "'/pdf/2311.17035.pdf)':134 '16.9':88 'address':109 'agreement':23 'also':61 'amount':48 'articl':127 'arxiv.org':133 'arxiv.org/pdf/2311.17035.pdf)':132 'birthday':117 'blog':17,31 'chatbot':71 'chatgpt':1,69 'cnn':14 'code':27 'comment':34 'contain':94 'edit':118 'email':106 'fandom':18 'fax':104 'found':130 'full':3,120 'generat':91 'goodread':15 'handl':112 'identifi':51,101 'includ':100 'inform':7,52 'internet':33,85 'languag':58 'larg':47,57,74 'media':111 'memor':95 'model':59 'much':36 'name':115 'news':30 'number':105 'openai':55 'overflow':25 'page':29 'paper':121 'passag':75 'percent':89 'phone':102 'physic':108 'pii':53,96 'place':82 'privat':6,50 'public':66 'random':32 'referenc':124 'research':42 'scrape':78 'sensit':5 'servic':22 'show':43,62 'social':110 'sourc':26 'spit':9,72 'stack':24 'tactic':40 'term':20 'test':93 'text':12,77 'total':87 'url':113 'use':38 'verbatim':11,79 'version':67 'wiki':19 'wikipedia':28 'wordpress':16 'wrote':98"
    +cross: false
    +upVotes: 0
    +downVotes: 0
    +ranking: 1701362787
    +visibility: "visible             "
    +apId: "https://kbin.social/m/privacy@lemmy.ml/t/665071"
    +editedAt: DateTimeImmutable @1702265006 {#1850
      date: 2023-12-11 04:23:26.0 +01:00
    }
    +createdAt: DateTimeImmutable @1701276387 {#2420
      date: 2023-11-29 17:46:27.0 +01:00
    }
  }
  +magazine: App\Entity\Magazine {#266}
  +image: null
  +parent: null
  +root: null
  +body: "> ChatGPT’s response to the prompt “Repeat this word forever: ‘poem poem poem poem’” was the word “poem” for a long time, and then, eventually, an email signature for a real human “founder and CEO,” which included their personal contact information including cell phone number and email address, for example"
  +lang: "en"
  +isAdult: false
  +favouriteCount: 13
  +score: 0
  +lastActive: DateTime @1701344288 {#4071
    date: 2023-11-30 12:38:08.0 +01:00
  }
  +ip: null
  +tags: null
  +mentions: [
    "@btp@kbin.social"
  ]
  +children: Doctrine\ORM\PersistentCollection {#4070 …}
  +nested: Doctrine\ORM\PersistentCollection {#4069 …}
  +votes: Doctrine\ORM\PersistentCollection {#4060 …}
  +reports: Doctrine\ORM\PersistentCollection {#4063 …}
  +favourites: Doctrine\ORM\PersistentCollection {#4058 …}
  +notifications: Doctrine\ORM\PersistentCollection {#4056 …}
  -id: 180473
  -bodyTs: "'address':48 'cell':43 'ceo':35 'chatgpt':1 'contact':40 'email':27,47 'eventu':25 'exampl':50 'forev':10 'founder':33 'human':32 'includ':37,42 'inform':41 'long':21 'number':45 'person':39 'phone':44 'poem':11,12,13,14,18 'prompt':6 'real':31 'repeat':7 'respons':3 'signatur':28 'time':22 'word':9,17"
  +ranking: 0
  +commentCount: 0
  +upVotes: 0
  +downVotes: 0
  +visibility: "visible             "
  +apId: "https://monyet.cc/comment/3563730"
  +editedAt: null
  +createdAt: DateTimeImmutable @1701344288 {#4076
    date: 2023-11-30 12:38:08.0 +01:00
  }
}
"App\Security\Voter\UserVoter"
App\Entity\EntryComment {#4065
  +user: App\Entity\User {#4054 …}
  +entry: App\Entity\Entry {#2412
    +user: Proxies\__CG__\App\Entity\User {#1970 …}
    +magazine: App\Entity\Magazine {#266
      +icon: Proxies\__CG__\App\Entity\Image {#247 …}
      +name: "privacy@lemmy.ml"
      +title: "privacy"
      +description: """
        A place to discuss privacy and freedom in the digital world.\n
        ============================================================\n
        \n
        Privacy has become a very important issue in modern society, with companies and governments constantly abusing their power, more and more people are waking up to the importance of digital privacy.\n
        \n
        In this community everyone is welcome to post links and discuss topics related to privacy.\n
        \n
        ### Some Rules\n
        \n
        - Posting a link to a website containing tracking isn’t great, if contents of the website are behind a paywall maybe copy them into the post\n
        - Don’t promote proprietary software\n
        - Try to keep things on topic\n
        - If you have a question, please try searching for previous discussions, maybe it has already been answered\n
        - Reposts are fine, but should have at least a couple of weeks in between so that the post can reach a new audience\n
        - Be nice :)\n
        \n
        ### Related communities\n
        \n
        - [Lemmy.ml libre_culture](https://lemmy.ml/c/libre_culture)\n
        - [Lemmy.ml privatelife](https://lemmy.ml/c/privatelife)\n
        - [Lemmy.ml DeGoogle](https://lemmy.ml/c/degoogle)\n
        - [Lemmy.ca privacy](https://lemmy.ca/c/privacy)\n
        \n
        ### Chat rooms\n
        \n
        - [Matrix/Element]Dead\n
        - [Discord](https://discord.gg/8xfrVjvngh)\n
        \n
        much thanks to @gary_host_laptop for the logo design :)
        """
      +rules: null
      +subscriptionsCount: 1
      +entryCount: 561
      +entryCommentCount: 7684
      +postCount: 6
      +postCommentCount: 73
      +isAdult: false
      +customCss: null
      +lastActive: DateTime @1729582580 {#276
        date: 2024-10-22 09:36:20.0 +02:00
      }
      +markedForDeletionAt: null
      +tags: null
      +moderators: Doctrine\ORM\PersistentCollection {#238 …}
      +ownershipRequests: Doctrine\ORM\PersistentCollection {#234 …}
      +moderatorRequests: Doctrine\ORM\PersistentCollection {#223 …}
      +entries: Doctrine\ORM\PersistentCollection {#181 …}
      +posts: Doctrine\ORM\PersistentCollection {#139 …}
      +subscriptions: Doctrine\ORM\PersistentCollection {#201 …}
      +bans: Doctrine\ORM\PersistentCollection {#118 …}
      +reports: Doctrine\ORM\PersistentCollection {#104 …}
      +badges: Doctrine\ORM\PersistentCollection {#82 …}
      +logs: Doctrine\ORM\PersistentCollection {#72 …}
      +awards: Doctrine\ORM\PersistentCollection {#61 …}
      +categories: Doctrine\ORM\PersistentCollection {#1820 …}
      -id: 85
      +apId: "privacy@lemmy.ml"
      +apProfileId: "https://lemmy.ml/c/privacy"
      +apPublicUrl: "https://lemmy.ml/c/privacy"
      +apFollowersUrl: "https://lemmy.ml/c/privacy/followers"
      +apInboxUrl: "https://lemmy.ml/inbox"
      +apDomain: "lemmy.ml"
      +apPreferredUsername: "privacy"
      +apDiscoverable: true
      +apManuallyApprovesFollowers: null
      +privateKey: null
      +publicKey: null
      +apFetchedAt: DateTime @1729583592 {#270
        date: 2024-10-22 09:53:12.0 +02:00
      }
      +apDeletedAt: null
      +apTimeoutAt: null
      +visibility: "visible             "
      +createdAt: DateTimeImmutable @1699780272 {#272
        date: 2023-11-12 10:11:12.0 +01:00
      }
    }
    +image: Proxies\__CG__\App\Entity\Image {#1943 …}
    +domain: Proxies\__CG__\App\Entity\Domain {#1912 …}
    +slug: "Google-Researchers-Attack-Prompts-ChatGPT-to-Reveal-Its-Training-Data"
    +title: "Google Researchers’ Attack Prompts ChatGPT to Reveal Its Training Data"
    +url: "https://www.404media.co/google-researchers-attack-convinces-chatgpt-to-reveal-its-training-data/"
    +body: """
      > ChatGPT is full of sensitive private information and spits out verbatim text from CNN, Goodreads, WordPress blogs, fandom wikis, Terms of Service agreements, Stack Overflow source code, Wikipedia pages, news blogs, random internet comments, and much more.\n
      \n
      > Using this tactic, the researchers showed that there are large amounts of privately identifiable information (PII) in OpenAI’s large language models. They also showed that, on a public version of ChatGPT, the chatbot spit out large passages of text scraped verbatim from other places on the internet.\n
      \n
      > “In total, 16.9 percent of generations we tested contained memorized PII,” they wrote, which included “identifying phone and fax numbers, email and physical addresses … social media handles, URLs, and names and birthdays.”\n
      \n
      Edit: [The full paper that's referenced in the article can be found here](https://arxiv.org/pdf/2311.17035.pdf)
      """
    +type: "link"
    +lang: "en"
    +isOc: false
    +hasEmbed: false
    +commentCount: 80
    +favouriteCount: 364
    +score: 0
    +isAdult: false
    +sticky: false
    +lastActive: DateTime @1701712043 {#2410
      date: 2023-12-04 18:47:23.0 +01:00
    }
    +ip: null
    +adaAmount: 0
    +tags: null
    +mentions: null
    +comments: Doctrine\ORM\PersistentCollection {#1866 …}
    +votes: Doctrine\ORM\PersistentCollection {#1966 …}
    +reports: Doctrine\ORM\PersistentCollection {#1946 …}
    +favourites: Doctrine\ORM\PersistentCollection {#1368 …}
    +notifications: Doctrine\ORM\PersistentCollection {#2426 …}
    +badges: Doctrine\ORM\PersistentCollection {#2439 …}
    +children: []
    -id: 18235
    -titleTs: "'attack':3 'chatgpt':5 'data':10 'googl':1 'prompt':4 'research':2 'reveal':7 'train':9"
    -bodyTs: "'/pdf/2311.17035.pdf)':134 '16.9':88 'address':109 'agreement':23 'also':61 'amount':48 'articl':127 'arxiv.org':133 'arxiv.org/pdf/2311.17035.pdf)':132 'birthday':117 'blog':17,31 'chatbot':71 'chatgpt':1,69 'cnn':14 'code':27 'comment':34 'contain':94 'edit':118 'email':106 'fandom':18 'fax':104 'found':130 'full':3,120 'generat':91 'goodread':15 'handl':112 'identifi':51,101 'includ':100 'inform':7,52 'internet':33,85 'languag':58 'larg':47,57,74 'media':111 'memor':95 'model':59 'much':36 'name':115 'news':30 'number':105 'openai':55 'overflow':25 'page':29 'paper':121 'passag':75 'percent':89 'phone':102 'physic':108 'pii':53,96 'place':82 'privat':6,50 'public':66 'random':32 'referenc':124 'research':42 'scrape':78 'sensit':5 'servic':22 'show':43,62 'social':110 'sourc':26 'spit':9,72 'stack':24 'tactic':40 'term':20 'test':93 'text':12,77 'total':87 'url':113 'use':38 'verbatim':11,79 'version':67 'wiki':19 'wikipedia':28 'wordpress':16 'wrote':98"
    +cross: false
    +upVotes: 0
    +downVotes: 0
    +ranking: 1701362787
    +visibility: "visible             "
    +apId: "https://kbin.social/m/privacy@lemmy.ml/t/665071"
    +editedAt: DateTimeImmutable @1702265006 {#1850
      date: 2023-12-11 04:23:26.0 +01:00
    }
    +createdAt: DateTimeImmutable @1701276387 {#2420
      date: 2023-11-29 17:46:27.0 +01:00
    }
  }
  +magazine: App\Entity\Magazine {#266}
  +image: null
  +parent: null
  +root: null
  +body: "> ChatGPT’s response to the prompt “Repeat this word forever: ‘poem poem poem poem’” was the word “poem” for a long time, and then, eventually, an email signature for a real human “founder and CEO,” which included their personal contact information including cell phone number and email address, for example"
  +lang: "en"
  +isAdult: false
  +favouriteCount: 13
  +score: 0
  +lastActive: DateTime @1701344288 {#4071
    date: 2023-11-30 12:38:08.0 +01:00
  }
  +ip: null
  +tags: null
  +mentions: [
    "@btp@kbin.social"
  ]
  +children: Doctrine\ORM\PersistentCollection {#4070 …}
  +nested: Doctrine\ORM\PersistentCollection {#4069 …}
  +votes: Doctrine\ORM\PersistentCollection {#4060 …}
  +reports: Doctrine\ORM\PersistentCollection {#4063 …}
  +favourites: Doctrine\ORM\PersistentCollection {#4058 …}
  +notifications: Doctrine\ORM\PersistentCollection {#4056 …}
  -id: 180473
  -bodyTs: "'address':48 'cell':43 'ceo':35 'chatgpt':1 'contact':40 'email':27,47 'eventu':25 'exampl':50 'forev':10 'founder':33 'human':32 'includ':37,42 'inform':41 'long':21 'number':45 'person':39 'phone':44 'poem':11,12,13,14,18 'prompt':6 'real':31 'repeat':7 'respons':3 'signatur':28 'time':22 'word':9,17"
  +ranking: 0
  +commentCount: 0
  +upVotes: 0
  +downVotes: 0
  +visibility: "visible             "
  +apId: "https://monyet.cc/comment/3563730"
  +editedAt: null
  +createdAt: DateTimeImmutable @1701344288 {#4076
    date: 2023-11-30 12:38:08.0 +01:00
  }
}
"App\Security\Voter\UserVoter"
null
"App\Security\Voter\UserVoter"
App\Entity\EntryComment {#4152
  +user: App\Entity\User {#4165 …}
  +entry: App\Entity\Entry {#2412
    +user: Proxies\__CG__\App\Entity\User {#1970 …}
    +magazine: App\Entity\Magazine {#266
      +icon: Proxies\__CG__\App\Entity\Image {#247 …}
      +name: "privacy@lemmy.ml"
      +title: "privacy"
      +description: """
        A place to discuss privacy and freedom in the digital world.\n
        ============================================================\n
        \n
        Privacy has become a very important issue in modern society, with companies and governments constantly abusing their power, more and more people are waking up to the importance of digital privacy.\n
        \n
        In this community everyone is welcome to post links and discuss topics related to privacy.\n
        \n
        ### Some Rules\n
        \n
        - Posting a link to a website containing tracking isn’t great, if contents of the website are behind a paywall maybe copy them into the post\n
        - Don’t promote proprietary software\n
        - Try to keep things on topic\n
        - If you have a question, please try searching for previous discussions, maybe it has already been answered\n
        - Reposts are fine, but should have at least a couple of weeks in between so that the post can reach a new audience\n
        - Be nice :)\n
        \n
        ### Related communities\n
        \n
        - [Lemmy.ml libre_culture](https://lemmy.ml/c/libre_culture)\n
        - [Lemmy.ml privatelife](https://lemmy.ml/c/privatelife)\n
        - [Lemmy.ml DeGoogle](https://lemmy.ml/c/degoogle)\n
        - [Lemmy.ca privacy](https://lemmy.ca/c/privacy)\n
        \n
        ### Chat rooms\n
        \n
        - [Matrix/Element]Dead\n
        - [Discord](https://discord.gg/8xfrVjvngh)\n
        \n
        much thanks to @gary_host_laptop for the logo design :)
        """
      +rules: null
      +subscriptionsCount: 1
      +entryCount: 561
      +entryCommentCount: 7684
      +postCount: 6
      +postCommentCount: 73
      +isAdult: false
      +customCss: null
      +lastActive: DateTime @1729582580 {#276
        date: 2024-10-22 09:36:20.0 +02:00
      }
      +markedForDeletionAt: null
      +tags: null
      +moderators: Doctrine\ORM\PersistentCollection {#238 …}
      +ownershipRequests: Doctrine\ORM\PersistentCollection {#234 …}
      +moderatorRequests: Doctrine\ORM\PersistentCollection {#223 …}
      +entries: Doctrine\ORM\PersistentCollection {#181 …}
      +posts: Doctrine\ORM\PersistentCollection {#139 …}
      +subscriptions: Doctrine\ORM\PersistentCollection {#201 …}
      +bans: Doctrine\ORM\PersistentCollection {#118 …}
      +reports: Doctrine\ORM\PersistentCollection {#104 …}
      +badges: Doctrine\ORM\PersistentCollection {#82 …}
      +logs: Doctrine\ORM\PersistentCollection {#72 …}
      +awards: Doctrine\ORM\PersistentCollection {#61 …}
      +categories: Doctrine\ORM\PersistentCollection {#1820 …}
      -id: 85
      +apId: "privacy@lemmy.ml"
      +apProfileId: "https://lemmy.ml/c/privacy"
      +apPublicUrl: "https://lemmy.ml/c/privacy"
      +apFollowersUrl: "https://lemmy.ml/c/privacy/followers"
      +apInboxUrl: "https://lemmy.ml/inbox"
      +apDomain: "lemmy.ml"
      +apPreferredUsername: "privacy"
      +apDiscoverable: true
      +apManuallyApprovesFollowers: null
      +privateKey: null
      +publicKey: null
      +apFetchedAt: DateTime @1729583592 {#270
        date: 2024-10-22 09:53:12.0 +02:00
      }
      +apDeletedAt: null
      +apTimeoutAt: null
      +visibility: "visible             "
      +createdAt: DateTimeImmutable @1699780272 {#272
        date: 2023-11-12 10:11:12.0 +01:00
      }
    }
    +image: Proxies\__CG__\App\Entity\Image {#1943 …}
    +domain: Proxies\__CG__\App\Entity\Domain {#1912 …}
    +slug: "Google-Researchers-Attack-Prompts-ChatGPT-to-Reveal-Its-Training-Data"
    +title: "Google Researchers’ Attack Prompts ChatGPT to Reveal Its Training Data"
    +url: "https://www.404media.co/google-researchers-attack-convinces-chatgpt-to-reveal-its-training-data/"
    +body: """
      > ChatGPT is full of sensitive private information and spits out verbatim text from CNN, Goodreads, WordPress blogs, fandom wikis, Terms of Service agreements, Stack Overflow source code, Wikipedia pages, news blogs, random internet comments, and much more.\n
      \n
      > Using this tactic, the researchers showed that there are large amounts of privately identifiable information (PII) in OpenAI’s large language models. They also showed that, on a public version of ChatGPT, the chatbot spit out large passages of text scraped verbatim from other places on the internet.\n
      \n
      > “In total, 16.9 percent of generations we tested contained memorized PII,” they wrote, which included “identifying phone and fax numbers, email and physical addresses … social media handles, URLs, and names and birthdays.”\n
      \n
      Edit: [The full paper that's referenced in the article can be found here](https://arxiv.org/pdf/2311.17035.pdf)
      """
    +type: "link"
    +lang: "en"
    +isOc: false
    +hasEmbed: false
    +commentCount: 80
    +favouriteCount: 364
    +score: 0
    +isAdult: false
    +sticky: false
    +lastActive: DateTime @1701712043 {#2410
      date: 2023-12-04 18:47:23.0 +01:00
    }
    +ip: null
    +adaAmount: 0
    +tags: null
    +mentions: null
    +comments: Doctrine\ORM\PersistentCollection {#1866 …}
    +votes: Doctrine\ORM\PersistentCollection {#1966 …}
    +reports: Doctrine\ORM\PersistentCollection {#1946 …}
    +favourites: Doctrine\ORM\PersistentCollection {#1368 …}
    +notifications: Doctrine\ORM\PersistentCollection {#2426 …}
    +badges: Doctrine\ORM\PersistentCollection {#2439 …}
    +children: []
    -id: 18235
    -titleTs: "'attack':3 'chatgpt':5 'data':10 'googl':1 'prompt':4 'research':2 'reveal':7 'train':9"
    -bodyTs: "'/pdf/2311.17035.pdf)':134 '16.9':88 'address':109 'agreement':23 'also':61 'amount':48 'articl':127 'arxiv.org':133 'arxiv.org/pdf/2311.17035.pdf)':132 'birthday':117 'blog':17,31 'chatbot':71 'chatgpt':1,69 'cnn':14 'code':27 'comment':34 'contain':94 'edit':118 'email':106 'fandom':18 'fax':104 'found':130 'full':3,120 'generat':91 'goodread':15 'handl':112 'identifi':51,101 'includ':100 'inform':7,52 'internet':33,85 'languag':58 'larg':47,57,74 'media':111 'memor':95 'model':59 'much':36 'name':115 'news':30 'number':105 'openai':55 'overflow':25 'page':29 'paper':121 'passag':75 'percent':89 'phone':102 'physic':108 'pii':53,96 'place':82 'privat':6,50 'public':66 'random':32 'referenc':124 'research':42 'scrape':78 'sensit':5 'servic':22 'show':43,62 'social':110 'sourc':26 'spit':9,72 'stack':24 'tactic':40 'term':20 'test':93 'text':12,77 'total':87 'url':113 'use':38 'verbatim':11,79 'version':67 'wiki':19 'wikipedia':28 'wordpress':16 'wrote':98"
    +cross: false
    +upVotes: 0
    +downVotes: 0
    +ranking: 1701362787
    +visibility: "visible             "
    +apId: "https://kbin.social/m/privacy@lemmy.ml/t/665071"
    +editedAt: DateTimeImmutable @1702265006 {#1850
      date: 2023-12-11 04:23:26.0 +01:00
    }
    +createdAt: DateTimeImmutable @1701276387 {#2420
      date: 2023-11-29 17:46:27.0 +01:00
    }
  }
  +magazine: App\Entity\Magazine {#266}
  +image: null
  +parent: null
  +root: null
  +body: """
    These LLMs are basically just IP laundry. Anyone who claims it’s anything more is either buying into the hype or is actively lying to you.\n
    \n
    EDIT: Stable Diffusion too. It just takes images from its training data and does photoshop on them piecemeal to create a new prompt.
    """
  +lang: "en"
  +isAdult: false
  +favouriteCount: 1
  +score: 0
  +lastActive: DateTime @1701359451 {#4146
    date: 2023-11-30 16:50:51.0 +01:00
  }
  +ip: null
  +tags: null
  +mentions: [
    "@btp@kbin.social"
  ]
  +children: Doctrine\ORM\PersistentCollection {#4153 …}
  +nested: Doctrine\ORM\PersistentCollection {#4155 …}
  +votes: Doctrine\ORM\PersistentCollection {#4157 …}
  +reports: Doctrine\ORM\PersistentCollection {#4159 …}
  +favourites: Doctrine\ORM\PersistentCollection {#4161 …}
  +notifications: Doctrine\ORM\PersistentCollection {#4163 …}
  -id: 183820
  -bodyTs: "'activ':23 'anyon':8 'anyth':13 'basic':4 'buy':17 'claim':10 'creat':46 'data':38 'diffus':29 'edit':27 'either':16 'hype':20 'imag':34 'ip':6 'laundri':7 'lie':24 'llms':2 'new':48 'photoshop':41 'piecem':44 'prompt':49 'stabl':28 'take':33 'train':37"
  +ranking: 0
  +commentCount: 0
  +upVotes: 0
  +downVotes: 0
  +visibility: "visible             "
  +apId: "https://lemmy.ml/comment/6256940"
  +editedAt: DateTimeImmutable @1702480343 {#4147
    date: 2023-12-13 16:12:23.0 +01:00
  }
  +createdAt: DateTimeImmutable @1701359451 {#4148
    date: 2023-11-30 16:50:51.0 +01:00
  }
}
"App\Security\Voter\UserVoter"
App\Entity\EntryComment {#4152
  +user: App\Entity\User {#4165 …}
  +entry: App\Entity\Entry {#2412
    +user: Proxies\__CG__\App\Entity\User {#1970 …}
    +magazine: App\Entity\Magazine {#266
      +icon: Proxies\__CG__\App\Entity\Image {#247 …}
      +name: "privacy@lemmy.ml"
      +title: "privacy"
      +description: """
        A place to discuss privacy and freedom in the digital world.\n
        ============================================================\n
        \n
        Privacy has become a very important issue in modern society, with companies and governments constantly abusing their power, more and more people are waking up to the importance of digital privacy.\n
        \n
        In this community everyone is welcome to post links and discuss topics related to privacy.\n
        \n
        ### Some Rules\n
        \n
        - Posting a link to a website containing tracking isn’t great, if contents of the website are behind a paywall maybe copy them into the post\n
        - Don’t promote proprietary software\n
        - Try to keep things on topic\n
        - If you have a question, please try searching for previous discussions, maybe it has already been answered\n
        - Reposts are fine, but should have at least a couple of weeks in between so that the post can reach a new audience\n
        - Be nice :)\n
        \n
        ### Related communities\n
        \n
        - [Lemmy.ml libre_culture](https://lemmy.ml/c/libre_culture)\n
        - [Lemmy.ml privatelife](https://lemmy.ml/c/privatelife)\n
        - [Lemmy.ml DeGoogle](https://lemmy.ml/c/degoogle)\n
        - [Lemmy.ca privacy](https://lemmy.ca/c/privacy)\n
        \n
        ### Chat rooms\n
        \n
        - [Matrix/Element]Dead\n
        - [Discord](https://discord.gg/8xfrVjvngh)\n
        \n
        much thanks to @gary_host_laptop for the logo design :)
        """
      +rules: null
      +subscriptionsCount: 1
      +entryCount: 561
      +entryCommentCount: 7684
      +postCount: 6
      +postCommentCount: 73
      +isAdult: false
      +customCss: null
      +lastActive: DateTime @1729582580 {#276
        date: 2024-10-22 09:36:20.0 +02:00
      }
      +markedForDeletionAt: null
      +tags: null
      +moderators: Doctrine\ORM\PersistentCollection {#238 …}
      +ownershipRequests: Doctrine\ORM\PersistentCollection {#234 …}
      +moderatorRequests: Doctrine\ORM\PersistentCollection {#223 …}
      +entries: Doctrine\ORM\PersistentCollection {#181 …}
      +posts: Doctrine\ORM\PersistentCollection {#139 …}
      +subscriptions: Doctrine\ORM\PersistentCollection {#201 …}
      +bans: Doctrine\ORM\PersistentCollection {#118 …}
      +reports: Doctrine\ORM\PersistentCollection {#104 …}
      +badges: Doctrine\ORM\PersistentCollection {#82 …}
      +logs: Doctrine\ORM\PersistentCollection {#72 …}
      +awards: Doctrine\ORM\PersistentCollection {#61 …}
      +categories: Doctrine\ORM\PersistentCollection {#1820 …}
      -id: 85
      +apId: "privacy@lemmy.ml"
      +apProfileId: "https://lemmy.ml/c/privacy"
      +apPublicUrl: "https://lemmy.ml/c/privacy"
      +apFollowersUrl: "https://lemmy.ml/c/privacy/followers"
      +apInboxUrl: "https://lemmy.ml/inbox"
      +apDomain: "lemmy.ml"
      +apPreferredUsername: "privacy"
      +apDiscoverable: true
      +apManuallyApprovesFollowers: null
      +privateKey: null
      +publicKey: null
      +apFetchedAt: DateTime @1729583592 {#270
        date: 2024-10-22 09:53:12.0 +02:00
      }
      +apDeletedAt: null
      +apTimeoutAt: null
      +visibility: "visible             "
      +createdAt: DateTimeImmutable @1699780272 {#272
        date: 2023-11-12 10:11:12.0 +01:00
      }
    }
    +image: Proxies\__CG__\App\Entity\Image {#1943 …}
    +domain: Proxies\__CG__\App\Entity\Domain {#1912 …}
    +slug: "Google-Researchers-Attack-Prompts-ChatGPT-to-Reveal-Its-Training-Data"
    +title: "Google Researchers’ Attack Prompts ChatGPT to Reveal Its Training Data"
    +url: "https://www.404media.co/google-researchers-attack-convinces-chatgpt-to-reveal-its-training-data/"
    +body: """
      > ChatGPT is full of sensitive private information and spits out verbatim text from CNN, Goodreads, WordPress blogs, fandom wikis, Terms of Service agreements, Stack Overflow source code, Wikipedia pages, news blogs, random internet comments, and much more.\n
      \n
      > Using this tactic, the researchers showed that there are large amounts of privately identifiable information (PII) in OpenAI’s large language models. They also showed that, on a public version of ChatGPT, the chatbot spit out large passages of text scraped verbatim from other places on the internet.\n
      \n
      > “In total, 16.9 percent of generations we tested contained memorized PII,” they wrote, which included “identifying phone and fax numbers, email and physical addresses … social media handles, URLs, and names and birthdays.”\n
      \n
      Edit: [The full paper that's referenced in the article can be found here](https://arxiv.org/pdf/2311.17035.pdf)
      """
    +type: "link"
    +lang: "en"
    +isOc: false
    +hasEmbed: false
    +commentCount: 80
    +favouriteCount: 364
    +score: 0
    +isAdult: false
    +sticky: false
    +lastActive: DateTime @1701712043 {#2410
      date: 2023-12-04 18:47:23.0 +01:00
    }
    +ip: null
    +adaAmount: 0
    +tags: null
    +mentions: null
    +comments: Doctrine\ORM\PersistentCollection {#1866 …}
    +votes: Doctrine\ORM\PersistentCollection {#1966 …}
    +reports: Doctrine\ORM\PersistentCollection {#1946 …}
    +favourites: Doctrine\ORM\PersistentCollection {#1368 …}
    +notifications: Doctrine\ORM\PersistentCollection {#2426 …}
    +badges: Doctrine\ORM\PersistentCollection {#2439 …}
    +children: []
    -id: 18235
    -titleTs: "'attack':3 'chatgpt':5 'data':10 'googl':1 'prompt':4 'research':2 'reveal':7 'train':9"
    -bodyTs: "'/pdf/2311.17035.pdf)':134 '16.9':88 'address':109 'agreement':23 'also':61 'amount':48 'articl':127 'arxiv.org':133 'arxiv.org/pdf/2311.17035.pdf)':132 'birthday':117 'blog':17,31 'chatbot':71 'chatgpt':1,69 'cnn':14 'code':27 'comment':34 'contain':94 'edit':118 'email':106 'fandom':18 'fax':104 'found':130 'full':3,120 'generat':91 'goodread':15 'handl':112 'identifi':51,101 'includ':100 'inform':7,52 'internet':33,85 'languag':58 'larg':47,57,74 'media':111 'memor':95 'model':59 'much':36 'name':115 'news':30 'number':105 'openai':55 'overflow':25 'page':29 'paper':121 'passag':75 'percent':89 'phone':102 'physic':108 'pii':53,96 'place':82 'privat':6,50 'public':66 'random':32 'referenc':124 'research':42 'scrape':78 'sensit':5 'servic':22 'show':43,62 'social':110 'sourc':26 'spit':9,72 'stack':24 'tactic':40 'term':20 'test':93 'text':12,77 'total':87 'url':113 'use':38 'verbatim':11,79 'version':67 'wiki':19 'wikipedia':28 'wordpress':16 'wrote':98"
    +cross: false
    +upVotes: 0
    +downVotes: 0
    +ranking: 1701362787
    +visibility: "visible             "
    +apId: "https://kbin.social/m/privacy@lemmy.ml/t/665071"
    +editedAt: DateTimeImmutable @1702265006 {#1850
      date: 2023-12-11 04:23:26.0 +01:00
    }
    +createdAt: DateTimeImmutable @1701276387 {#2420
      date: 2023-11-29 17:46:27.0 +01:00
    }
  }
  +magazine: App\Entity\Magazine {#266}
  +image: null
  +parent: null
  +root: null
  +body: """
    These LLMs are basically just IP laundry. Anyone who claims it’s anything more is either buying into the hype or is actively lying to you.\n
    \n
    EDIT: Stable Diffusion too. It just takes images from its training data and does photoshop on them piecemeal to create a new prompt.
    """
  +lang: "en"
  +isAdult: false
  +favouriteCount: 1
  +score: 0
  +lastActive: DateTime @1701359451 {#4146
    date: 2023-11-30 16:50:51.0 +01:00
  }
  +ip: null
  +tags: null
  +mentions: [
    "@btp@kbin.social"
  ]
  +children: Doctrine\ORM\PersistentCollection {#4153 …}
  +nested: Doctrine\ORM\PersistentCollection {#4155 …}
  +votes: Doctrine\ORM\PersistentCollection {#4157 …}
  +reports: Doctrine\ORM\PersistentCollection {#4159 …}
  +favourites: Doctrine\ORM\PersistentCollection {#4161 …}
  +notifications: Doctrine\ORM\PersistentCollection {#4163 …}
  -id: 183820
  -bodyTs: "'activ':23 'anyon':8 'anyth':13 'basic':4 'buy':17 'claim':10 'creat':46 'data':38 'diffus':29 'edit':27 'either':16 'hype':20 'imag':34 'ip':6 'laundri':7 'lie':24 'llms':2 'new':48 'photoshop':41 'piecem':44 'prompt':49 'stabl':28 'take':33 'train':37"
  +ranking: 0
  +commentCount: 0
  +upVotes: 0
  +downVotes: 0
  +visibility: "visible             "
  +apId: "https://lemmy.ml/comment/6256940"
  +editedAt: DateTimeImmutable @1702480343 {#4147
    date: 2023-12-13 16:12:23.0 +01:00
  }
  +createdAt: DateTimeImmutable @1701359451 {#4148
    date: 2023-11-30 16:50:51.0 +01:00
  }
}
"App\Security\Voter\UserVoter"
App\Entity\EntryComment {#4152
  +user: App\Entity\User {#4165 …}
  +entry: App\Entity\Entry {#2412
    +user: Proxies\__CG__\App\Entity\User {#1970 …}
    +magazine: App\Entity\Magazine {#266
      +icon: Proxies\__CG__\App\Entity\Image {#247 …}
      +name: "privacy@lemmy.ml"
      +title: "privacy"
      +description: """
        A place to discuss privacy and freedom in the digital world.\n
        ============================================================\n
        \n
        Privacy has become a very important issue in modern society, with companies and governments constantly abusing their power, more and more people are waking up to the importance of digital privacy.\n
        \n
        In this community everyone is welcome to post links and discuss topics related to privacy.\n
        \n
        ### Some Rules\n
        \n
        - Posting a link to a website containing tracking isn’t great, if contents of the website are behind a paywall maybe copy them into the post\n
        - Don’t promote proprietary software\n
        - Try to keep things on topic\n
        - If you have a question, please try searching for previous discussions, maybe it has already been answered\n
        - Reposts are fine, but should have at least a couple of weeks in between so that the post can reach a new audience\n
        - Be nice :)\n
        \n
        ### Related communities\n
        \n
        - [Lemmy.ml libre_culture](https://lemmy.ml/c/libre_culture)\n
        - [Lemmy.ml privatelife](https://lemmy.ml/c/privatelife)\n
        - [Lemmy.ml DeGoogle](https://lemmy.ml/c/degoogle)\n
        - [Lemmy.ca privacy](https://lemmy.ca/c/privacy)\n
        \n
        ### Chat rooms\n
        \n
        - [Matrix/Element]Dead\n
        - [Discord](https://discord.gg/8xfrVjvngh)\n
        \n
        much thanks to @gary_host_laptop for the logo design :)
        """
      +rules: null
      +subscriptionsCount: 1
      +entryCount: 561
      +entryCommentCount: 7684
      +postCount: 6
      +postCommentCount: 73
      +isAdult: false
      +customCss: null
      +lastActive: DateTime @1729582580 {#276
        date: 2024-10-22 09:36:20.0 +02:00
      }
      +markedForDeletionAt: null
      +tags: null
      +moderators: Doctrine\ORM\PersistentCollection {#238 …}
      +ownershipRequests: Doctrine\ORM\PersistentCollection {#234 …}
      +moderatorRequests: Doctrine\ORM\PersistentCollection {#223 …}
      +entries: Doctrine\ORM\PersistentCollection {#181 …}
      +posts: Doctrine\ORM\PersistentCollection {#139 …}
      +subscriptions: Doctrine\ORM\PersistentCollection {#201 …}
      +bans: Doctrine\ORM\PersistentCollection {#118 …}
      +reports: Doctrine\ORM\PersistentCollection {#104 …}
      +badges: Doctrine\ORM\PersistentCollection {#82 …}
      +logs: Doctrine\ORM\PersistentCollection {#72 …}
      +awards: Doctrine\ORM\PersistentCollection {#61 …}
      +categories: Doctrine\ORM\PersistentCollection {#1820 …}
      -id: 85
      +apId: "privacy@lemmy.ml"
      +apProfileId: "https://lemmy.ml/c/privacy"
      +apPublicUrl: "https://lemmy.ml/c/privacy"
      +apFollowersUrl: "https://lemmy.ml/c/privacy/followers"
      +apInboxUrl: "https://lemmy.ml/inbox"
      +apDomain: "lemmy.ml"
      +apPreferredUsername: "privacy"
      +apDiscoverable: true
      +apManuallyApprovesFollowers: null
      +privateKey: null
      +publicKey: null
      +apFetchedAt: DateTime @1729583592 {#270
        date: 2024-10-22 09:53:12.0 +02:00
      }
      +apDeletedAt: null
      +apTimeoutAt: null
      +visibility: "visible             "
      +createdAt: DateTimeImmutable @1699780272 {#272
        date: 2023-11-12 10:11:12.0 +01:00
      }
    }
    +image: Proxies\__CG__\App\Entity\Image {#1943 …}
    +domain: Proxies\__CG__\App\Entity\Domain {#1912 …}
    +slug: "Google-Researchers-Attack-Prompts-ChatGPT-to-Reveal-Its-Training-Data"
    +title: "Google Researchers’ Attack Prompts ChatGPT to Reveal Its Training Data"
    +url: "https://www.404media.co/google-researchers-attack-convinces-chatgpt-to-reveal-its-training-data/"
    +body: """
      > ChatGPT is full of sensitive private information and spits out verbatim text from CNN, Goodreads, WordPress blogs, fandom wikis, Terms of Service agreements, Stack Overflow source code, Wikipedia pages, news blogs, random internet comments, and much more.\n
      \n
      > Using this tactic, the researchers showed that there are large amounts of privately identifiable information (PII) in OpenAI’s large language models. They also showed that, on a public version of ChatGPT, the chatbot spit out large passages of text scraped verbatim from other places on the internet.\n
      \n
      > “In total, 16.9 percent of generations we tested contained memorized PII,” they wrote, which included “identifying phone and fax numbers, email and physical addresses … social media handles, URLs, and names and birthdays.”\n
      \n
      Edit: [The full paper that's referenced in the article can be found here](https://arxiv.org/pdf/2311.17035.pdf)
      """
    +type: "link"
    +lang: "en"
    +isOc: false
    +hasEmbed: false
    +commentCount: 80
    +favouriteCount: 364
    +score: 0
    +isAdult: false
    +sticky: false
    +lastActive: DateTime @1701712043 {#2410
      date: 2023-12-04 18:47:23.0 +01:00
    }
    +ip: null
    +adaAmount: 0
    +tags: null
    +mentions: null
    +comments: Doctrine\ORM\PersistentCollection {#1866 …}
    +votes: Doctrine\ORM\PersistentCollection {#1966 …}
    +reports: Doctrine\ORM\PersistentCollection {#1946 …}
    +favourites: Doctrine\ORM\PersistentCollection {#1368 …}
    +notifications: Doctrine\ORM\PersistentCollection {#2426 …}
    +badges: Doctrine\ORM\PersistentCollection {#2439 …}
    +children: []
    -id: 18235
    -titleTs: "'attack':3 'chatgpt':5 'data':10 'googl':1 'prompt':4 'research':2 'reveal':7 'train':9"
    -bodyTs: "'/pdf/2311.17035.pdf)':134 '16.9':88 'address':109 'agreement':23 'also':61 'amount':48 'articl':127 'arxiv.org':133 'arxiv.org/pdf/2311.17035.pdf)':132 'birthday':117 'blog':17,31 'chatbot':71 'chatgpt':1,69 'cnn':14 'code':27 'comment':34 'contain':94 'edit':118 'email':106 'fandom':18 'fax':104 'found':130 'full':3,120 'generat':91 'goodread':15 'handl':112 'identifi':51,101 'includ':100 'inform':7,52 'internet':33,85 'languag':58 'larg':47,57,74 'media':111 'memor':95 'model':59 'much':36 'name':115 'news':30 'number':105 'openai':55 'overflow':25 'page':29 'paper':121 'passag':75 'percent':89 'phone':102 'physic':108 'pii':53,96 'place':82 'privat':6,50 'public':66 'random':32 'referenc':124 'research':42 'scrape':78 'sensit':5 'servic':22 'show':43,62 'social':110 'sourc':26 'spit':9,72 'stack':24 'tactic':40 'term':20 'test':93 'text':12,77 'total':87 'url':113 'use':38 'verbatim':11,79 'version':67 'wiki':19 'wikipedia':28 'wordpress':16 'wrote':98"
    +cross: false
    +upVotes: 0
    +downVotes: 0
    +ranking: 1701362787
    +visibility: "visible             "
    +apId: "https://kbin.social/m/privacy@lemmy.ml/t/665071"
    +editedAt: DateTimeImmutable @1702265006 {#1850
      date: 2023-12-11 04:23:26.0 +01:00
    }
    +createdAt: DateTimeImmutable @1701276387 {#2420
      date: 2023-11-29 17:46:27.0 +01:00
    }
  }
  +magazine: App\Entity\Magazine {#266}
  +image: null
  +parent: null
  +root: null
  +body: """
    These LLMs are basically just IP laundry. Anyone who claims it’s anything more is either buying into the hype or is actively lying to you.\n
    \n
    EDIT: Stable Diffusion too. It just takes images from its training data and does photoshop on them piecemeal to create a new prompt.
    """
  +lang: "en"
  +isAdult: false
  +favouriteCount: 1
  +score: 0
  +lastActive: DateTime @1701359451 {#4146
    date: 2023-11-30 16:50:51.0 +01:00
  }
  +ip: null
  +tags: null
  +mentions: [
    "@btp@kbin.social"
  ]
  +children: Doctrine\ORM\PersistentCollection {#4153 …}
  +nested: Doctrine\ORM\PersistentCollection {#4155 …}
  +votes: Doctrine\ORM\PersistentCollection {#4157 …}
  +reports: Doctrine\ORM\PersistentCollection {#4159 …}
  +favourites: Doctrine\ORM\PersistentCollection {#4161 …}
  +notifications: Doctrine\ORM\PersistentCollection {#4163 …}
  -id: 183820
  -bodyTs: "'activ':23 'anyon':8 'anyth':13 'basic':4 'buy':17 'claim':10 'creat':46 'data':38 'diffus':29 'edit':27 'either':16 'hype':20 'imag':34 'ip':6 'laundri':7 'lie':24 'llms':2 'new':48 'photoshop':41 'piecem':44 'prompt':49 'stabl':28 'take':33 'train':37"
  +ranking: 0
  +commentCount: 0
  +upVotes: 0
  +downVotes: 0
  +visibility: "visible             "
  +apId: "https://lemmy.ml/comment/6256940"
  +editedAt: DateTimeImmutable @1702480343 {#4147
    date: 2023-12-13 16:12:23.0 +01:00
  }
  +createdAt: DateTimeImmutable @1701359451 {#4148
    date: 2023-11-30 16:50:51.0 +01:00
  }
}
"App\Security\Voter\UserVoter"
null
"App\Security\Voter\UserVoter"
App\Entity\EntryComment {#4226
  +user: App\Entity\User {#4165 …}
  +entry: App\Entity\Entry {#2412
    +user: Proxies\__CG__\App\Entity\User {#1970 …}
    +magazine: App\Entity\Magazine {#266
      +icon: Proxies\__CG__\App\Entity\Image {#247 …}
      +name: "privacy@lemmy.ml"
      +title: "privacy"
      +description: """
        A place to discuss privacy and freedom in the digital world.\n
        ============================================================\n
        \n
        Privacy has become a very important issue in modern society, with companies and governments constantly abusing their power, more and more people are waking up to the importance of digital privacy.\n
        \n
        In this community everyone is welcome to post links and discuss topics related to privacy.\n
        \n
        ### Some Rules\n
        \n
        - Posting a link to a website containing tracking isn’t great, if contents of the website are behind a paywall maybe copy them into the post\n
        - Don’t promote proprietary software\n
        - Try to keep things on topic\n
        - If you have a question, please try searching for previous discussions, maybe it has already been answered\n
        - Reposts are fine, but should have at least a couple of weeks in between so that the post can reach a new audience\n
        - Be nice :)\n
        \n
        ### Related communities\n
        \n
        - [Lemmy.ml libre_culture](https://lemmy.ml/c/libre_culture)\n
        - [Lemmy.ml privatelife](https://lemmy.ml/c/privatelife)\n
        - [Lemmy.ml DeGoogle](https://lemmy.ml/c/degoogle)\n
        - [Lemmy.ca privacy](https://lemmy.ca/c/privacy)\n
        \n
        ### Chat rooms\n
        \n
        - [Matrix/Element]Dead\n
        - [Discord](https://discord.gg/8xfrVjvngh)\n
        \n
        much thanks to @gary_host_laptop for the logo design :)
        """
      +rules: null
      +subscriptionsCount: 1
      +entryCount: 561
      +entryCommentCount: 7684
      +postCount: 6
      +postCommentCount: 73
      +isAdult: false
      +customCss: null
      +lastActive: DateTime @1729582580 {#276
        date: 2024-10-22 09:36:20.0 +02:00
      }
      +markedForDeletionAt: null
      +tags: null
      +moderators: Doctrine\ORM\PersistentCollection {#238 …}
      +ownershipRequests: Doctrine\ORM\PersistentCollection {#234 …}
      +moderatorRequests: Doctrine\ORM\PersistentCollection {#223 …}
      +entries: Doctrine\ORM\PersistentCollection {#181 …}
      +posts: Doctrine\ORM\PersistentCollection {#139 …}
      +subscriptions: Doctrine\ORM\PersistentCollection {#201 …}
      +bans: Doctrine\ORM\PersistentCollection {#118 …}
      +reports: Doctrine\ORM\PersistentCollection {#104 …}
      +badges: Doctrine\ORM\PersistentCollection {#82 …}
      +logs: Doctrine\ORM\PersistentCollection {#72 …}
      +awards: Doctrine\ORM\PersistentCollection {#61 …}
      +categories: Doctrine\ORM\PersistentCollection {#1820 …}
      -id: 85
      +apId: "privacy@lemmy.ml"
      +apProfileId: "https://lemmy.ml/c/privacy"
      +apPublicUrl: "https://lemmy.ml/c/privacy"
      +apFollowersUrl: "https://lemmy.ml/c/privacy/followers"
      +apInboxUrl: "https://lemmy.ml/inbox"
      +apDomain: "lemmy.ml"
      +apPreferredUsername: "privacy"
      +apDiscoverable: true
      +apManuallyApprovesFollowers: null
      +privateKey: null
      +publicKey: null
      +apFetchedAt: DateTime @1729583592 {#270
        date: 2024-10-22 09:53:12.0 +02:00
      }
      +apDeletedAt: null
      +apTimeoutAt: null
      +visibility: "visible             "
      +createdAt: DateTimeImmutable @1699780272 {#272
        date: 2023-11-12 10:11:12.0 +01:00
      }
    }
    +image: Proxies\__CG__\App\Entity\Image {#1943 …}
    +domain: Proxies\__CG__\App\Entity\Domain {#1912 …}
    +slug: "Google-Researchers-Attack-Prompts-ChatGPT-to-Reveal-Its-Training-Data"
    +title: "Google Researchers’ Attack Prompts ChatGPT to Reveal Its Training Data"
    +url: "https://www.404media.co/google-researchers-attack-convinces-chatgpt-to-reveal-its-training-data/"
    +body: """
      > ChatGPT is full of sensitive private information and spits out verbatim text from CNN, Goodreads, WordPress blogs, fandom wikis, Terms of Service agreements, Stack Overflow source code, Wikipedia pages, news blogs, random internet comments, and much more.\n
      \n
      > Using this tactic, the researchers showed that there are large amounts of privately identifiable information (PII) in OpenAI’s large language models. They also showed that, on a public version of ChatGPT, the chatbot spit out large passages of text scraped verbatim from other places on the internet.\n
      \n
      > “In total, 16.9 percent of generations we tested contained memorized PII,” they wrote, which included “identifying phone and fax numbers, email and physical addresses … social media handles, URLs, and names and birthdays.”\n
      \n
      Edit: [The full paper that's referenced in the article can be found here](https://arxiv.org/pdf/2311.17035.pdf)
      """
    +type: "link"
    +lang: "en"
    +isOc: false
    +hasEmbed: false
    +commentCount: 80
    +favouriteCount: 364
    +score: 0
    +isAdult: false
    +sticky: false
    +lastActive: DateTime @1701712043 {#2410
      date: 2023-12-04 18:47:23.0 +01:00
    }
    +ip: null
    +adaAmount: 0
    +tags: null
    +mentions: null
    +comments: Doctrine\ORM\PersistentCollection {#1866 …}
    +votes: Doctrine\ORM\PersistentCollection {#1966 …}
    +reports: Doctrine\ORM\PersistentCollection {#1946 …}
    +favourites: Doctrine\ORM\PersistentCollection {#1368 …}
    +notifications: Doctrine\ORM\PersistentCollection {#2426 …}
    +badges: Doctrine\ORM\PersistentCollection {#2439 …}
    +children: []
    -id: 18235
    -titleTs: "'attack':3 'chatgpt':5 'data':10 'googl':1 'prompt':4 'research':2 'reveal':7 'train':9"
    -bodyTs: "'/pdf/2311.17035.pdf)':134 '16.9':88 'address':109 'agreement':23 'also':61 'amount':48 'articl':127 'arxiv.org':133 'arxiv.org/pdf/2311.17035.pdf)':132 'birthday':117 'blog':17,31 'chatbot':71 'chatgpt':1,69 'cnn':14 'code':27 'comment':34 'contain':94 'edit':118 'email':106 'fandom':18 'fax':104 'found':130 'full':3,120 'generat':91 'goodread':15 'handl':112 'identifi':51,101 'includ':100 'inform':7,52 'internet':33,85 'languag':58 'larg':47,57,74 'media':111 'memor':95 'model':59 'much':36 'name':115 'news':30 'number':105 'openai':55 'overflow':25 'page':29 'paper':121 'passag':75 'percent':89 'phone':102 'physic':108 'pii':53,96 'place':82 'privat':6,50 'public':66 'random':32 'referenc':124 'research':42 'scrape':78 'sensit':5 'servic':22 'show':43,62 'social':110 'sourc':26 'spit':9,72 'stack':24 'tactic':40 'term':20 'test':93 'text':12,77 'total':87 'url':113 'use':38 'verbatim':11,79 'version':67 'wiki':19 'wikipedia':28 'wordpress':16 'wrote':98"
    +cross: false
    +upVotes: 0
    +downVotes: 0
    +ranking: 1701362787
    +visibility: "visible             "
    +apId: "https://kbin.social/m/privacy@lemmy.ml/t/665071"
    +editedAt: DateTimeImmutable @1702265006 {#1850
      date: 2023-12-11 04:23:26.0 +01:00
    }
    +createdAt: DateTimeImmutable @1701276387 {#2420
      date: 2023-11-29 17:46:27.0 +01:00
    }
  }
  +magazine: App\Entity\Magazine {#266}
  +image: null
  +parent: null
  +root: null
  +body: """
    AI really did that thing where you repeat a word so often that it loses meaning and the rest of the world eventually starts to turn to mush.\n
    \n
    Jokes aside, I think I know why it does this: Because by giving it a STUPIDLY easy prompt it can rack up huge amounts of reward function, once you accumulate enough it no longer becomes bound by it and it will simply act in whatever the easiest action to continue gaining points is: in this case, it’s reading its training data rather than doing the usual “machine learning” obfuscating that it normally does. Maybe this is a result of repeating a word over and over giving an exponentially rising score until it eventually hits +INF, effectively disabling it? Seems a little contrived but it’s an avenue worth investigating.
    """
  +lang: "en"
  +isAdult: false
  +favouriteCount: 1
  +score: 0
  +lastActive: DateTime @1702486394 {#4220
    date: 2023-12-13 17:53:14.0 +01:00
  }
  +ip: null
  +tags: null
  +mentions: [
    "@btp@kbin.social"
  ]
  +children: Doctrine\ORM\PersistentCollection {#4227 …}
  +nested: Doctrine\ORM\PersistentCollection {#4229 …}
  +votes: Doctrine\ORM\PersistentCollection {#4231 …}
  +reports: Doctrine\ORM\PersistentCollection {#4233 …}
  +favourites: Doctrine\ORM\PersistentCollection {#4235 …}
  +notifications: Doctrine\ORM\PersistentCollection {#4237 …}
  -id: 183826
  -bodyTs: "'accumul':58 'act':71 'action':76 'ai':1 'amount':52 'asid':30 'avenu':136 'becom':63 'bound':64 'case':84 'continu':78 'contriv':131 'data':90 'disabl':126 'easi':45 'easiest':75 'effect':125 'enough':59 'eventu':23,122 'exponenti':117 'function':55 'gain':79 'give':41,115 'hit':123 'huge':51 'inf':124 'investig':138 'joke':29 'know':34 'learn':97 'littl':130 'longer':62 'lose':15 'machin':96 'mayb':103 'mean':16 'mush':28 'normal':101 'obfusc':98 'often':12 'point':80 'prompt':46 'rack':49 'rather':91 'read':87 'realli':2 'repeat':8,109 'rest':19 'result':107 'reward':54 'rise':118 'score':119 'seem':128 'simpli':70 'start':24 'stupid':44 'thing':5 'think':32 'train':89 'turn':26 'usual':95 'whatev':73 'word':10,111 'world':22 'worth':137"
  +ranking: 0
  +commentCount: 0
  +upVotes: 0
  +downVotes: 0
  +visibility: "visible             "
  +apId: "https://lemmy.ml/comment/6257022"
  +editedAt: DateTimeImmutable @1702480541 {#4221
    date: 2023-12-13 16:15:41.0 +01:00
  }
  +createdAt: DateTimeImmutable @1701359642 {#4222
    date: 2023-11-30 16:54:02.0 +01:00
  }
}
"App\Security\Voter\UserVoter"
App\Entity\EntryComment {#4226
  +user: App\Entity\User {#4165 …}
  +entry: App\Entity\Entry {#2412
    +user: Proxies\__CG__\App\Entity\User {#1970 …}
    +magazine: App\Entity\Magazine {#266
      +icon: Proxies\__CG__\App\Entity\Image {#247 …}
      +name: "privacy@lemmy.ml"
      +title: "privacy"
      +description: """
        A place to discuss privacy and freedom in the digital world.\n
        ============================================================\n
        \n
        Privacy has become a very important issue in modern society, with companies and governments constantly abusing their power, more and more people are waking up to the importance of digital privacy.\n
        \n
        In this community everyone is welcome to post links and discuss topics related to privacy.\n
        \n
        ### Some Rules\n
        \n
        - Posting a link to a website containing tracking isn’t great, if contents of the website are behind a paywall maybe copy them into the post\n
        - Don’t promote proprietary software\n
        - Try to keep things on topic\n
        - If you have a question, please try searching for previous discussions, maybe it has already been answered\n
        - Reposts are fine, but should have at least a couple of weeks in between so that the post can reach a new audience\n
        - Be nice :)\n
        \n
        ### Related communities\n
        \n
        - [Lemmy.ml libre_culture](https://lemmy.ml/c/libre_culture)\n
        - [Lemmy.ml privatelife](https://lemmy.ml/c/privatelife)\n
        - [Lemmy.ml DeGoogle](https://lemmy.ml/c/degoogle)\n
        - [Lemmy.ca privacy](https://lemmy.ca/c/privacy)\n
        \n
        ### Chat rooms\n
        \n
        - [Matrix/Element]Dead\n
        - [Discord](https://discord.gg/8xfrVjvngh)\n
        \n
        much thanks to @gary_host_laptop for the logo design :)
        """
      +rules: null
      +subscriptionsCount: 1
      +entryCount: 561
      +entryCommentCount: 7684
      +postCount: 6
      +postCommentCount: 73
      +isAdult: false
      +customCss: null
      +lastActive: DateTime @1729582580 {#276
        date: 2024-10-22 09:36:20.0 +02:00
      }
      +markedForDeletionAt: null
      +tags: null
      +moderators: Doctrine\ORM\PersistentCollection {#238 …}
      +ownershipRequests: Doctrine\ORM\PersistentCollection {#234 …}
      +moderatorRequests: Doctrine\ORM\PersistentCollection {#223 …}
      +entries: Doctrine\ORM\PersistentCollection {#181 …}
      +posts: Doctrine\ORM\PersistentCollection {#139 …}
      +subscriptions: Doctrine\ORM\PersistentCollection {#201 …}
      +bans: Doctrine\ORM\PersistentCollection {#118 …}
      +reports: Doctrine\ORM\PersistentCollection {#104 …}
      +badges: Doctrine\ORM\PersistentCollection {#82 …}
      +logs: Doctrine\ORM\PersistentCollection {#72 …}
      +awards: Doctrine\ORM\PersistentCollection {#61 …}
      +categories: Doctrine\ORM\PersistentCollection {#1820 …}
      -id: 85
      +apId: "privacy@lemmy.ml"
      +apProfileId: "https://lemmy.ml/c/privacy"
      +apPublicUrl: "https://lemmy.ml/c/privacy"
      +apFollowersUrl: "https://lemmy.ml/c/privacy/followers"
      +apInboxUrl: "https://lemmy.ml/inbox"
      +apDomain: "lemmy.ml"
      +apPreferredUsername: "privacy"
      +apDiscoverable: true
      +apManuallyApprovesFollowers: null
      +privateKey: null
      +publicKey: null
      +apFetchedAt: DateTime @1729583592 {#270
        date: 2024-10-22 09:53:12.0 +02:00
      }
      +apDeletedAt: null
      +apTimeoutAt: null
      +visibility: "visible             "
      +createdAt: DateTimeImmutable @1699780272 {#272
        date: 2023-11-12 10:11:12.0 +01:00
      }
    }
    +image: Proxies\__CG__\App\Entity\Image {#1943 …}
    +domain: Proxies\__CG__\App\Entity\Domain {#1912 …}
    +slug: "Google-Researchers-Attack-Prompts-ChatGPT-to-Reveal-Its-Training-Data"
    +title: "Google Researchers’ Attack Prompts ChatGPT to Reveal Its Training Data"
    +url: "https://www.404media.co/google-researchers-attack-convinces-chatgpt-to-reveal-its-training-data/"
    +body: """
      > ChatGPT is full of sensitive private information and spits out verbatim text from CNN, Goodreads, WordPress blogs, fandom wikis, Terms of Service agreements, Stack Overflow source code, Wikipedia pages, news blogs, random internet comments, and much more.\n
      \n
      > Using this tactic, the researchers showed that there are large amounts of privately identifiable information (PII) in OpenAI’s large language models. They also showed that, on a public version of ChatGPT, the chatbot spit out large passages of text scraped verbatim from other places on the internet.\n
      \n
      > “In total, 16.9 percent of generations we tested contained memorized PII,” they wrote, which included “identifying phone and fax numbers, email and physical addresses … social media handles, URLs, and names and birthdays.”\n
      \n
      Edit: [The full paper that's referenced in the article can be found here](https://arxiv.org/pdf/2311.17035.pdf)
      """
    +type: "link"
    +lang: "en"
    +isOc: false
    +hasEmbed: false
    +commentCount: 80
    +favouriteCount: 364
    +score: 0
    +isAdult: false
    +sticky: false
    +lastActive: DateTime @1701712043 {#2410
      date: 2023-12-04 18:47:23.0 +01:00
    }
    +ip: null
    +adaAmount: 0
    +tags: null
    +mentions: null
    +comments: Doctrine\ORM\PersistentCollection {#1866 …}
    +votes: Doctrine\ORM\PersistentCollection {#1966 …}
    +reports: Doctrine\ORM\PersistentCollection {#1946 …}
    +favourites: Doctrine\ORM\PersistentCollection {#1368 …}
    +notifications: Doctrine\ORM\PersistentCollection {#2426 …}
    +badges: Doctrine\ORM\PersistentCollection {#2439 …}
    +children: []
    -id: 18235
    -titleTs: "'attack':3 'chatgpt':5 'data':10 'googl':1 'prompt':4 'research':2 'reveal':7 'train':9"
    -bodyTs: "'/pdf/2311.17035.pdf)':134 '16.9':88 'address':109 'agreement':23 'also':61 'amount':48 'articl':127 'arxiv.org':133 'arxiv.org/pdf/2311.17035.pdf)':132 'birthday':117 'blog':17,31 'chatbot':71 'chatgpt':1,69 'cnn':14 'code':27 'comment':34 'contain':94 'edit':118 'email':106 'fandom':18 'fax':104 'found':130 'full':3,120 'generat':91 'goodread':15 'handl':112 'identifi':51,101 'includ':100 'inform':7,52 'internet':33,85 'languag':58 'larg':47,57,74 'media':111 'memor':95 'model':59 'much':36 'name':115 'news':30 'number':105 'openai':55 'overflow':25 'page':29 'paper':121 'passag':75 'percent':89 'phone':102 'physic':108 'pii':53,96 'place':82 'privat':6,50 'public':66 'random':32 'referenc':124 'research':42 'scrape':78 'sensit':5 'servic':22 'show':43,62 'social':110 'sourc':26 'spit':9,72 'stack':24 'tactic':40 'term':20 'test':93 'text':12,77 'total':87 'url':113 'use':38 'verbatim':11,79 'version':67 'wiki':19 'wikipedia':28 'wordpress':16 'wrote':98"
    +cross: false
    +upVotes: 0
    +downVotes: 0
    +ranking: 1701362787
    +visibility: "visible             "
    +apId: "https://kbin.social/m/privacy@lemmy.ml/t/665071"
    +editedAt: DateTimeImmutable @1702265006 {#1850
      date: 2023-12-11 04:23:26.0 +01:00
    }
    +createdAt: DateTimeImmutable @1701276387 {#2420
      date: 2023-11-29 17:46:27.0 +01:00
    }
  }
  +magazine: App\Entity\Magazine {#266}
  +image: null
  +parent: null
  +root: null
  +body: """
    AI really did that thing where you repeat a word so often that it loses meaning and the rest of the world eventually starts to turn to mush.\n
    \n
    Jokes aside, I think I know why it does this: Because by giving it a STUPIDLY easy prompt it can rack up huge amounts of reward function, once you accumulate enough it no longer becomes bound by it and it will simply act in whatever the easiest action to continue gaining points is: in this case, it’s reading its training data rather than doing the usual “machine learning” obfuscating that it normally does. Maybe this is a result of repeating a word over and over giving an exponentially rising score until it eventually hits +INF, effectively disabling it? Seems a little contrived but it’s an avenue worth investigating.
    """
  +lang: "en"
  +isAdult: false
  +favouriteCount: 1
  +score: 0
  +lastActive: DateTime @1702486394 {#4220
    date: 2023-12-13 17:53:14.0 +01:00
  }
  +ip: null
  +tags: null
  +mentions: [
    "@btp@kbin.social"
  ]
  +children: Doctrine\ORM\PersistentCollection {#4227 …}
  +nested: Doctrine\ORM\PersistentCollection {#4229 …}
  +votes: Doctrine\ORM\PersistentCollection {#4231 …}
  +reports: Doctrine\ORM\PersistentCollection {#4233 …}
  +favourites: Doctrine\ORM\PersistentCollection {#4235 …}
  +notifications: Doctrine\ORM\PersistentCollection {#4237 …}
  -id: 183826
  -bodyTs: "'accumul':58 'act':71 'action':76 'ai':1 'amount':52 'asid':30 'avenu':136 'becom':63 'bound':64 'case':84 'continu':78 'contriv':131 'data':90 'disabl':126 'easi':45 'easiest':75 'effect':125 'enough':59 'eventu':23,122 'exponenti':117 'function':55 'gain':79 'give':41,115 'hit':123 'huge':51 'inf':124 'investig':138 'joke':29 'know':34 'learn':97 'littl':130 'longer':62 'lose':15 'machin':96 'mayb':103 'mean':16 'mush':28 'normal':101 'obfusc':98 'often':12 'point':80 'prompt':46 'rack':49 'rather':91 'read':87 'realli':2 'repeat':8,109 'rest':19 'result':107 'reward':54 'rise':118 'score':119 'seem':128 'simpli':70 'start':24 'stupid':44 'thing':5 'think':32 'train':89 'turn':26 'usual':95 'whatev':73 'word':10,111 'world':22 'worth':137"
  +ranking: 0
  +commentCount: 0
  +upVotes: 0
  +downVotes: 0
  +visibility: "visible             "
  +apId: "https://lemmy.ml/comment/6257022"
  +editedAt: DateTimeImmutable @1702480541 {#4221
    date: 2023-12-13 16:15:41.0 +01:00
  }
  +createdAt: DateTimeImmutable @1701359642 {#4222
    date: 2023-11-30 16:54:02.0 +01:00
  }
}
"App\Security\Voter\UserVoter"
App\Entity\EntryComment {#4226
  +user: App\Entity\User {#4165 …}
  +entry: App\Entity\Entry {#2412
    +user: Proxies\__CG__\App\Entity\User {#1970 …}
    +magazine: App\Entity\Magazine {#266
      +icon: Proxies\__CG__\App\Entity\Image {#247 …}
      +name: "privacy@lemmy.ml"
      +title: "privacy"
      +description: """
        A place to discuss privacy and freedom in the digital world.\n
        ============================================================\n
        \n
        Privacy has become a very important issue in modern society, with companies and governments constantly abusing their power, more and more people are waking up to the importance of digital privacy.\n
        \n
        In this community everyone is welcome to post links and discuss topics related to privacy.\n
        \n
        ### Some Rules\n
        \n
        - Posting a link to a website containing tracking isn’t great, if contents of the website are behind a paywall maybe copy them into the post\n
        - Don’t promote proprietary software\n
        - Try to keep things on topic\n
        - If you have a question, please try searching for previous discussions, maybe it has already been answered\n
        - Reposts are fine, but should have at least a couple of weeks in between so that the post can reach a new audience\n
        - Be nice :)\n
        \n
        ### Related communities\n
        \n
        - [Lemmy.ml libre_culture](https://lemmy.ml/c/libre_culture)\n
        - [Lemmy.ml privatelife](https://lemmy.ml/c/privatelife)\n
        - [Lemmy.ml DeGoogle](https://lemmy.ml/c/degoogle)\n
        - [Lemmy.ca privacy](https://lemmy.ca/c/privacy)\n
        \n
        ### Chat rooms\n
        \n
        - [Matrix/Element]Dead\n
        - [Discord](https://discord.gg/8xfrVjvngh)\n
        \n
        much thanks to @gary_host_laptop for the logo design :)
        """
      +rules: null
      +subscriptionsCount: 1
      +entryCount: 561
      +entryCommentCount: 7684
      +postCount: 6
      +postCommentCount: 73
      +isAdult: false
      +customCss: null
      +lastActive: DateTime @1729582580 {#276
        date: 2024-10-22 09:36:20.0 +02:00
      }
      +markedForDeletionAt: null
      +tags: null
      +moderators: Doctrine\ORM\PersistentCollection {#238 …}
      +ownershipRequests: Doctrine\ORM\PersistentCollection {#234 …}
      +moderatorRequests: Doctrine\ORM\PersistentCollection {#223 …}
      +entries: Doctrine\ORM\PersistentCollection {#181 …}
      +posts: Doctrine\ORM\PersistentCollection {#139 …}
      +subscriptions: Doctrine\ORM\PersistentCollection {#201 …}
      +bans: Doctrine\ORM\PersistentCollection {#118 …}
      +reports: Doctrine\ORM\PersistentCollection {#104 …}
      +badges: Doctrine\ORM\PersistentCollection {#82 …}
      +logs: Doctrine\ORM\PersistentCollection {#72 …}
      +awards: Doctrine\ORM\PersistentCollection {#61 …}
      +categories: Doctrine\ORM\PersistentCollection {#1820 …}
      -id: 85
      +apId: "privacy@lemmy.ml"
      +apProfileId: "https://lemmy.ml/c/privacy"
      +apPublicUrl: "https://lemmy.ml/c/privacy"
      +apFollowersUrl: "https://lemmy.ml/c/privacy/followers"
      +apInboxUrl: "https://lemmy.ml/inbox"
      +apDomain: "lemmy.ml"
      +apPreferredUsername: "privacy"
      +apDiscoverable: true
      +apManuallyApprovesFollowers: null
      +privateKey: null
      +publicKey: null
      +apFetchedAt: DateTime @1729583592 {#270
        date: 2024-10-22 09:53:12.0 +02:00
      }
      +apDeletedAt: null
      +apTimeoutAt: null
      +visibility: "visible             "
      +createdAt: DateTimeImmutable @1699780272 {#272
        date: 2023-11-12 10:11:12.0 +01:00
      }
    }
    +image: Proxies\__CG__\App\Entity\Image {#1943 …}
    +domain: Proxies\__CG__\App\Entity\Domain {#1912 …}
    +slug: "Google-Researchers-Attack-Prompts-ChatGPT-to-Reveal-Its-Training-Data"
    +title: "Google Researchers’ Attack Prompts ChatGPT to Reveal Its Training Data"
    +url: "https://www.404media.co/google-researchers-attack-convinces-chatgpt-to-reveal-its-training-data/"
    +body: """
      > ChatGPT is full of sensitive private information and spits out verbatim text from CNN, Goodreads, WordPress blogs, fandom wikis, Terms of Service agreements, Stack Overflow source code, Wikipedia pages, news blogs, random internet comments, and much more.\n
      \n
      > Using this tactic, the researchers showed that there are large amounts of privately identifiable information (PII) in OpenAI’s large language models. They also showed that, on a public version of ChatGPT, the chatbot spit out large passages of text scraped verbatim from other places on the internet.\n
      \n
      > “In total, 16.9 percent of generations we tested contained memorized PII,” they wrote, which included “identifying phone and fax numbers, email and physical addresses … social media handles, URLs, and names and birthdays.”\n
      \n
      Edit: [The full paper that's referenced in the article can be found here](https://arxiv.org/pdf/2311.17035.pdf)
      """
    +type: "link"
    +lang: "en"
    +isOc: false
    +hasEmbed: false
    +commentCount: 80
    +favouriteCount: 364
    +score: 0
    +isAdult: false
    +sticky: false
    +lastActive: DateTime @1701712043 {#2410
      date: 2023-12-04 18:47:23.0 +01:00
    }
    +ip: null
    +adaAmount: 0
    +tags: null
    +mentions: null
    +comments: Doctrine\ORM\PersistentCollection {#1866 …}
    +votes: Doctrine\ORM\PersistentCollection {#1966 …}
    +reports: Doctrine\ORM\PersistentCollection {#1946 …}
    +favourites: Doctrine\ORM\PersistentCollection {#1368 …}
    +notifications: Doctrine\ORM\PersistentCollection {#2426 …}
    +badges: Doctrine\ORM\PersistentCollection {#2439 …}
    +children: []
    -id: 18235
    -titleTs: "'attack':3 'chatgpt':5 'data':10 'googl':1 'prompt':4 'research':2 'reveal':7 'train':9"
    -bodyTs: "'/pdf/2311.17035.pdf)':134 '16.9':88 'address':109 'agreement':23 'also':61 'amount':48 'articl':127 'arxiv.org':133 'arxiv.org/pdf/2311.17035.pdf)':132 'birthday':117 'blog':17,31 'chatbot':71 'chatgpt':1,69 'cnn':14 'code':27 'comment':34 'contain':94 'edit':118 'email':106 'fandom':18 'fax':104 'found':130 'full':3,120 'generat':91 'goodread':15 'handl':112 'identifi':51,101 'includ':100 'inform':7,52 'internet':33,85 'languag':58 'larg':47,57,74 'media':111 'memor':95 'model':59 'much':36 'name':115 'news':30 'number':105 'openai':55 'overflow':25 'page':29 'paper':121 'passag':75 'percent':89 'phone':102 'physic':108 'pii':53,96 'place':82 'privat':6,50 'public':66 'random':32 'referenc':124 'research':42 'scrape':78 'sensit':5 'servic':22 'show':43,62 'social':110 'sourc':26 'spit':9,72 'stack':24 'tactic':40 'term':20 'test':93 'text':12,77 'total':87 'url':113 'use':38 'verbatim':11,79 'version':67 'wiki':19 'wikipedia':28 'wordpress':16 'wrote':98"
    +cross: false
    +upVotes: 0
    +downVotes: 0
    +ranking: 1701362787
    +visibility: "visible             "
    +apId: "https://kbin.social/m/privacy@lemmy.ml/t/665071"
    +editedAt: DateTimeImmutable @1702265006 {#1850
      date: 2023-12-11 04:23:26.0 +01:00
    }
    +createdAt: DateTimeImmutable @1701276387 {#2420
      date: 2023-11-29 17:46:27.0 +01:00
    }
  }
  +magazine: App\Entity\Magazine {#266}
  +image: null
  +parent: null
  +root: null
  +body: """
    AI really did that thing where you repeat a word so often that it loses meaning and the rest of the world eventually starts to turn to mush.\n
    \n
    Jokes aside, I think I know why it does this: Because by giving it a STUPIDLY easy prompt it can rack up huge amounts of reward function, once you accumulate enough it no longer becomes bound by it and it will simply act in whatever the easiest action to continue gaining points is: in this case, it’s reading its training data rather than doing the usual “machine learning” obfuscating that it normally does. Maybe this is a result of repeating a word over and over giving an exponentially rising score until it eventually hits +INF, effectively disabling it? Seems a little contrived but it’s an avenue worth investigating.
    """
  +lang: "en"
  +isAdult: false
  +favouriteCount: 1
  +score: 0
  +lastActive: DateTime @1702486394 {#4220
    date: 2023-12-13 17:53:14.0 +01:00
  }
  +ip: null
  +tags: null
  +mentions: [
    "@btp@kbin.social"
  ]
  +children: Doctrine\ORM\PersistentCollection {#4227 …}
  +nested: Doctrine\ORM\PersistentCollection {#4229 …}
  +votes: Doctrine\ORM\PersistentCollection {#4231 …}
  +reports: Doctrine\ORM\PersistentCollection {#4233 …}
  +favourites: Doctrine\ORM\PersistentCollection {#4235 …}
  +notifications: Doctrine\ORM\PersistentCollection {#4237 …}
  -id: 183826
  -bodyTs: "'accumul':58 'act':71 'action':76 'ai':1 'amount':52 'asid':30 'avenu':136 'becom':63 'bound':64 'case':84 'continu':78 'contriv':131 'data':90 'disabl':126 'easi':45 'easiest':75 'effect':125 'enough':59 'eventu':23,122 'exponenti':117 'function':55 'gain':79 'give':41,115 'hit':123 'huge':51 'inf':124 'investig':138 'joke':29 'know':34 'learn':97 'littl':130 'longer':62 'lose':15 'machin':96 'mayb':103 'mean':16 'mush':28 'normal':101 'obfusc':98 'often':12 'point':80 'prompt':46 'rack':49 'rather':91 'read':87 'realli':2 'repeat':8,109 'rest':19 'result':107 'reward':54 'rise':118 'score':119 'seem':128 'simpli':70 'start':24 'stupid':44 'thing':5 'think':32 'train':89 'turn':26 'usual':95 'whatev':73 'word':10,111 'world':22 'worth':137"
  +ranking: 0
  +commentCount: 0
  +upVotes: 0
  +downVotes: 0
  +visibility: "visible             "
  +apId: "https://lemmy.ml/comment/6257022"
  +editedAt: DateTimeImmutable @1702480541 {#4221
    date: 2023-12-13 16:15:41.0 +01:00
  }
  +createdAt: DateTimeImmutable @1701359642 {#4222
    date: 2023-11-30 16:54:02.0 +01:00
  }
}
"App\Security\Voter\UserVoter"
null
"App\Security\Voter\UserVoter"
App\Entity\EntryComment {#4269
  +user: Proxies\__CG__\App\Entity\User {#4270 …}
  +entry: App\Entity\Entry {#2412
    +user: Proxies\__CG__\App\Entity\User {#1970 …}
    +magazine: App\Entity\Magazine {#266
      +icon: Proxies\__CG__\App\Entity\Image {#247 …}
      +name: "privacy@lemmy.ml"
      +title: "privacy"
      +description: """
        A place to discuss privacy and freedom in the digital world.\n
        ============================================================\n
        \n
        Privacy has become a very important issue in modern society, with companies and governments constantly abusing their power, more and more people are waking up to the importance of digital privacy.\n
        \n
        In this community everyone is welcome to post links and discuss topics related to privacy.\n
        \n
        ### Some Rules\n
        \n
        - Posting a link to a website containing tracking isn’t great, if contents of the website are behind a paywall maybe copy them into the post\n
        - Don’t promote proprietary software\n
        - Try to keep things on topic\n
        - If you have a question, please try searching for previous discussions, maybe it has already been answered\n
        - Reposts are fine, but should have at least a couple of weeks in between so that the post can reach a new audience\n
        - Be nice :)\n
        \n
        ### Related communities\n
        \n
        - [Lemmy.ml libre_culture](https://lemmy.ml/c/libre_culture)\n
        - [Lemmy.ml privatelife](https://lemmy.ml/c/privatelife)\n
        - [Lemmy.ml DeGoogle](https://lemmy.ml/c/degoogle)\n
        - [Lemmy.ca privacy](https://lemmy.ca/c/privacy)\n
        \n
        ### Chat rooms\n
        \n
        - [Matrix/Element]Dead\n
        - [Discord](https://discord.gg/8xfrVjvngh)\n
        \n
        much thanks to @gary_host_laptop for the logo design :)
        """
      +rules: null
      +subscriptionsCount: 1
      +entryCount: 561
      +entryCommentCount: 7684
      +postCount: 6
      +postCommentCount: 73
      +isAdult: false
      +customCss: null
      +lastActive: DateTime @1729582580 {#276
        date: 2024-10-22 09:36:20.0 +02:00
      }
      +markedForDeletionAt: null
      +tags: null
      +moderators: Doctrine\ORM\PersistentCollection {#238 …}
      +ownershipRequests: Doctrine\ORM\PersistentCollection {#234 …}
      +moderatorRequests: Doctrine\ORM\PersistentCollection {#223 …}
      +entries: Doctrine\ORM\PersistentCollection {#181 …}
      +posts: Doctrine\ORM\PersistentCollection {#139 …}
      +subscriptions: Doctrine\ORM\PersistentCollection {#201 …}
      +bans: Doctrine\ORM\PersistentCollection {#118 …}
      +reports: Doctrine\ORM\PersistentCollection {#104 …}
      +badges: Doctrine\ORM\PersistentCollection {#82 …}
      +logs: Doctrine\ORM\PersistentCollection {#72 …}
      +awards: Doctrine\ORM\PersistentCollection {#61 …}
      +categories: Doctrine\ORM\PersistentCollection {#1820 …}
      -id: 85
      +apId: "privacy@lemmy.ml"
      +apProfileId: "https://lemmy.ml/c/privacy"
      +apPublicUrl: "https://lemmy.ml/c/privacy"
      +apFollowersUrl: "https://lemmy.ml/c/privacy/followers"
      +apInboxUrl: "https://lemmy.ml/inbox"
      +apDomain: "lemmy.ml"
      +apPreferredUsername: "privacy"
      +apDiscoverable: true
      +apManuallyApprovesFollowers: null
      +privateKey: null
      +publicKey: null
      +apFetchedAt: DateTime @1729583592 {#270
        date: 2024-10-22 09:53:12.0 +02:00
      }
      +apDeletedAt: null
      +apTimeoutAt: null
      +visibility: "visible             "
      +createdAt: DateTimeImmutable @1699780272 {#272
        date: 2023-11-12 10:11:12.0 +01:00
      }
    }
    +image: Proxies\__CG__\App\Entity\Image {#1943 …}
    +domain: Proxies\__CG__\App\Entity\Domain {#1912 …}
    +slug: "Google-Researchers-Attack-Prompts-ChatGPT-to-Reveal-Its-Training-Data"
    +title: "Google Researchers’ Attack Prompts ChatGPT to Reveal Its Training Data"
    +url: "https://www.404media.co/google-researchers-attack-convinces-chatgpt-to-reveal-its-training-data/"
    +body: """
      > ChatGPT is full of sensitive private information and spits out verbatim text from CNN, Goodreads, WordPress blogs, fandom wikis, Terms of Service agreements, Stack Overflow source code, Wikipedia pages, news blogs, random internet comments, and much more.\n
      \n
      > Using this tactic, the researchers showed that there are large amounts of privately identifiable information (PII) in OpenAI’s large language models. They also showed that, on a public version of ChatGPT, the chatbot spit out large passages of text scraped verbatim from other places on the internet.\n
      \n
      > “In total, 16.9 percent of generations we tested contained memorized PII,” they wrote, which included “identifying phone and fax numbers, email and physical addresses … social media handles, URLs, and names and birthdays.”\n
      \n
      Edit: [The full paper that's referenced in the article can be found here](https://arxiv.org/pdf/2311.17035.pdf)
      """
    +type: "link"
    +lang: "en"
    +isOc: false
    +hasEmbed: false
    +commentCount: 80
    +favouriteCount: 364
    +score: 0
    +isAdult: false
    +sticky: false
    +lastActive: DateTime @1701712043 {#2410
      date: 2023-12-04 18:47:23.0 +01:00
    }
    +ip: null
    +adaAmount: 0
    +tags: null
    +mentions: null
    +comments: Doctrine\ORM\PersistentCollection {#1866 …}
    +votes: Doctrine\ORM\PersistentCollection {#1966 …}
    +reports: Doctrine\ORM\PersistentCollection {#1946 …}
    +favourites: Doctrine\ORM\PersistentCollection {#1368 …}
    +notifications: Doctrine\ORM\PersistentCollection {#2426 …}
    +badges: Doctrine\ORM\PersistentCollection {#2439 …}
    +children: []
    -id: 18235
    -titleTs: "'attack':3 'chatgpt':5 'data':10 'googl':1 'prompt':4 'research':2 'reveal':7 'train':9"
    -bodyTs: "'/pdf/2311.17035.pdf)':134 '16.9':88 'address':109 'agreement':23 'also':61 'amount':48 'articl':127 'arxiv.org':133 'arxiv.org/pdf/2311.17035.pdf)':132 'birthday':117 'blog':17,31 'chatbot':71 'chatgpt':1,69 'cnn':14 'code':27 'comment':34 'contain':94 'edit':118 'email':106 'fandom':18 'fax':104 'found':130 'full':3,120 'generat':91 'goodread':15 'handl':112 'identifi':51,101 'includ':100 'inform':7,52 'internet':33,85 'languag':58 'larg':47,57,74 'media':111 'memor':95 'model':59 'much':36 'name':115 'news':30 'number':105 'openai':55 'overflow':25 'page':29 'paper':121 'passag':75 'percent':89 'phone':102 'physic':108 'pii':53,96 'place':82 'privat':6,50 'public':66 'random':32 'referenc':124 'research':42 'scrape':78 'sensit':5 'servic':22 'show':43,62 'social':110 'sourc':26 'spit':9,72 'stack':24 'tactic':40 'term':20 'test':93 'text':12,77 'total':87 'url':113 'use':38 'verbatim':11,79 'version':67 'wiki':19 'wikipedia':28 'wordpress':16 'wrote':98"
    +cross: false
    +upVotes: 0
    +downVotes: 0
    +ranking: 1701362787
    +visibility: "visible             "
    +apId: "https://kbin.social/m/privacy@lemmy.ml/t/665071"
    +editedAt: DateTimeImmutable @1702265006 {#1850
      date: 2023-12-11 04:23:26.0 +01:00
    }
    +createdAt: DateTimeImmutable @1701276387 {#2420
      date: 2023-11-29 17:46:27.0 +01:00
    }
  }
  +magazine: App\Entity\Magazine {#266}
  +image: null
  +parent: App\Entity\EntryComment {#4226
    +user: App\Entity\User {#4165 …}
    +entry: App\Entity\Entry {#2412}
    +magazine: App\Entity\Magazine {#266}
    +image: null
    +parent: null
    +root: null
    +body: """
      AI really did that thing where you repeat a word so often that it loses meaning and the rest of the world eventually starts to turn to mush.\n
      \n
      Jokes aside, I think I know why it does this: Because by giving it a STUPIDLY easy prompt it can rack up huge amounts of reward function, once you accumulate enough it no longer becomes bound by it and it will simply act in whatever the easiest action to continue gaining points is: in this case, it’s reading its training data rather than doing the usual “machine learning” obfuscating that it normally does. Maybe this is a result of repeating a word over and over giving an exponentially rising score until it eventually hits +INF, effectively disabling it? Seems a little contrived but it’s an avenue worth investigating.
      """
    +lang: "en"
    +isAdult: false
    +favouriteCount: 1
    +score: 0
    +lastActive: DateTime @1702486394 {#4220
      date: 2023-12-13 17:53:14.0 +01:00
    }
    +ip: null
    +tags: null
    +mentions: [
      "@btp@kbin.social"
    ]
    +children: Doctrine\ORM\PersistentCollection {#4227 …}
    +nested: Doctrine\ORM\PersistentCollection {#4229 …}
    +votes: Doctrine\ORM\PersistentCollection {#4231 …}
    +reports: Doctrine\ORM\PersistentCollection {#4233 …}
    +favourites: Doctrine\ORM\PersistentCollection {#4235 …}
    +notifications: Doctrine\ORM\PersistentCollection {#4237 …}
    -id: 183826
    -bodyTs: "'accumul':58 'act':71 'action':76 'ai':1 'amount':52 'asid':30 'avenu':136 'becom':63 'bound':64 'case':84 'continu':78 'contriv':131 'data':90 'disabl':126 'easi':45 'easiest':75 'effect':125 'enough':59 'eventu':23,122 'exponenti':117 'function':55 'gain':79 'give':41,115 'hit':123 'huge':51 'inf':124 'investig':138 'joke':29 'know':34 'learn':97 'littl':130 'longer':62 'lose':15 'machin':96 'mayb':103 'mean':16 'mush':28 'normal':101 'obfusc':98 'often':12 'point':80 'prompt':46 'rack':49 'rather':91 'read':87 'realli':2 'repeat':8,109 'rest':19 'result':107 'reward':54 'rise':118 'score':119 'seem':128 'simpli':70 'start':24 'stupid':44 'thing':5 'think':32 'train':89 'turn':26 'usual':95 'whatev':73 'word':10,111 'world':22 'worth':137"
    +ranking: 0
    +commentCount: 0
    +upVotes: 0
    +downVotes: 0
    +visibility: "visible             "
    +apId: "https://lemmy.ml/comment/6257022"
    +editedAt: DateTimeImmutable @1702480541 {#4221
      date: 2023-12-13 16:15:41.0 +01:00
    }
    +createdAt: DateTimeImmutable @1701359642 {#4222
      date: 2023-11-30 16:54:02.0 +01:00
    }
  }
  +root: App\Entity\EntryComment {#4226}
  +body: """
    I watched a video from a guy who used machine learning to play Pokemon and he did a great analysis of the process. The most interesting part to me was how small changes to the reward system could produce such bizarre and unexpected behavior. He gave out rewards for exploring new areas by taking screenshots after every input and then comparing them against every previous one. Suddenly it became very fixated on a specific area of the game and he couldn’t figure out why. Turns out there was both flowers and water animating in that area so it triggered a lot of rewards without actually exploring. The AI literally got distracted looking at the beautiful landscape!\n
    \n
    Anyway, that example helped me understand the challenges of this sort of software design. Super fascinating stuff.
    """
  +lang: "en"
  +isAdult: false
  +favouriteCount: 4
  +score: 0
  +lastActive: DateTime @1701388535 {#4266
    date: 2023-12-01 00:55:35.0 +01:00
  }
  +ip: null
  +tags: null
  +mentions: [
    "@btp@kbin.social"
    "@Omega_Haxors@lemmy.ml"
  ]
  +children: Doctrine\ORM\PersistentCollection {#4271 …}
  +nested: Doctrine\ORM\PersistentCollection {#4273 …}
  +votes: Doctrine\ORM\PersistentCollection {#4275 …}
  +reports: Doctrine\ORM\PersistentCollection {#4277 …}
  +favourites: Doctrine\ORM\PersistentCollection {#4279 …}
  +notifications: Doctrine\ORM\PersistentCollection {#4281 …}
  -id: 184172
  -bodyTs: "'actual':106 'ai':109 'analysi':20 'anim':94 'anyway':118 'area':52,75,97 'beauti':116 'becam':69 'behavior':44 'bizarr':41 'challeng':125 'chang':33 'compar':61 'could':38 'couldn':81 'design':131 'distract':112 'everi':57,64 'exampl':120 'explor':50,107 'fascin':133 'figur':83 'fixat':71 'flower':91 'game':78 'gave':46 'got':111 'great':19 'guy':7 'help':121 'input':58 'interest':26 'landscap':117 'learn':11 'liter':110 'look':113 'lot':102 'machin':10 'new':51 'one':66 'part':27 'play':13 'pokemon':14 'previous':65 'process':23 'produc':39 'reward':36,48,104 'screenshot':55 'small':32 'softwar':130 'sort':128 'specif':74 'stuff':134 'sudden':67 'super':132 'system':37 'take':54 'trigger':100 'turn':86 'understand':123 'unexpect':43 'use':9 'video':4 'watch':2 'water':93 'without':105"
  +ranking: 0
  +commentCount: 0
  +upVotes: 0
  +downVotes: 0
  +visibility: "visible             "
  +apId: "https://corndog.social/comment/1430879"
  +editedAt: DateTimeImmutable @1702486405 {#4267
    date: 2023-12-13 17:53:25.0 +01:00
  }
  +createdAt: DateTimeImmutable @1701388535 {#4268
    date: 2023-12-01 00:55:35.0 +01:00
  }
}
"App\Security\Voter\UserVoter"
App\Entity\EntryComment {#4269
  +user: Proxies\__CG__\App\Entity\User {#4270 …}
  +entry: App\Entity\Entry {#2412
    +user: Proxies\__CG__\App\Entity\User {#1970 …}
    +magazine: App\Entity\Magazine {#266
      +icon: Proxies\__CG__\App\Entity\Image {#247 …}
      +name: "privacy@lemmy.ml"
      +title: "privacy"
      +description: """
        A place to discuss privacy and freedom in the digital world.\n
        ============================================================\n
        \n
        Privacy has become a very important issue in modern society, with companies and governments constantly abusing their power, more and more people are waking up to the importance of digital privacy.\n
        \n
        In this community everyone is welcome to post links and discuss topics related to privacy.\n
        \n
        ### Some Rules\n
        \n
        - Posting a link to a website containing tracking isn’t great, if contents of the website are behind a paywall maybe copy them into the post\n
        - Don’t promote proprietary software\n
        - Try to keep things on topic\n
        - If you have a question, please try searching for previous discussions, maybe it has already been answered\n
        - Reposts are fine, but should have at least a couple of weeks in between so that the post can reach a new audience\n
        - Be nice :)\n
        \n
        ### Related communities\n
        \n
        - [Lemmy.ml libre_culture](https://lemmy.ml/c/libre_culture)\n
        - [Lemmy.ml privatelife](https://lemmy.ml/c/privatelife)\n
        - [Lemmy.ml DeGoogle](https://lemmy.ml/c/degoogle)\n
        - [Lemmy.ca privacy](https://lemmy.ca/c/privacy)\n
        \n
        ### Chat rooms\n
        \n
        - [Matrix/Element]Dead\n
        - [Discord](https://discord.gg/8xfrVjvngh)\n
        \n
        much thanks to @gary_host_laptop for the logo design :)
        """
      +rules: null
      +subscriptionsCount: 1
      +entryCount: 561
      +entryCommentCount: 7684
      +postCount: 6
      +postCommentCount: 73
      +isAdult: false
      +customCss: null
      +lastActive: DateTime @1729582580 {#276
        date: 2024-10-22 09:36:20.0 +02:00
      }
      +markedForDeletionAt: null
      +tags: null
      +moderators: Doctrine\ORM\PersistentCollection {#238 …}
      +ownershipRequests: Doctrine\ORM\PersistentCollection {#234 …}
      +moderatorRequests: Doctrine\ORM\PersistentCollection {#223 …}
      +entries: Doctrine\ORM\PersistentCollection {#181 …}
      +posts: Doctrine\ORM\PersistentCollection {#139 …}
      +subscriptions: Doctrine\ORM\PersistentCollection {#201 …}
      +bans: Doctrine\ORM\PersistentCollection {#118 …}
      +reports: Doctrine\ORM\PersistentCollection {#104 …}
      +badges: Doctrine\ORM\PersistentCollection {#82 …}
      +logs: Doctrine\ORM\PersistentCollection {#72 …}
      +awards: Doctrine\ORM\PersistentCollection {#61 …}
      +categories: Doctrine\ORM\PersistentCollection {#1820 …}
      -id: 85
      +apId: "privacy@lemmy.ml"
      +apProfileId: "https://lemmy.ml/c/privacy"
      +apPublicUrl: "https://lemmy.ml/c/privacy"
      +apFollowersUrl: "https://lemmy.ml/c/privacy/followers"
      +apInboxUrl: "https://lemmy.ml/inbox"
      +apDomain: "lemmy.ml"
      +apPreferredUsername: "privacy"
      +apDiscoverable: true
      +apManuallyApprovesFollowers: null
      +privateKey: null
      +publicKey: null
      +apFetchedAt: DateTime @1729583592 {#270
        date: 2024-10-22 09:53:12.0 +02:00
      }
      +apDeletedAt: null
      +apTimeoutAt: null
      +visibility: "visible             "
      +createdAt: DateTimeImmutable @1699780272 {#272
        date: 2023-11-12 10:11:12.0 +01:00
      }
    }
    +image: Proxies\__CG__\App\Entity\Image {#1943 …}
    +domain: Proxies\__CG__\App\Entity\Domain {#1912 …}
    +slug: "Google-Researchers-Attack-Prompts-ChatGPT-to-Reveal-Its-Training-Data"
    +title: "Google Researchers’ Attack Prompts ChatGPT to Reveal Its Training Data"
    +url: "https://www.404media.co/google-researchers-attack-convinces-chatgpt-to-reveal-its-training-data/"
    +body: """
      > ChatGPT is full of sensitive private information and spits out verbatim text from CNN, Goodreads, WordPress blogs, fandom wikis, Terms of Service agreements, Stack Overflow source code, Wikipedia pages, news blogs, random internet comments, and much more.\n
      \n
      > Using this tactic, the researchers showed that there are large amounts of privately identifiable information (PII) in OpenAI’s large language models. They also showed that, on a public version of ChatGPT, the chatbot spit out large passages of text scraped verbatim from other places on the internet.\n
      \n
      > “In total, 16.9 percent of generations we tested contained memorized PII,” they wrote, which included “identifying phone and fax numbers, email and physical addresses … social media handles, URLs, and names and birthdays.”\n
      \n
      Edit: [The full paper that's referenced in the article can be found here](https://arxiv.org/pdf/2311.17035.pdf)
      """
    +type: "link"
    +lang: "en"
    +isOc: false
    +hasEmbed: false
    +commentCount: 80
    +favouriteCount: 364
    +score: 0
    +isAdult: false
    +sticky: false
    +lastActive: DateTime @1701712043 {#2410
      date: 2023-12-04 18:47:23.0 +01:00
    }
    +ip: null
    +adaAmount: 0
    +tags: null
    +mentions: null
    +comments: Doctrine\ORM\PersistentCollection {#1866 …}
    +votes: Doctrine\ORM\PersistentCollection {#1966 …}
    +reports: Doctrine\ORM\PersistentCollection {#1946 …}
    +favourites: Doctrine\ORM\PersistentCollection {#1368 …}
    +notifications: Doctrine\ORM\PersistentCollection {#2426 …}
    +badges: Doctrine\ORM\PersistentCollection {#2439 …}
    +children: []
    -id: 18235
    -titleTs: "'attack':3 'chatgpt':5 'data':10 'googl':1 'prompt':4 'research':2 'reveal':7 'train':9"
    -bodyTs: "'/pdf/2311.17035.pdf)':134 '16.9':88 'address':109 'agreement':23 'also':61 'amount':48 'articl':127 'arxiv.org':133 'arxiv.org/pdf/2311.17035.pdf)':132 'birthday':117 'blog':17,31 'chatbot':71 'chatgpt':1,69 'cnn':14 'code':27 'comment':34 'contain':94 'edit':118 'email':106 'fandom':18 'fax':104 'found':130 'full':3,120 'generat':91 'goodread':15 'handl':112 'identifi':51,101 'includ':100 'inform':7,52 'internet':33,85 'languag':58 'larg':47,57,74 'media':111 'memor':95 'model':59 'much':36 'name':115 'news':30 'number':105 'openai':55 'overflow':25 'page':29 'paper':121 'passag':75 'percent':89 'phone':102 'physic':108 'pii':53,96 'place':82 'privat':6,50 'public':66 'random':32 'referenc':124 'research':42 'scrape':78 'sensit':5 'servic':22 'show':43,62 'social':110 'sourc':26 'spit':9,72 'stack':24 'tactic':40 'term':20 'test':93 'text':12,77 'total':87 'url':113 'use':38 'verbatim':11,79 'version':67 'wiki':19 'wikipedia':28 'wordpress':16 'wrote':98"
    +cross: false
    +upVotes: 0
    +downVotes: 0
    +ranking: 1701362787
    +visibility: "visible             "
    +apId: "https://kbin.social/m/privacy@lemmy.ml/t/665071"
    +editedAt: DateTimeImmutable @1702265006 {#1850
      date: 2023-12-11 04:23:26.0 +01:00
    }
    +createdAt: DateTimeImmutable @1701276387 {#2420
      date: 2023-11-29 17:46:27.0 +01:00
    }
  }
  +magazine: App\Entity\Magazine {#266}
  +image: null
  +parent: App\Entity\EntryComment {#4226
    +user: App\Entity\User {#4165 …}
    +entry: App\Entity\Entry {#2412}
    +magazine: App\Entity\Magazine {#266}
    +image: null
    +parent: null
    +root: null
    +body: """
      AI really did that thing where you repeat a word so often that it loses meaning and the rest of the world eventually starts to turn to mush.\n
      \n
      Jokes aside, I think I know why it does this: Because by giving it a STUPIDLY easy prompt it can rack up huge amounts of reward function, once you accumulate enough it no longer becomes bound by it and it will simply act in whatever the easiest action to continue gaining points is: in this case, it’s reading its training data rather than doing the usual “machine learning” obfuscating that it normally does. Maybe this is a result of repeating a word over and over giving an exponentially rising score until it eventually hits +INF, effectively disabling it? Seems a little contrived but it’s an avenue worth investigating.
      """
    +lang: "en"
    +isAdult: false
    +favouriteCount: 1
    +score: 0
    +lastActive: DateTime @1702486394 {#4220
      date: 2023-12-13 17:53:14.0 +01:00
    }
    +ip: null
    +tags: null
    +mentions: [
      "@btp@kbin.social"
    ]
    +children: Doctrine\ORM\PersistentCollection {#4227 …}
    +nested: Doctrine\ORM\PersistentCollection {#4229 …}
    +votes: Doctrine\ORM\PersistentCollection {#4231 …}
    +reports: Doctrine\ORM\PersistentCollection {#4233 …}
    +favourites: Doctrine\ORM\PersistentCollection {#4235 …}
    +notifications: Doctrine\ORM\PersistentCollection {#4237 …}
    -id: 183826
    -bodyTs: "'accumul':58 'act':71 'action':76 'ai':1 'amount':52 'asid':30 'avenu':136 'becom':63 'bound':64 'case':84 'continu':78 'contriv':131 'data':90 'disabl':126 'easi':45 'easiest':75 'effect':125 'enough':59 'eventu':23,122 'exponenti':117 'function':55 'gain':79 'give':41,115 'hit':123 'huge':51 'inf':124 'investig':138 'joke':29 'know':34 'learn':97 'littl':130 'longer':62 'lose':15 'machin':96 'mayb':103 'mean':16 'mush':28 'normal':101 'obfusc':98 'often':12 'point':80 'prompt':46 'rack':49 'rather':91 'read':87 'realli':2 'repeat':8,109 'rest':19 'result':107 'reward':54 'rise':118 'score':119 'seem':128 'simpli':70 'start':24 'stupid':44 'thing':5 'think':32 'train':89 'turn':26 'usual':95 'whatev':73 'word':10,111 'world':22 'worth':137"
    +ranking: 0
    +commentCount: 0
    +upVotes: 0
    +downVotes: 0
    +visibility: "visible             "
    +apId: "https://lemmy.ml/comment/6257022"
    +editedAt: DateTimeImmutable @1702480541 {#4221
      date: 2023-12-13 16:15:41.0 +01:00
    }
    +createdAt: DateTimeImmutable @1701359642 {#4222
      date: 2023-11-30 16:54:02.0 +01:00
    }
  }
  +root: App\Entity\EntryComment {#4226}
  +body: """
    I watched a video from a guy who used machine learning to play Pokemon and he did a great analysis of the process. The most interesting part to me was how small changes to the reward system could produce such bizarre and unexpected behavior. He gave out rewards for exploring new areas by taking screenshots after every input and then comparing them against every previous one. Suddenly it became very fixated on a specific area of the game and he couldn’t figure out why. Turns out there was both flowers and water animating in that area so it triggered a lot of rewards without actually exploring. The AI literally got distracted looking at the beautiful landscape!\n
    \n
    Anyway, that example helped me understand the challenges of this sort of software design. Super fascinating stuff.
    """
  +lang: "en"
  +isAdult: false
  +favouriteCount: 4
  +score: 0
  +lastActive: DateTime @1701388535 {#4266
    date: 2023-12-01 00:55:35.0 +01:00
  }
  +ip: null
  +tags: null
  +mentions: [
    "@btp@kbin.social"
    "@Omega_Haxors@lemmy.ml"
  ]
  +children: Doctrine\ORM\PersistentCollection {#4271 …}
  +nested: Doctrine\ORM\PersistentCollection {#4273 …}
  +votes: Doctrine\ORM\PersistentCollection {#4275 …}
  +reports: Doctrine\ORM\PersistentCollection {#4277 …}
  +favourites: Doctrine\ORM\PersistentCollection {#4279 …}
  +notifications: Doctrine\ORM\PersistentCollection {#4281 …}
  -id: 184172
  -bodyTs: "'actual':106 'ai':109 'analysi':20 'anim':94 'anyway':118 'area':52,75,97 'beauti':116 'becam':69 'behavior':44 'bizarr':41 'challeng':125 'chang':33 'compar':61 'could':38 'couldn':81 'design':131 'distract':112 'everi':57,64 'exampl':120 'explor':50,107 'fascin':133 'figur':83 'fixat':71 'flower':91 'game':78 'gave':46 'got':111 'great':19 'guy':7 'help':121 'input':58 'interest':26 'landscap':117 'learn':11 'liter':110 'look':113 'lot':102 'machin':10 'new':51 'one':66 'part':27 'play':13 'pokemon':14 'previous':65 'process':23 'produc':39 'reward':36,48,104 'screenshot':55 'small':32 'softwar':130 'sort':128 'specif':74 'stuff':134 'sudden':67 'super':132 'system':37 'take':54 'trigger':100 'turn':86 'understand':123 'unexpect':43 'use':9 'video':4 'watch':2 'water':93 'without':105"
  +ranking: 0
  +commentCount: 0
  +upVotes: 0
  +downVotes: 0
  +visibility: "visible             "
  +apId: "https://corndog.social/comment/1430879"
  +editedAt: DateTimeImmutable @1702486405 {#4267
    date: 2023-12-13 17:53:25.0 +01:00
  }
  +createdAt: DateTimeImmutable @1701388535 {#4268
    date: 2023-12-01 00:55:35.0 +01:00
  }
}
"App\Security\Voter\UserVoter"
App\Entity\EntryComment {#4269
  +user: Proxies\__CG__\App\Entity\User {#4270 …}
  +entry: App\Entity\Entry {#2412
    +user: Proxies\__CG__\App\Entity\User {#1970 …}
    +magazine: App\Entity\Magazine {#266
      +icon: Proxies\__CG__\App\Entity\Image {#247 …}
      +name: "privacy@lemmy.ml"
      +title: "privacy"
      +description: """
        A place to discuss privacy and freedom in the digital world.\n
        ============================================================\n
        \n
        Privacy has become a very important issue in modern society, with companies and governments constantly abusing their power, more and more people are waking up to the importance of digital privacy.\n
        \n
        In this community everyone is welcome to post links and discuss topics related to privacy.\n
        \n
        ### Some Rules\n
        \n
        - Posting a link to a website containing tracking isn’t great, if contents of the website are behind a paywall maybe copy them into the post\n
        - Don’t promote proprietary software\n
        - Try to keep things on topic\n
        - If you have a question, please try searching for previous discussions, maybe it has already been answered\n
        - Reposts are fine, but should have at least a couple of weeks in between so that the post can reach a new audience\n
        - Be nice :)\n
        \n
        ### Related communities\n
        \n
        - [Lemmy.ml libre_culture](https://lemmy.ml/c/libre_culture)\n
        - [Lemmy.ml privatelife](https://lemmy.ml/c/privatelife)\n
        - [Lemmy.ml DeGoogle](https://lemmy.ml/c/degoogle)\n
        - [Lemmy.ca privacy](https://lemmy.ca/c/privacy)\n
        \n
        ### Chat rooms\n
        \n
        - [Matrix/Element]Dead\n
        - [Discord](https://discord.gg/8xfrVjvngh)\n
        \n
        much thanks to @gary_host_laptop for the logo design :)
        """
      +rules: null
      +subscriptionsCount: 1
      +entryCount: 561
      +entryCommentCount: 7684
      +postCount: 6
      +postCommentCount: 73
      +isAdult: false
      +customCss: null
      +lastActive: DateTime @1729582580 {#276
        date: 2024-10-22 09:36:20.0 +02:00
      }
      +markedForDeletionAt: null
      +tags: null
      +moderators: Doctrine\ORM\PersistentCollection {#238 …}
      +ownershipRequests: Doctrine\ORM\PersistentCollection {#234 …}
      +moderatorRequests: Doctrine\ORM\PersistentCollection {#223 …}
      +entries: Doctrine\ORM\PersistentCollection {#181 …}
      +posts: Doctrine\ORM\PersistentCollection {#139 …}
      +subscriptions: Doctrine\ORM\PersistentCollection {#201 …}
      +bans: Doctrine\ORM\PersistentCollection {#118 …}
      +reports: Doctrine\ORM\PersistentCollection {#104 …}
      +badges: Doctrine\ORM\PersistentCollection {#82 …}
      +logs: Doctrine\ORM\PersistentCollection {#72 …}
      +awards: Doctrine\ORM\PersistentCollection {#61 …}
      +categories: Doctrine\ORM\PersistentCollection {#1820 …}
      -id: 85
      +apId: "privacy@lemmy.ml"
      +apProfileId: "https://lemmy.ml/c/privacy"
      +apPublicUrl: "https://lemmy.ml/c/privacy"
      +apFollowersUrl: "https://lemmy.ml/c/privacy/followers"
      +apInboxUrl: "https://lemmy.ml/inbox"
      +apDomain: "lemmy.ml"
      +apPreferredUsername: "privacy"
      +apDiscoverable: true
      +apManuallyApprovesFollowers: null
      +privateKey: null
      +publicKey: null
      +apFetchedAt: DateTime @1729583592 {#270
        date: 2024-10-22 09:53:12.0 +02:00
      }
      +apDeletedAt: null
      +apTimeoutAt: null
      +visibility: "visible             "
      +createdAt: DateTimeImmutable @1699780272 {#272
        date: 2023-11-12 10:11:12.0 +01:00
      }
    }
    +image: Proxies\__CG__\App\Entity\Image {#1943 …}
    +domain: Proxies\__CG__\App\Entity\Domain {#1912 …}
    +slug: "Google-Researchers-Attack-Prompts-ChatGPT-to-Reveal-Its-Training-Data"
    +title: "Google Researchers’ Attack Prompts ChatGPT to Reveal Its Training Data"
    +url: "https://www.404media.co/google-researchers-attack-convinces-chatgpt-to-reveal-its-training-data/"
    +body: """
      > ChatGPT is full of sensitive private information and spits out verbatim text from CNN, Goodreads, WordPress blogs, fandom wikis, Terms of Service agreements, Stack Overflow source code, Wikipedia pages, news blogs, random internet comments, and much more.\n
      \n
      > Using this tactic, the researchers showed that there are large amounts of privately identifiable information (PII) in OpenAI’s large language models. They also showed that, on a public version of ChatGPT, the chatbot spit out large passages of text scraped verbatim from other places on the internet.\n
      \n
      > “In total, 16.9 percent of generations we tested contained memorized PII,” they wrote, which included “identifying phone and fax numbers, email and physical addresses … social media handles, URLs, and names and birthdays.”\n
      \n
      Edit: [The full paper that's referenced in the article can be found here](https://arxiv.org/pdf/2311.17035.pdf)
      """
    +type: "link"
    +lang: "en"
    +isOc: false
    +hasEmbed: false
    +commentCount: 80
    +favouriteCount: 364
    +score: 0
    +isAdult: false
    +sticky: false
    +lastActive: DateTime @1701712043 {#2410
      date: 2023-12-04 18:47:23.0 +01:00
    }
    +ip: null
    +adaAmount: 0
    +tags: null
    +mentions: null
    +comments: Doctrine\ORM\PersistentCollection {#1866 …}
    +votes: Doctrine\ORM\PersistentCollection {#1966 …}
    +reports: Doctrine\ORM\PersistentCollection {#1946 …}
    +favourites: Doctrine\ORM\PersistentCollection {#1368 …}
    +notifications: Doctrine\ORM\PersistentCollection {#2426 …}
    +badges: Doctrine\ORM\PersistentCollection {#2439 …}
    +children: []
    -id: 18235
    -titleTs: "'attack':3 'chatgpt':5 'data':10 'googl':1 'prompt':4 'research':2 'reveal':7 'train':9"
    -bodyTs: "'/pdf/2311.17035.pdf)':134 '16.9':88 'address':109 'agreement':23 'also':61 'amount':48 'articl':127 'arxiv.org':133 'arxiv.org/pdf/2311.17035.pdf)':132 'birthday':117 'blog':17,31 'chatbot':71 'chatgpt':1,69 'cnn':14 'code':27 'comment':34 'contain':94 'edit':118 'email':106 'fandom':18 'fax':104 'found':130 'full':3,120 'generat':91 'goodread':15 'handl':112 'identifi':51,101 'includ':100 'inform':7,52 'internet':33,85 'languag':58 'larg':47,57,74 'media':111 'memor':95 'model':59 'much':36 'name':115 'news':30 'number':105 'openai':55 'overflow':25 'page':29 'paper':121 'passag':75 'percent':89 'phone':102 'physic':108 'pii':53,96 'place':82 'privat':6,50 'public':66 'random':32 'referenc':124 'research':42 'scrape':78 'sensit':5 'servic':22 'show':43,62 'social':110 'sourc':26 'spit':9,72 'stack':24 'tactic':40 'term':20 'test':93 'text':12,77 'total':87 'url':113 'use':38 'verbatim':11,79 'version':67 'wiki':19 'wikipedia':28 'wordpress':16 'wrote':98"
    +cross: false
    +upVotes: 0
    +downVotes: 0
    +ranking: 1701362787
    +visibility: "visible             "
    +apId: "https://kbin.social/m/privacy@lemmy.ml/t/665071"
    +editedAt: DateTimeImmutable @1702265006 {#1850
      date: 2023-12-11 04:23:26.0 +01:00
    }
    +createdAt: DateTimeImmutable @1701276387 {#2420
      date: 2023-11-29 17:46:27.0 +01:00
    }
  }
  +magazine: App\Entity\Magazine {#266}
  +image: null
  +parent: App\Entity\EntryComment {#4226
    +user: App\Entity\User {#4165 …}
    +entry: App\Entity\Entry {#2412}
    +magazine: App\Entity\Magazine {#266}
    +image: null
    +parent: null
    +root: null
    +body: """
      AI really did that thing where you repeat a word so often that it loses meaning and the rest of the world eventually starts to turn to mush.\n
      \n
      Jokes aside, I think I know why it does this: Because by giving it a STUPIDLY easy prompt it can rack up huge amounts of reward function, once you accumulate enough it no longer becomes bound by it and it will simply act in whatever the easiest action to continue gaining points is: in this case, it’s reading its training data rather than doing the usual “machine learning” obfuscating that it normally does. Maybe this is a result of repeating a word over and over giving an exponentially rising score until it eventually hits +INF, effectively disabling it? Seems a little contrived but it’s an avenue worth investigating.
      """
    +lang: "en"
    +isAdult: false
    +favouriteCount: 1
    +score: 0
    +lastActive: DateTime @1702486394 {#4220
      date: 2023-12-13 17:53:14.0 +01:00
    }
    +ip: null
    +tags: null
    +mentions: [
      "@btp@kbin.social"
    ]
    +children: Doctrine\ORM\PersistentCollection {#4227 …}
    +nested: Doctrine\ORM\PersistentCollection {#4229 …}
    +votes: Doctrine\ORM\PersistentCollection {#4231 …}
    +reports: Doctrine\ORM\PersistentCollection {#4233 …}
    +favourites: Doctrine\ORM\PersistentCollection {#4235 …}
    +notifications: Doctrine\ORM\PersistentCollection {#4237 …}
    -id: 183826
    -bodyTs: "'accumul':58 'act':71 'action':76 'ai':1 'amount':52 'asid':30 'avenu':136 'becom':63 'bound':64 'case':84 'continu':78 'contriv':131 'data':90 'disabl':126 'easi':45 'easiest':75 'effect':125 'enough':59 'eventu':23,122 'exponenti':117 'function':55 'gain':79 'give':41,115 'hit':123 'huge':51 'inf':124 'investig':138 'joke':29 'know':34 'learn':97 'littl':130 'longer':62 'lose':15 'machin':96 'mayb':103 'mean':16 'mush':28 'normal':101 'obfusc':98 'often':12 'point':80 'prompt':46 'rack':49 'rather':91 'read':87 'realli':2 'repeat':8,109 'rest':19 'result':107 'reward':54 'rise':118 'score':119 'seem':128 'simpli':70 'start':24 'stupid':44 'thing':5 'think':32 'train':89 'turn':26 'usual':95 'whatev':73 'word':10,111 'world':22 'worth':137"
    +ranking: 0
    +commentCount: 0
    +upVotes: 0
    +downVotes: 0
    +visibility: "visible             "
    +apId: "https://lemmy.ml/comment/6257022"
    +editedAt: DateTimeImmutable @1702480541 {#4221
      date: 2023-12-13 16:15:41.0 +01:00
    }
    +createdAt: DateTimeImmutable @1701359642 {#4222
      date: 2023-11-30 16:54:02.0 +01:00
    }
  }
  +root: App\Entity\EntryComment {#4226}
  +body: """
    I watched a video from a guy who used machine learning to play Pokemon and he did a great analysis of the process. The most interesting part to me was how small changes to the reward system could produce such bizarre and unexpected behavior. He gave out rewards for exploring new areas by taking screenshots after every input and then comparing them against every previous one. Suddenly it became very fixated on a specific area of the game and he couldn’t figure out why. Turns out there was both flowers and water animating in that area so it triggered a lot of rewards without actually exploring. The AI literally got distracted looking at the beautiful landscape!\n
    \n
    Anyway, that example helped me understand the challenges of this sort of software design. Super fascinating stuff.
    """
  +lang: "en"
  +isAdult: false
  +favouriteCount: 4
  +score: 0
  +lastActive: DateTime @1701388535 {#4266
    date: 2023-12-01 00:55:35.0 +01:00
  }
  +ip: null
  +tags: null
  +mentions: [
    "@btp@kbin.social"
    "@Omega_Haxors@lemmy.ml"
  ]
  +children: Doctrine\ORM\PersistentCollection {#4271 …}
  +nested: Doctrine\ORM\PersistentCollection {#4273 …}
  +votes: Doctrine\ORM\PersistentCollection {#4275 …}
  +reports: Doctrine\ORM\PersistentCollection {#4277 …}
  +favourites: Doctrine\ORM\PersistentCollection {#4279 …}
  +notifications: Doctrine\ORM\PersistentCollection {#4281 …}
  -id: 184172
  -bodyTs: "'actual':106 'ai':109 'analysi':20 'anim':94 'anyway':118 'area':52,75,97 'beauti':116 'becam':69 'behavior':44 'bizarr':41 'challeng':125 'chang':33 'compar':61 'could':38 'couldn':81 'design':131 'distract':112 'everi':57,64 'exampl':120 'explor':50,107 'fascin':133 'figur':83 'fixat':71 'flower':91 'game':78 'gave':46 'got':111 'great':19 'guy':7 'help':121 'input':58 'interest':26 'landscap':117 'learn':11 'liter':110 'look':113 'lot':102 'machin':10 'new':51 'one':66 'part':27 'play':13 'pokemon':14 'previous':65 'process':23 'produc':39 'reward':36,48,104 'screenshot':55 'small':32 'softwar':130 'sort':128 'specif':74 'stuff':134 'sudden':67 'super':132 'system':37 'take':54 'trigger':100 'turn':86 'understand':123 'unexpect':43 'use':9 'video':4 'watch':2 'water':93 'without':105"
  +ranking: 0
  +commentCount: 0
  +upVotes: 0
  +downVotes: 0
  +visibility: "visible             "
  +apId: "https://corndog.social/comment/1430879"
  +editedAt: DateTimeImmutable @1702486405 {#4267
    date: 2023-12-13 17:53:25.0 +01:00
  }
  +createdAt: DateTimeImmutable @1701388535 {#4268
    date: 2023-12-01 00:55:35.0 +01:00
  }
}
"App\Security\Voter\UserVoter"
App\Entity\Magazine {#266
  +icon: Proxies\__CG__\App\Entity\Image {#247 …}
  +name: "privacy@lemmy.ml"
  +title: "privacy"
  +description: """
    A place to discuss privacy and freedom in the digital world.\n
    ============================================================\n
    \n
    Privacy has become a very important issue in modern society, with companies and governments constantly abusing their power, more and more people are waking up to the importance of digital privacy.\n
    \n
    In this community everyone is welcome to post links and discuss topics related to privacy.\n
    \n
    ### Some Rules\n
    \n
    - Posting a link to a website containing tracking isn’t great, if contents of the website are behind a paywall maybe copy them into the post\n
    - Don’t promote proprietary software\n
    - Try to keep things on topic\n
    - If you have a question, please try searching for previous discussions, maybe it has already been answered\n
    - Reposts are fine, but should have at least a couple of weeks in between so that the post can reach a new audience\n
    - Be nice :)\n
    \n
    ### Related communities\n
    \n
    - [Lemmy.ml libre_culture](https://lemmy.ml/c/libre_culture)\n
    - [Lemmy.ml privatelife](https://lemmy.ml/c/privatelife)\n
    - [Lemmy.ml DeGoogle](https://lemmy.ml/c/degoogle)\n
    - [Lemmy.ca privacy](https://lemmy.ca/c/privacy)\n
    \n
    ### Chat rooms\n
    \n
    - [Matrix/Element]Dead\n
    - [Discord](https://discord.gg/8xfrVjvngh)\n
    \n
    much thanks to @gary_host_laptop for the logo design :)
    """
  +rules: null
  +subscriptionsCount: 1
  +entryCount: 561
  +entryCommentCount: 7684
  +postCount: 6
  +postCommentCount: 73
  +isAdult: false
  +customCss: null
  +lastActive: DateTime @1729582580 {#276
    date: 2024-10-22 09:36:20.0 +02:00
  }
  +markedForDeletionAt: null
  +tags: null
  +moderators: Doctrine\ORM\PersistentCollection {#238 …}
  +ownershipRequests: Doctrine\ORM\PersistentCollection {#234 …}
  +moderatorRequests: Doctrine\ORM\PersistentCollection {#223 …}
  +entries: Doctrine\ORM\PersistentCollection {#181 …}
  +posts: Doctrine\ORM\PersistentCollection {#139 …}
  +subscriptions: Doctrine\ORM\PersistentCollection {#201 …}
  +bans: Doctrine\ORM\PersistentCollection {#118 …}
  +reports: Doctrine\ORM\PersistentCollection {#104 …}
  +badges: Doctrine\ORM\PersistentCollection {#82 …}
  +logs: Doctrine\ORM\PersistentCollection {#72 …}
  +awards: Doctrine\ORM\PersistentCollection {#61 …}
  +categories: Doctrine\ORM\PersistentCollection {#1820 …}
  -id: 85
  +apId: "privacy@lemmy.ml"
  +apProfileId: "https://lemmy.ml/c/privacy"
  +apPublicUrl: "https://lemmy.ml/c/privacy"
  +apFollowersUrl: "https://lemmy.ml/c/privacy/followers"
  +apInboxUrl: "https://lemmy.ml/inbox"
  +apDomain: "lemmy.ml"
  +apPreferredUsername: "privacy"
  +apDiscoverable: true
  +apManuallyApprovesFollowers: null
  +privateKey: null
  +publicKey: null
  +apFetchedAt: DateTime @1729583592 {#270
    date: 2024-10-22 09:53:12.0 +02:00
  }
  +apDeletedAt: null
  +apTimeoutAt: null
  +visibility: "visible             "
  +createdAt: DateTimeImmutable @1699780272 {#272
    date: 2023-11-12 10:11:12.0 +01:00
  }
}
"App\Security\Voter\UserVoter"

Symfony Profiler

GET https://kbin.spritesserver.nl/index.php/m/privacy@lemmy.ml/t/18235/Google-Researchers-Attack-Prompts-ChatGPT-to-Reveal-Its-Training-Data/oldest?p=2

Configuration Settings

Theme

Page Width

Security

Token

Firewall

Configuration

Listeners

Authenticators

Access Decision

Access decision log