Outside of unpredictable things like hacks which are bound to happen, always the potential of something with widespread impact, 2024 will be a year of increasing “AI” venture capital investment and some widely used online services are going to pivot or completely rebrand.
Cross modality is what is missing. We have models that can produce text, hear things, and see things really really well. Facilitating communication between these individual models, and probably creating some executive model to utilize them, is what’s missing. We’re probably still a few years from beginning to touch general intelligence
It probably won’t happen until we move to new hardware architectures.
I do think LLMs are a great springboard for AGI, but I don’t think the current hardware allows for models to cross the hump to AGI.
There’s not enough recursive self-interaction in the network to encode nonlinear representations. So we saw this past year a number of impressive papers exhibiting linear representations of world models extrapolated from training data, but there hasn’t yet been any nonlinear representations discovered and I don’t think there will be.
But when we switch to either optoelectronics or colocating processing with memory at a node basis, that next generation of algorithms taking advantage of the hardware may allow for the final missing piece of modern LLMs in extrapolating data from the training set, pulling nonlinear representations of world models from the data (things like saying “I don’t know” will be more prominent in that next generation of models).
From there, we’ll quickly get to AGI, but until then I’m skeptical that classical/traditional hardware will get us there.
I just want good, cheap, and mass production ready solid state batteries of any sort. Right now, anything that’s on the market either isn’t good or isn’t cheap, and none of it is mass produced… Often all three.
If we get over the hurdle of something we can mass produce for cheap that’s as good as, or better than the existing lithium tech that we have, I’m in.
I like this one. I’ve been hoping for some host-your -own AI models that I can dump into a system with a bucket of TPUs and a decent GPU for processing and get my own version of something like chat GPT at home then train it on the entire collective works of documentation and help articles about the software I usually do support for so it can act as a defacto repository of “natural language” chat/search for troubleshooting.
He also made the Oculus lineup so cheap by subsidizing the costs with selling your data, that many VR startups couldn’t compete. Now there’s only a handful of groups still making any VR gear. Immediately after that he killed all PC based VR, though you can still do it with an add-on cable or wirelessly (which sucks on most WiFi), as an afterthought.
This just locked everyone into the Oculus ecosystem and Facebook by extension, bricking more than a few headsets in the process. Now you either have to pay thousands to boutique VR outfits, or buy an Oculus and sell your soul to Zuckerberg for a cut rate product.
I hold Mark solely responsible for killing VR as a consumer product.
I wouldn’t be surprised if VR booms soon in a hyper-competitive environment like phones pre-2016. We already had a boom, but there was a tiny market for decent VR. Now that Meta, Apple, and Samsung are making decent headsets at different price points, it’s only a matter of time before Huawei, Xiaomi, and Oppo start doing crazy stuff that sells well. Hopefully, Valve and HTC become even bigger players as well.
I’m really hoping my email service gains more traction. People are so accustomed to crappy email providers that the current providers don’t really innovate anymore. I think it will actually have a chance of catching on, because it’s such a different way of doing email.
Right now I’m self funding it. I’m building out the enterprise features, like custom domains and domain user management, so I can start marketing to businesses. I’d like to be able to fund it to profitability without any outside investors.
It’s actually surprisingly affordable to run an email service. Right now my biggest expense is the MySQL server. With one server node plus one backup node I should be able to handle several hundred active users.
Release of Windows 12, possibly backtracks Windows 11 decision of requiring TPM.
I hope so, I built my own PC less than 4 years ago and it can’t run windows 11. I don’t care that much at the moment because I’m not a fan of some of the UI choices (and I only use Windows for gaming anyways) but once support is dropped for Windows 10 I’ll need options.
Oh of course, Linux is my everyday machine (I have 2 separate hard drives in my tower). I just haven’t taken the time to figure out Steam yet, and there are some pieces of work software that either work like shit on Linux or aren’t available at all (yes, I know Wine is a thing but it’s not perfect)
Work is what keeps me away from Linux. That and games. I know things have gotten a lot better for gaming on Linux and that’s great, it’s still limiting, quite bluntly. The games I want to play are not all available on Linux, and some have been more or less abandoned, so they may never work on Linux properly.
Work is the worst offender, since a lot of “business productivity” software seems to require Windows, since nobody in business runs anything other than Windows… Except if you’re running web services, then it’s usually Linux… Almost every app has a “cloud” component now that relies on something running Linux… But you can’t get the client software for Linux because fffffffuuuuuuuuuuu
Ugh, I feel you on the work thing. We use the Microsoft suite and although technically there are online versions of the software, it’s fucking terrible compared to the desktop version, especially Teams (and sometimes I just flat out can’t get Teams to work in the browser since it doesn’t play nice with Firefox). And no, I can’t just use Libre Office because it will fuck up any previous formatting of word docs, or in the case of Excel there will be functions that aren’t supported.
I’ve just accepted that I need to be my own IT support for anything Linux in most day to day applications. Calling or emailing customer service inevitably gets me the answer that Linux is not supported.
Yep, then all the specialty application that are made, especially for peripherals like scanners… Forget about it.
You might be able to get it to function at a basic level, but all the settings and customizable features are not going to exist, and you will also be up a creek workout a paddle if you need support, as you’ve correctly noted.
Linux is a wonderful operating system, and it does what it does very well. The fact is, all the business desktop application software companies stick to Windows because that’s what most people have, and most stay with Windows because business app developers don’t support anything else. The only time I’ve known of any users who had something different, it was almost always a Mac, and they always had parallels or some similar windows virtualization software installed because even their mac isn’t supported.
It still caused issues, but it mostly worked at least.
Add comment