Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Open source AI is the path forward

about.fb.com

Post reply on HN
Print view View original

761–770 of 936 posts

Re: Open source AI is the path forward

#761

Post by makingstuffs »

baq wrote: post #752
have_faith wrote: post #729

Earlier quoted context omitted.

The feed certainly is, but I suspect most activity left on Facebook is happening in group pages. Groups are the only thing I still log in for as some of them, particularly the local ones, have no other way of taking part. They are also commonly membership by request and actively moderated. If I had the time (and energy) I might put some effort into advocating to moving to something else, but it will be an uphill batt…

I stopped going on facebook a few years ago and don't miss it; I don't even need messenger as everyone migrated to whatsapp (yes I know, normal people don't want to move to signal, but got quite a few techy friends to migrate). The FB-only groups are indeed a problem, I'm delegating them to my wife. IF I ever had to go to FB for anything, I'd probably install a wall-removing browser extension. Mobile app is of course…

> IF I ever had to go to FB for anything, I'd probably install a wall-removing browser extension. Mobile app is of course out of question.

You’ll probably find you can no longer make an account. I’m in the same boat as you (not used and haven’t missed in over a decade), however, my partner needed an account to manage an ad campaign for a client and neither of us were able to make one. Both tried a load of different things and, ultimately, gave up. Had to tell the client what they needed over a video call

Re: Open source AI is the path forward

#762

Post by fishermanbill »

ChanderG wrote: post #730

I think all this discussion around Open-source AI is a total distraction from the elephants in the room. Let's list what you need to run/play around with something like Llama: 1. Software: this is all Pytorch/HF, so completely open-source. This is total parity between what corporates have and what the public has. 2. Model weights: Meta and a few other orgs release open models - as opposed to OpenAI's closed models. S…

But there is an insidiousness to Meta calling their software 'open source'. It feels as if they are riding on the coat tails of the term as if they are being altruistic, when in fact they are being no more altruistic than any large corporation that wants to capture market share via their financial muscle - which I suppose touches on your last point.

Re: Open source AI is the path forward

#763

Post by llm_trw »

TeMPOraL wrote: post #705
llm_trw wrote: post #595

Earlier quoted context omitted.

This is bending the definition to the other extreme. Linux doesn't ship you the compiler you need to build the binaries either, that doesn't mean it's closed source. LLMs are fundamentally different to software and using terms from software just muddies the waters.

And LLMs don't ship with a Python distribution. Linux sources :: dataset that goes into training Linux sources' build confs and scripts :: training code + hyperparameters GCC :: Python + PyTorch or whatever they use in training Compiled Linux kernel binary :: model weights

Just because you keep saying it doesn't make it true.

LLMs are not software any more than photographs are.

Re: Open source AI is the path forward

#764

Post by sbeaks »

sensanaty wrote: post #718

Meanwhile Facebook is flooded with AI-generated slop with hundreds of thousands of other bots interacting with it to boost it to whoever is insane enough to still use that putrid hellhole of a mass-data-harvesting platform. Dead internet theory is very much happening in real time, and I dread what's about to come since the world has collectively decided to lose their minds with this AI crap. And people on this site a…

Are we going to end up with rererecapture where to post you need something on your phone/laptop measuring typing speed and scanning your face with an IR cam? Or a checkmark showing typed out by hand? Wouldn't get rid of ai content but may slow down bots posting it.

Re: Open source AI is the path forward

#765

Post by pera »

I wish Meta stopped using the "open source" misnomer for free of charge weights. In the US the FTC already uses the term Open-Weights, and it seems the industry is also adopting this term (e.g. Mistral).

Someone can correct me here but AFAIK we don't even know which datasets are used to train these models, so why should we even use "open" to describe Llama? This is more similar to a freeware than an open-source project.

[1] https://www.ftc.gov/policy/advocacy-research/tech-at-ftc/202...

Re: Open source AI is the path forward

#766

Post by GreenWatermelon »

nomel wrote: post #520
nothrowaways wrote: post #474

Earlier quoted context omitted.

Weight is the new code.

I think saying it's the new binary is closer to the truth. You can't reproduce it, but you can use it. In this new version, you can even nudge it a bit to do something a little different. New stuff, so probably not good to force old words, with known meanings, onto new stuff.

The model is more akin to a python script than a compiled C binary. This is how I see it:

Training Code and dataset are analogous to the developer who wrote the script

Model and weights are end product that is then released

Inference Code is the runtime that could execute the code. That would be e.g. PyTorch, which can import the weights and run inference.

Re: Open source AI is the path forward

#767

Post by Zambyte »

root_axis wrote: post #560
Zambyte wrote: post #238

Earlier quoted context omitted.

> If so, then how can current ML models be open source? The source of a language model is the text it was trained on. Llama models are not open source (contrary to their claims), they are open weight.

No. The text is an asset used by the source to train the model. The source can process arbitrary text. Text is just text, it was written for communication purposes, software (defined by source code) processes that text in a particular way to train a model.

In programming, "source" and "asset" have specific meanings that conflict with how you used them.

Source is the input to some built artifact. It is the source of that artifact. As in: where the artifact comes from. Textual input is absolutely the source of the ML model. What you are using "source" as is analogous to the source of the compiler in traditional programming.

Asset is an artifact used as input, that is revered verbatim by the output. For example, a logo baked into an application to be rendered in the UI. The compilation of the program doesn't make a new logo, it just moves the asset into the built artifact.

Re: Open source AI is the path forward

#768

Post by narrator »

msnkarthik wrote: post #680

Interesting discussion! While I agree with Zuckerberg's vision, the comments raise valid concerns. The point about GPU accessibility and cost is crucial. Public clusters are great, but sustainable funding and equitable access are essential to avoid exacerbating existing inequalities. I also resonate with the call for CUDA alternatives. Breaking the dependence on proprietary technology is key for a truly open AI ecosy…

3nm chip fabs take years to build. You don't just go to AWS and spin one up. This is the very hard part about AI that breaks a lot of the usual tech assumptions. We have entered a world where suddenly there isn't enough compute, because it's just too damn hard to build capacity and that's different from the past 40 years.

Re: Open source AI is the path forward

#769

Post by benrutter »

pera wrote: post #765

I wish Meta stopped using the "open source" misnomer for free of charge weights. In the US the FTC already uses the term Open-Weights , and it seems the industry is also adopting this term (e.g. Mistral). Someone can correct me here but AFAIK we don't even know which datasets are used to train these models, so why should we even use "open" to describe Llama? This is more similar to a freeware than an open-source proj…

This is such a good point. The industry is really putting the term "open source" through the ringer at the moment but I don't see any justification for considering the final weight output a "source" anymore than releasing a compiled binary would be open source.

In fairness to Llama, the source code itself (though not the training data) is available to access, although not really under a license that many would consider open source.

Re: Open source AI is the path forward

#770

Post by Shinchy »

LauraMedia wrote: post #742
LoveMortuus wrote: post #741

Earlier quoted context omitted.

In regards to the dead internet hypothesis, the content that you're enjoying today, will still be there tomorrow. What I mean is if you, for example, like Mr.Beast, AI is not going to replace him and the content that he produces. Now, he might use AI to boost the productivity of his company, but the end result is still "making the best video ever" as he's often said.

The big problem with this is that content is harder and harder to find. Try to find a non-AI generated reply to a viral post on Twitter, you're looking at having to scroll down 5-6 1080p screens to finally get to some actual stuff people wrote. The content you're enjoying today still exists, but it's a needle in a haystack of AI spam

This is the exact thing I keep telling people. It's all well and good saying human made content will still be around, but it will be covered in a tidal wave of cheaply generated AI hogwash.
Post reply on HN