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

OpenAI researchers warned board of AI breakthrough ahead of CEO ouster

reuters.com

Post reply on HN
Print view View original

611–620 of 1001 posts

Re: OpenAI researchers warned board of AI breakthrough ahead of CEO ouster

#611

Post by wraptile »

TerrifiedMouse wrote: post #496
dizzydes wrote: post #253

This matches far better with the board's letter re: firing Sam than a simple power struggle or disagreement on commercialisation. Seeing a huge breakthrough and then not reporting it to the board, who then find out via staff letter certainly counts as a "lack of candour".... As an aside, assuming a doomsday scenario, how long can secrets like this stay outside of the hands of bad actors? On a scale of 1 to enriched u…

To quote Reddit user jstadig, > The thing that most worries me about technology is not the technology itself but the greed of those who run it. Someone slimy with limitless ambition like Altman seems to be the worst person to be in charge of things like this.

Why do you perceive Altman as "slimy with limitless ambition"? I've always perceived him as being quite humble from his interviews and podcast appearances.

Re: OpenAI researchers warned board of AI breakthrough ahead of CEO ouster

#612

Post by richardw »

wbhart wrote: post #601
richardw wrote: post #556

Earlier quoted context omitted.

On backtracking, I thought tree-of-thought enabled that? "considering multiple different reasoning paths and self-evaluating choices to decide the next course of action, as well as looking ahead or backtracking when necessary to make global choices" https://arxiv.org/abs/2305.10601 Generally with you though, this thing is not leading to real smarts and that's accepted by many. Yes, it'll fill in a few gaps with expon…

Yes, there are various approaches like tree-of-thought. They don't fundamentally solve the problem because there are just too many paths to explore and inference is just too slow and too expensive to explore 10,000 or 100,000 paths just for basic problems that no one wanted to solve anyway. The problem with solving such problems with LLMs is that if the solution to the problem is unlike problems seen in training, the…

Thank you for the thoughtful response

Re: OpenAI researchers warned board of AI breakthrough ahead of CEO ouster

#613

Post by dwaltrip »

wbhart wrote: post #565
neilk wrote: post #532

Earlier quoted context omitted.

I am neither a mathematician or LLM creator but I do know how to evaluate interesting tech claims. The absolute best case scenario for a new technology is that it when it seems like a toy for nerds, and doesn't outperform anything we have today, but the scaling path is clear. Its problems just won't matter if it does that one thing with scaling. The web is a pretty good hypermedia platform, but a disastrously bad pla…

How on earth could you evaluate the scaling path with too little information. That's my point. You can't possibly know that a technology can solve a given kind of problem if it can only so far solve a completely different kind of problem which is largely unrelated! Saying that performance on grade-school problems is predictive of performance on complex reasoning tasks (including theorem proving) is like saying that a…

For the current generative AI wave, this is how I understand it:

1. The scaling path is decreased val/test loss during training.

2. We have seen multiples times that large decreases in this loss have resulted in very impressive improvements in model capability across a diverse set of tasks (e.g. gpt-1 through gpt-4, and many other examples).

3. By now, there is tons of robust data demonstrating really nice relationships between model size, quantity of data, length of training, quality of data, etc and decreased loss. Evidence keeps building that most multi-billion param LLMs are probably undertrained, perhaps significantly so.

4. Ergo, we should expect continued capability improvement with continued scaling. Make a bigger model, get more data, get higher data quality, and/or train for longer and we will see improved capabilities. The graphs demand that it is so.

---

This is the fundamental scaling hypothesis that labs like OpenAI and Anthropic have been operating off of for the past 5+ years. They looked at the early versions of the curves mentioned above, extended the lines, and said, "Huh... These lines are so sharp. Why wouldn't it keep going? It seems like it would."

And they were right. The scaling curves may break at some point. But they don't show indications of that yet.

Lastly, all of this is largely just taking existing model architectures and scaling up. Neural nets are a very young technology. There will be better architectures in the future.

Re: OpenAI researchers warned board of AI breakthrough ahead of CEO ouster

#614

Post by doktrin »

skepticATX wrote: post #228

> OpenAI spokesperson Lindsey Held Bolton refuted that notion in a statement shared with The Verge: “Mira told employees what the media reports were about but she did not comment on the accuracy of the information.” > Separately, a person familiar with the matter told The Verge that the board never received a letter about such a breakthrough and that the company’s research progress didn’t play a role in Altman’s sudd…

We’ll find out sooner or later. Personally, if the Verge and their “source” turn out to be incorrect I’ll permanently file them away under the “gossip rag” folder.

Re: OpenAI researchers warned board of AI breakthrough ahead of CEO ouster

#617

Post by mollerhoj »

Pet theory: Could it be some bio hacking/disease construction thing they have discovered? I know nothing of the subject (which is probably why it seem a possibility to me) but what if they have discovered something to that makes engineering pandemics significantly easier?

Re: OpenAI researchers warned board of AI breakthrough ahead of CEO ouster

#618

Post by stephenboyd »

wbhart wrote: post #491

I feel very comfortable saying, as a mathematician, that the ability to solve grade school maths problems would not be at all a predictor of ability to solve real mathematical problems at a research level. The reason LLMs fail at solving mathematical problems is because: 1) they are terrible at arithmetic, 2) they are terrible at algebra, but most importantly, 3) they are terrible at complex reasoning (more specifica…

Did they say it was an LLM? I didn’t see that in the reporting.

Re: OpenAI researchers warned board of AI breakthrough ahead of CEO ouster

#620

Post by reroute22 »

theonemind wrote: post #463
xanderlewis wrote: post #37

Earlier quoted context omitted.

Mathematical logic doesn’t refer to ‘complex word problems with big numbers’. The reason that LLMs struggle (well, are hopeless currently) with mathematics problems is not because it involves arithmetic (although this is already enough of a struggle) but because it requires a conceptual understanding of the ideas rather than just being about manipulating language — a lot of the goings-on are hidden in the semantic wo…

That's an interesting observation. It seems that in theory, you could train it to do math if you encoded literally everything in syntax and started at the beginning, like Principia Mathematica 50 pages proving 1+1=2 type beginning, and then the rest of known mathematics encoded similarly, and used that as the training data, although the context window limitations might still stop it from doing anything useful, and it…

https://arxiv.org/abs/2310.02989 ("xVal: A Continuous Number Encoding for Large Language Models", Submitted on 4 Oct 2023)
Post reply on HN