Earlier quoted context omitted.
Best case: Hallucination Worst case (more probable): Lying
Hanlon's Razor
Lying is a stupid way of selling something and making money
OpenAI researcher announced GPT-5 math breakthrough that never happened
41–50 of 258 posts
Re: OpenAI researcher announced GPT-5 math breakthrough that never happened
#42I try not to lose sight of the first time that I heard (some years back) that people were using this new LLM thing for DM'ing ("dungeon mastering", leading) a game of Dungeons and Dragons. I thought, this must be bullshit or some kind of witchcraft.
Definitely not anti-AI here. I think I have been disappointed though, since then, to slowly learn that they're (still) little beyond that.
Still amazing though. And better than a Google search (IMHO).
Re: OpenAI researcher announced GPT-5 math breakthrough that never happened
#43How fing obvious was it that AI slop did not do anything other than scarpe some websites.
I felt like I was going crazy when people uncritically accepted the original claim from OpenAI. Have people actually used these models?
Re: OpenAI researcher announced GPT-5 math breakthrough that never happened
#44This entire thing has been pretty disingenuous on both sides of the fence. All the anti-AI (or anti OpenAI) people are doing victory laps, but what GPT-5 Pro did is still very valuable. 1) What good is your open problem set if really its a trivial "google search" away from being solved. Why are they not catching any blame here? 2) These answers still weren't perfectly laid out for the most part. GPT-5 was still doing…
> What good is your open problem set if really its a trivial "google search" away from being solved. Why are they not catching any blame here? They are a community run database, not the sole arbiter and source of this information. We learned the most basic research back in highschool, I'd hope researchers from top institutions now working for one of the biggest frontier labs can do the same prior to making a claim, b…
[dead]
Re: OpenAI researcher announced GPT-5 math breakthrough that never happened
#45Earlier quoted context omitted.
Humans "hallucinate" in the AI way constantly, which is why I don't see them as a barrier to LLMs replacing humans in many contexts. It really isn't unusual for a human to make stuff up or be unaware of stuff.
A human being informed of a mistake will usually be able to resolve it and learn something in the process, whereas an LLM is more likely to spiral into nonsense
You must know better humans than I do.
Re: OpenAI researcher announced GPT-5 math breakthrough that never happened
#46Yann LeCun's "Hoisted by their own GPTards" is fantastic.
Re: OpenAI researcher announced GPT-5 math breakthrough that never happened
#47This entire thing has been pretty disingenuous on both sides of the fence. All the anti-AI (or anti OpenAI) people are doing victory laps, but what GPT-5 Pro did is still very valuable. 1) What good is your open problem set if really its a trivial "google search" away from being solved. Why are they not catching any blame here? 2) These answers still weren't perfectly laid out for the most part. GPT-5 was still doing…
Framing this question as "AI good" OR "AI bad" is culture-war thinking.
The real problem here is that there's clearly a strong incentive for the big labs to deceive the public (and/or themselves) about the actual scientific and technical capabilities of LLMs. As Karpathy pointed out on the recent Dwarkesh podcast, LLMs are quite terrible at novel problems, but this has become sort of an "Emperor's new clothes" situation where nobody with a financial stake will actually admit that, even though it's common knowledge if you actually work with these things.
And this directly leads to the misallocation of billions of dollars and potentially trillions in economic damage as companies align their 5-year strategies towards capabilities that are (right now) still science fiction.
The truth is at stake.
Re: OpenAI researcher announced GPT-5 math breakthrough that never happened
#48Re: OpenAI researcher announced GPT-5 math breakthrough that never happened
#49The sad truth about this incident is that it reveals that OpenAI does not have a serious effort to actually work on unsolved math problems.
How so? I wouldn't put much stock into a roque employee announcing something wrong.
Re: OpenAI researcher announced GPT-5 math breakthrough that never happened
#50Another instance of openAI manipulating results to prolong their unsustainable circular hype bubble. The inevitable collapse could be even more devastating than the 2008 financial crisis. All while so vast resources are being wasted on non-verifiable gen AI slob, while real approaches (neuro-symbolic like DeepMind's AlphaFold) are mostly ignored financially because they don't generate the quick stock market increases…
People keep spouting this, but I don't see how the AI bubble bursting would be all that devastating. 2008 was a systemic breakdown rippling through the foundations of the financial system. It would lead to a market crash (80% of gains this year were big tech/AI) and likely a full recession in the US, but nothing nearly as dramatic as a global systemic crisis. In contrast to the dot com bubble, the huge AI spending is…
I think my theory into contagion would be that There’s been lots of talk about these companies starting to rack up debt, and I think AI is so tied into the US GDP that things like -
If the stock market crashes, there’s lots of talk about how wealth and debt are interlinked. Could the crash be general enough to start calls on debt backed by stocks.
My recollection in 2008 was that we didn’t learn about how bad it was until after. The tech companies have been so desperate for a win, I wonder if some of them are over their skis in some way, and if there are banks that are risking it all on AI. (We know for some tech bros think the bet on AI is a longtermist like bet; closer to religion than reason and that it’s worth risking everything because the payback could be in the hundreds of trillions)
Combine this with the fact that AI is like what - 30% of the US economy? Magnificent 7 are 60%?
What happens if sustainable PE ratios in tech collapse. Does it take out Tesla?
Maybe the contagion is just the impact on the US economy which, classically anyways has been intermingled with everything.
I would bet almost everything that there is some lie at the center of this thing that we aren’t really aware of yet.