3 ms·

> It seems increasingly apparent that we are reaching the limits of throwing more data at more GPUs Yes. This is exactly why I'm skeptical of AI doomerism/savi

> It seems increasingly apparent that we are reaching the limits of throwing more data at more GPUs Yes. This is exactly why I'm skeptical of AI doomerism/saviorism. Too many people have been looking at the pace of LLM development over the last two (2) years, modeled it as an exponential growth function, and come to the conclusion that AGI is inevitable in the next ${1-5} years and we're headed for ${(dys|u)topia}. But all that assumes that we can extrapolate a pattern of long-term exponential growth from less than two years of data. It's simply not possible to project in that way, and we're already seeing that OpenAI has pivoted from improving on GPT-4's benchmarks to reducing cost, while competitors (including free ones) catch up. All the evidence suggests we've been slowing the rate of growth in capabilities of SOTA LLMs for at least the past year, which means predictions based on exponential growth all need to be reevaluated.

cjalmeida 2y ago
Indeed.All exponential growth curves are sigmoids in disguise.
- nicman23 2y ago
  except when it isn't and we ded :P
  - kridsdale3 2y ago
    I don't think Special Relativity would allow that.
- ToValueFunfetti 2y ago
  This is something that is definitionally true in a finite universe, but doesn't carry a lot of useful predictive value in practice unless you can identify when the flattening will occur. If you have a machine that converts mass into energy and then uses that energy to increase the rate at which it operates, you could rightfully say that it will level off well before consuming all of the mass in the universe. You just can't say that next week after it has consumed all of the mass of Earth.
RicoElectrico 2y ago
I don't think we are approaching limits, if you take off the English-centric glasses. You can query LLMs about pretty basic questions about Polish language or literature and it's gonna either bullshit or say it doesn't know the answer. Example: w której gwarze jest słowo ekspres i co znaczy? Słowo "ekspres" występuje w gwarze śląskiej i oznacza tam ekspres do kawy. Jest to skrót od nazwy "ekspres do kawy", czyli urządzenia służącego do szybkiego przygotowania kawy. The correct answer is that "ekspres" is a zipper in Łódź dialect.
- andrepd 2y ago
  Tbf, you can ask it basic questions in English and it will also bullshit you.
- nprateem 2y ago
  That's just same same but different, not a step change towards significant cognitive ability.
- lolinder 2y ago
  What this means is just that Polish support (and probably most other languages besides English) in the models is behind SOTA. We can gradually get those languages closer to SOTA, but that doesn't bring us closer to AGI.
jeremyjh 2y ago
I'm also wondering about the extent to which we are simply burning venture capital versus actually charging subscription prices that are sustainable long-term. Its easy to sell dollars for $0.75 but you can only do that for so long.
dvansoye 2y ago
What about synthetic data?
impossiblefork 2y ago
Notice though, that all these improvements have been with pretty basic transformer models that output all their tokens-- no internal thoughts, no search, no architecture improvements and things are only fed through them once. But we could add internal thoughts-- we could make the model generate tokens that aren't part of its output but are there for it to better figure out its next token. This was tried QuietSTAR. Hochreiter is also active with alternative models, and there's all the microchip design companies, Groq, Etched, etc. trying to speed up models and reduce model running cost. Therefore, I think there's room for very great improvements. They may not come right away, but there are so many obvious paths to improve things that I think it's unreasonable to think progress has stalled. Also, presumably GPT-5 isn't far away.
- lolinder 2y ago
  > Also, presumably GPT-5 isn't far away. Why do we presume that? People were saying this right before 4o and then what came out was not 5 but instead a major improvement on cost for 4. Is there any specific reason to believe OpenAI has a model coming soon that will be a major step up in capabilities?
  - impossiblefork 2y ago
    OpenAI have made statements saying they've begun training it, as they explain here: https://openai.com/index/openai-board-forms-safety-and-security-committee/ https://openai.com/index/openai-board-forms-safety-and-secur... I assume that this won't take forever, but will be done this year. A couple of months, not more.
- audunw 2y ago
  > But we could add internal thoughts It feels like there’s an assumption in the community that this will be almost trivial. I suspect it will be one of the hardest tasks humanity has ever endeavoured. I’m guessing it has already been tried many times in internal development. I suspect if you start creating a feedback loop with these models they will tend to become very unstable very fast. We already see with these more linear LLMs that they can be extremely sensitive to the values of parameters like the temperature settings, and can go “crazy” fairly easily. With feedback loops it could become much harder to prevent these AIs from spinning out of control. And no I don’t mean in the “become an evil paperclip maximiser” kind of way. Just plain unproductive insanity. I think I can summarise my vision of the future in one sentence: AI psychologists will become a huge profession, and it will be just as difficult and nebulous as being a human psychologist.
jpadkins 2y ago
> we're already seeing that OpenAI has pivoted from improving on GPT-4's benchmarks to reducing cost, while competitors (including free ones) catch up. What if they have two teams? One dedicated to optimizing (cost, speed, etc) the current model and a different team working on the next frontier model? I don't think we know the growth curve until we see gpt5.
- lolinder 2y ago
  > I don't think we know the growth curve until we see gpt5. I'm prepared to be wrong, but I think that the fact that we still haven't seen GPT-5 or even had a proper teaser for it 16 months after GPT-4 is evidence that the growth curve is slowing. The teasers that the media assumed were for GPT-5 seem to have actually been for GPT-4o [0]: > Lex Fridman(01:06:13) So when is GPT-5 coming out again? > Sam Altman(01:06:15) I don’t know. That’s the honest answer. > Lex Fridman(01:06:18) Oh, that’s the honest answer. Blink twice if it’s this year. > Sam Altman(01:06:30) We will release an amazing new model this year. I don’t know what we’ll call it. > Lex Fridman(01:06:36) So that goes to the question of, what’s the way we release this thing? > Sam Altman(01:06:41) We’ll release in the coming months many different things. I think that’d be very cool. I think before we talk about a GPT-5-like model called that, or not called that, or a little bit worse or a little bit better than what you’d expect from a GPT-5, I think we have a lot of other important things to release first. Note that last response. That's not the sound of a CEO who has an amazing v5 of their product lined up, that's the sound of a CEO who's trying to figure out how to brand the model that they're working on that will be cheaper but not substantially better. [0] https://arstechnica.com/information-technology/2024/03/openais-gpt-5-may-launch-this-summer-upgrading-chatgpt-along-the-way/ https://arstechnica.com/information-technology/2024/03/opena...

cjalmeida 2y ago

Indeed.All exponential growth curves are sigmoids in disguise.

nicman23 2y ago
except when it isn't and we ded :P
- kridsdale3 2y ago
  I don't think Special Relativity would allow that.
ToValueFunfetti 2y ago
This is something that is definitionally true in a finite universe, but doesn't carry a lot of useful predictive value in practice unless you can identify when the flattening will occur. If you have a machine that converts mass into energy and then uses that energy to increase the rate at which it operates, you could rightfully say that it will level off well before consuming all of the mass in the universe. You just can't say that next week after it has consumed all of the mass of Earth.

RicoElectrico 2y ago

I don't think we are approaching limits, if you take off the English-centric glasses. You can query LLMs about pretty basic questions about Polish language or literature and it's gonna either bullshit or say it doesn't know the answer. Example: w której gwarze jest słowo ekspres i co znaczy? Słowo "ekspres" występuje w gwarze śląskiej i oznacza tam ekspres do kawy. Jest to skrót od nazwy "ekspres do kawy", czyli urządzenia służącego do szybkiego przygotowania kawy. The correct answer is that "ekspres" is a zipper in Łódź dialect.

andrepd 2y ago
Tbf, you can ask it basic questions in English and it will also bullshit you.
nprateem 2y ago
That's just same same but different, not a step change towards significant cognitive ability.
lolinder 2y ago
What this means is just that Polish support (and probably most other languages besides English) in the models is behind SOTA. We can gradually get those languages closer to SOTA, but that doesn't bring us closer to AGI.

jeremyjh 2y ago

I'm also wondering about the extent to which we are simply burning venture capital versus actually charging subscription prices that are sustainable long-term. Its easy to sell dollars for $0.75 but you can only do that for so long.

dvansoye 2y ago

What about synthetic data?

impossiblefork 2y ago

Notice though, that all these improvements have been with pretty basic transformer models that output all their tokens-- no internal thoughts, no search, no architecture improvements and things are only fed through them once. But we could add internal thoughts-- we could make the model generate tokens that aren't part of its output but are there for it to better figure out its next token. This was tried QuietSTAR. Hochreiter is also active with alternative models, and there's all the microchip design companies, Groq, Etched, etc. trying to speed up models and reduce model running cost. Therefore, I think there's room for very great improvements. They may not come right away, but there are so many obvious paths to improve things that I think it's unreasonable to think progress has stalled. Also, presumably GPT-5 isn't far away.

lolinder 2y ago
> Also, presumably GPT-5 isn't far away. Why do we presume that? People were saying this right before 4o and then what came out was not 5 but instead a major improvement on cost for 4. Is there any specific reason to believe OpenAI has a model coming soon that will be a major step up in capabilities?
- impossiblefork 2y ago
  OpenAI have made statements saying they've begun training it, as they explain here: https://openai.com/index/openai-board-forms-safety-and-security-committee/ https://openai.com/index/openai-board-forms-safety-and-secur... I assume that this won't take forever, but will be done this year. A couple of months, not more.
audunw 2y ago
> But we could add internal thoughts It feels like there’s an assumption in the community that this will be almost trivial. I suspect it will be one of the hardest tasks humanity has ever endeavoured. I’m guessing it has already been tried many times in internal development. I suspect if you start creating a feedback loop with these models they will tend to become very unstable very fast. We already see with these more linear LLMs that they can be extremely sensitive to the values of parameters like the temperature settings, and can go “crazy” fairly easily. With feedback loops it could become much harder to prevent these AIs from spinning out of control. And no I don’t mean in the “become an evil paperclip maximiser” kind of way. Just plain unproductive insanity. I think I can summarise my vision of the future in one sentence: AI psychologists will become a huge profession, and it will be just as difficult and nebulous as being a human psychologist.

jpadkins 2y ago

> we're already seeing that OpenAI has pivoted from improving on GPT-4's benchmarks to reducing cost, while competitors (including free ones) catch up. What if they have two teams? One dedicated to optimizing (cost, speed, etc) the current model and a different team working on the next frontier model? I don't think we know the growth curve until we see gpt5.

lolinder 2y ago
> I don't think we know the growth curve until we see gpt5. I'm prepared to be wrong, but I think that the fact that we still haven't seen GPT-5 or even had a proper teaser for it 16 months after GPT-4 is evidence that the growth curve is slowing. The teasers that the media assumed were for GPT-5 seem to have actually been for GPT-4o [0]: > Lex Fridman(01:06:13) So when is GPT-5 coming out again? > Sam Altman(01:06:15) I don’t know. That’s the honest answer. > Lex Fridman(01:06:18) Oh, that’s the honest answer. Blink twice if it’s this year. > Sam Altman(01:06:30) We will release an amazing new model this year. I don’t know what we’ll call it. > Lex Fridman(01:06:36) So that goes to the question of, what’s the way we release this thing? > Sam Altman(01:06:41) We’ll release in the coming months many different things. I think that’d be very cool. I think before we talk about a GPT-5-like model called that, or not called that, or a little bit worse or a little bit better than what you’d expect from a GPT-5, I think we have a lot of other important things to release first. Note that last response. That's not the sound of a CEO who has an amazing v5 of their product lined up, that's the sound of a CEO who's trying to figure out how to brand the model that they're working on that will be cheaper but not substantially better. [0] https://arstechnica.com/information-technology/2024/03/openais-gpt-5-may-launch-this-summer-upgrading-chatgpt-along-the-way/ https://arstechnica.com/information-technology/2024/03/opena...