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

Notes on OpenAI's new o1 chain-of-thought models

simonwillison.net

Post reply on HN
Print view View original

311–320 of 659 posts

Re: Notes on OpenAI's new o1 chain-of-thought models

#311

Post by kristianp »

roomey wrote: post #223
grey-area wrote: post #199

Earlier quoted context omitted.

The 'riddle': A woman and her son are in a car accident. The woman is sadly killed. The boy is rushed to hospital. When the doctor sees the boy he says "I can't operate on this child, he is my son". How is this possible? GPT Answer: The doctor is the boy's mother Real Answer: Boy = Son, Woman = Mother (and her son), Doctor = Father (he says...he is my son) This is not in fact a riddle (though presented as one) and th…

Why couldn't the doctor be the boys mother? There is no indication of the sex of the doctor, and families that consist of two mothers do actually exist and probably doesn't even count as that unusual.

So the riddle could have two answers: mother or father? Usually riddles have only one definitive answer. There's nothing in the wording of the riddle that excludes the doctor being the father.

Re: Notes on OpenAI's new o1 chain-of-thought models

#312

Post by jari_mustonen »

Once again, there’s a lot of safety talk. For example, OpenAI’s collaborations with NGOs and government agencies are being highlighted in the release notes. While it’s crucial to prevent AI from facilitating genuinely harmful activities—like instructing someone on building a nuclear bomb, there is an elephant in the room regarding safety talk: Evidence suggests that these safety protocols sometimes censor specific political perspectives.

OpenAI and other AI vendors should recognize the widespread suspicion that safety policies are being used to push political agendas. Concrete remedies are called for—for example, clearly defining what “safety” means and specifying prohibited content to reduce suspicions of hidden agendas.

Openly engaging with the public to address concerns about bias and manipulation is a crucial step. If biases are due to innocent reasons like technical limitations, they should be explained. However, if there’s evidence of political bias within teams testing AI systems, it should be acknowledged, and corrective actions should be taken publicly to restore trust.

Re: Notes on OpenAI's new o1 chain-of-thought models

#313

Post by nsokolsky »

lucianbr wrote: post #294
mediaman wrote: post #63

Earlier quoted context omitted.

The simple answer is: I don't care. I'll statistically figure out what the typical total cost per call is from experience, and that's what matters. Who cares if they lie about it, if the model's cost per call fits my budget? If it starts costing $1 per call, and that's too high, then I just won't use it commercially. Whether it was $1 because they inflated the token count or because it just actually took a lot of tok…

The thing is it might increase in cost after you've decided to use it commercially, and have invested a lot of time and resources in it. Now it's very hard to move to something else, but very easy for OpenAI to increase your cost arbitrarily. The statistics you made are not binding for them.

Same applies to every other API in the world, yes.

Re: Notes on OpenAI's new o1 chain-of-thought models

#315

Post by gloosx »

Fraterkes wrote: post #255
gloosx wrote: post #191

Earlier quoted context omitted.

That's a classical school question, word-by-word, except multiplication is replaced by division

Order doesn't matter with multiplication (eg: (20 * 5) * 2 == (5 * 2) * 20) but it obviously does with division ((20/5)/2 != (2/5)/20) so the question doesn't make sense. It's you making grade-school level mistakes here.

The question makes perfect sense. Here it is written in logical language. I'm curious at which point does it stop making sense for you?

  numbers divided together  
    ↓----------↓ 
    ((a / b / c) = a + b + c) ← numbers added together
  | ((a / c / b) = a + b + c)
  | ((b / a / c) = a + b + c)
  | ((b / c / a) = a + b + c)
  | ((c / a / b) = a + b + c)
  | ((c / b / a) = a + b + c)
  | ((a / (b / c)) = a + b + c)
  | ((a / (c / b)) = a + b + c)
  | ((b / (a / c)) = a + b + c)
  | ((b / (c / a)) = a + b + c)
  | ((c / (a / b)) = a + b + c)
  | ((c / (b / a)) = a + b + c) = true

Re: Notes on OpenAI's new o1 chain-of-thought models

#316

Post by KoolKat23 »

bambax wrote: post #281

Near the end, the quote from OpenAI researcher Jason Wei seems damning to me: > Results on AIME and GPQA are really strong, but that doesn’t necessarily translate to something that a user can feel. Even as someone working in science, it’s not easy to find the slice of prompts where GPT-4o fails, o1 does well, and I can grade the answer. But when you do find such prompts, o1 feels totally magical. We all need to find…

> Results are "strong" but can't be felt by the user? What does that even mean?

This means it often doesn't provide the answer the user is looking for. In my opinion, it's an alignment problem, people are very presumptuous and leave out a lot of detail in their request. Like the "which is bigger - 9.8 or 9.11? question, if you ask "numerically which is bigger - 9.8 or 9.11?" It gets the correct answer, basically it prioritizes a different meaning for bigger.

> But the last sentence is the worst: "we all need to find harder prompts". If I understand it correctly, it means we should go looking for new problems / craft specific questions that would let these new models shine. But why? Why would we do that? Wouldn't our time be better spent trying to solve our actual, current problems, using any tool available?

Without better questions we can't test and prove that it is getting more intelligent or is just wrong. If it is more intelligent than us it it might provide answers that don't make sense to us but are actually clever, 4d chess as they say. Again an alignment problem, better questions aid with solving that.

Re: Notes on OpenAI's new o1 chain-of-thought models

#317

Post by singingfish »

williamdclt wrote: post #164
MPSimmons wrote: post #106

Earlier quoted context omitted.

The failure is in how you're using it. I don't mean this as a personal attack, but more to shed light on what's happening. A lot of people use LLMs as a search engine. It makes sense - it's basically a lossy compressed database of everything its ever read, and it generates output that is statistically likely - varying degrees of likeliness depending on the temperature, as well as how many times the particular weights…

> Treat it as a naive but intelligent intern That’s the problem: it’s a _terrible_ intern. A good intern will ask clarifying questions, tell me “I don’t know” or “I’m not sure I did it right”. LLMs do none of that, they will take whatever you ask and give a reasonable-sounding output that might be anything between brilliant and nonsense. With an intern, I don’t need to measure how good my prompting is, we’ll usually…

> LLMs do none of that, they will take whatever you ask and give a reasonable-sounding output that might be anything between brilliant and nonsense.

This is exactly why I’ve been objecting so much to the use of the term “hallucination” and maintain that “confabulation” is accurate. People who have spent enough time with acutelypsychotic people, and people experiencing the effects of long term alcohol related brain damage, and trying to tell computers what to do will understand why.

Re: Notes on OpenAI's new o1 chain-of-thought models

#318

Post by dailykoder »

I am mostly only an LLM user with technical background. I don't have much in-depth knowledge. So I have questions about this take:

>the output token allowance has been increased dramatically—to 32,768 for o1-preview and 65,536 for the supposedly smaller o1-mini!

So the text says reasoning and output tokens are the same, as in you pay for both. But does the increase say that it can actually do more, or does it just mean it is able to output more text?

Because by now I am just bored of GPT4o output, because I don't have the time to read through a multi-paragraph text that explains to me stuff that I already know, when I only want to have a short, technical answer. But maybe that's just what it can't do, give exact answers. I am still not convinced by AI.

Re: Notes on OpenAI's new o1 chain-of-thought models

#319

Post by singingfish »

roomey wrote: post #223
grey-area wrote: post #199

Earlier quoted context omitted.

The 'riddle': A woman and her son are in a car accident. The woman is sadly killed. The boy is rushed to hospital. When the doctor sees the boy he says "I can't operate on this child, he is my son". How is this possible? GPT Answer: The doctor is the boy's mother Real Answer: Boy = Son, Woman = Mother (and her son), Doctor = Father (he says...he is my son) This is not in fact a riddle (though presented as one) and th…

Why couldn't the doctor be the boys mother? There is no indication of the sex of the doctor, and families that consist of two mothers do actually exist and probably doesn't even count as that unusual.

Speaking as a 50-something year old man whose mother finished her career in medicine and the very pointy end of politics, when I first heard this joke in the 1980s it stumped me and made me feel really stupid. But my 1970s kindergarten class mates who told me “your mum can’t be a doctor, she has to be a nurse” were clearly seriously misinformed then. I believe that things are somewhat better now but not as good as they should be …

Re: Notes on OpenAI's new o1 chain-of-thought models

#320

Post by oersted »

throwaway40602 wrote: post #288
oersted wrote: post #265

Earlier quoted context omitted.

Kagi is most likely evaluating it mainly on deriving an answer for the user from search result snippets. Indeed, GPT-4o is plenty good at this already, and o1 would only perform better on particular types of hard requests, while being so much slower. If you look at Appendix A in the o1 post [1], this becomes quite clear. There's a huge jump in performance in "puzzle" tasks like competitive maths or programming. But t…

hey buddy, you're talking to owner of kagi, and the kagi benchmark is a traditional one

My bad, you are right, should have looked into it better, I was too dismissive. Still I think that highlighting those charts from OpenAI is important.
Post reply on HN