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

591–600 of 659 posts

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

#591

Post by ww2supercut »

jasondigitized wrote: post #564
williamdclt wrote: post #164

Earlier quoted context omitted.

> 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…

Have you ever worked with an intern? They have personalities and expectations that need to be managed. They get sick. The get tired. They want to punch you if you treat them like a 24-7 bird dog. It's so much easier to not let perfect be the enemy of the good and just rapid fire ALL day at a LLM for any and everything I need help with. You can also just not use the LLM. Interns need to be 'fed' work or the ROI ends u…

The LLMs are getting better and better at a certain kind of task, but there's a subset of tasks that I'd still much rather have any human than an LLM, today. Even something simple, like "Find me the top 5 highest grossing movies of 2023" it will take a long time before I trust an LLM's answer, without having a human intern verify the output.

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

#592

Post by franze »

SeanAnderson wrote: post #530
franze wrote: post #402

Just coded this this morning using chatgpt o1 - it is the reimplementation of an old idea now music, multiple dots, more and more bug fixes honestly, chatgpt is now a better coder than i ever was or will be https://lsd.franzai.com/

Neat idea. The ball frequently passes through solid lines though.

fixed, just asked chatgpt to come up with a better physics engine and collision detection algorithm

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

#593

Post by HarHarVeryFunny »

m101 wrote: post #588
HarHarVeryFunny wrote: post #451

I think Rich Sutton's bitter lesson will prove to apply here, and what we really need to advance machine learning capabilities are more general and powerful models capable of learning for themselves - better able to extract and use knowledge from the firehose of data available from the real world (ultimately via some form of closed-loop deployment where they can act and incrementally learn from their own actions). Wh…

It works like our own minds in that we also think, test, go back, try again. This doesn't seem like a failing but just a recognition that thought can proceed in that way.

The "failing" here isn't the short term functional gains, but rather the choice of architectural direction. Trying to add reasoning as an ad-hoc wrapper around the base model, based on some fixed reasoning heuristics (built in biases) is really a dead-end approach. It would be better to invest in a more powerful architecture capable of learning at runtime to reason for itself.

Bespoke hand-crafted models/agents can never compete with ones that can just be scaled and learn for themselves.

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

#594

Post by madeofpalk »

OutOfHere wrote: post #37
jumploops wrote: post #3

> the idea that I can run a complex prompt and have key details of how that prompt was evaluated hidden from me feels like a big step backwards. As a developer, this is highly concerning, as it makes it much harder to debug where/how the “reasoning” went wrong. The pricing is also silly, because I’m paying for tokens I can’t see. As a user, I don’t really care. LLMs are already magic boxes and I usually only care abo…

> As a user, I don’t really care. Tell me: Just how is it fair for a user to pay for the reasoning tokens without actually seeing them? If they are not shared, the service can bill you anything they want for them!

Pricing for many things in life is abstrated away.

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

#595

Post by jrflowers »

MPSimmons wrote: post #106
layer8 wrote: post #18

The o1-preview model still hallucinates non-existing libraries and functions for me, and is quickly wrong about facts that aren't well-represented on the web. It's the usual string of "You're absolutely correct, and I apologize for the oversight in my previous response. [Let me make another guess.]" While the reasoning may have been improved, this doesn't solve the problem of the model having no way to assess if what…

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…

> The failure is in how you're using it

This isn’t true because, as you can read in the first sentence of the post you’re responding to, GP did give it a task like you recommend here

> Provide it data, give it a task, and let it surprise you with its output.

And it fails the task. Specifically it fails it by hallucinating important parts of accomplishing it.

> hallucinates non-existing libraries and functions

This post only makes sense if your advice to “let it surprise you with its output” is mandatory, like you’re using it wrong if you do not make yourself feel impressed by it.

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

#596

Post by ttul »

8n4vidtmkvmk wrote: post #181
ttul wrote: post #48

Earlier quoted context omitted.

It seems OpenAI has decided to keep the CoT results a secret. If they were to allow the model to call out to tools to help fill in the CoT steps, then this might reveal what the model is thinking - something they do not want the outside world to know about. I could imagine OpenAI might allow their own vetted tools to be used, but perhaps it will be a while (if ever) before developers are allowed to hook up their own…

It's only a matter of time. When some other company releases the tool, they likely will too.

I have to agree with you here. OpenAI may be playing for competitive advantage more than for the good of humanity by hiding the results.

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

#597

Post by la64710 »

starbugs wrote: post #578
TechDebtDevin wrote: post #97

Earlier quoted context omitted.

Okay, what is a thought then?

Something in the mind. (Didn't make that up. It's one of the definitions of Merriam Webster: https://www.merriam-webster.com/dictionary/thought )

Exactly and now please don’t say AI has a mind …

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

#598

Post by la64710 »

sebzim4500 wrote: post #277
la64710 wrote: post #79

Please please please stop saying thought. This has nothing to do with the word thought. When we say the word thought it means something. Please don’t use the same word for whatever AI is doing and trivialize the word. Invent a new word if needed but for Pete’s sake be accurate and truthful.

No one gets this emotional about astrophysicists calling almost everything 'metal' and this is definitely less bad than that.

It’s way worse than that .. next you know we will be taking about AI’s mind and AI’s soul and how have a soul purer than us … just so they can sell you a few damn chips.

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

#599

Post by la64710 »

poikroequ wrote: post #120
la64710 wrote: post #79

Please please please stop saying thought. This has nothing to do with the word thought. When we say the word thought it means something. Please don’t use the same word for whatever AI is doing and trivialize the word. Invent a new word if needed but for Pete’s sake be accurate and truthful.

It's called terminology. Every field has words that mean very different things from the layman's definition. It's nothing to get upset about.

Not upset but saddened and disappointed … this is how snake oil was sold.

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

#600

Post by poikroequ »

starbugs wrote: post #578
TechDebtDevin wrote: post #97

Earlier quoted context omitted.

Okay, what is a thought then?

Something in the mind. (Didn't make that up. It's one of the definitions of Merriam Webster: https://www.merriam-webster.com/dictionary/thought )

Way to cherry pick the one definition that fits your argument and ignore all the other definitions which kinda fit for what this model is doing.
Post reply on HN