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

GPT-5 is behind schedule

wsj.com

Post reply on HN
Print view View original

191–200 of 1001 posts

Re: GPT-5 is behind schedule

#191

Post by glenstein »

The lack of tech literacy in this article is a bit concerning:

>Some researchers take this so seriously they won’t work on planes, coffee shops or anyplace where someone could peer over their shoulder and catch a glimpse of their work.

I'm almost certain that originally this was meant to be a reference to public wifi networks, as planes and coffee shops are often the frequently cited prototypical examples. They made it literally into a matter of someone looking over their shoulder, which loses so much in translation it's almost how you would write this as a joke to illustrate someone missing the point.

>OpenAI and its brash chief executive, Sam Altman

This also strikes me as nonsense. It's the first I've ever heard of someone describing Sam Altman as brash. The only way I can see them getting there is (1) tech executives are often brash (2) Altman is a tech executive (3) let's just go ahead and call him brash.

Nevertheless if this history of GPT5 and/or o3 training is accurate, it strikes me as significant news, but perhaps a missed opportunity to say more about the pertinent dynamics that explain why the training isn't working and/or to talk in interestingly specific ways about strategies for training, synthetic data, or other such things.

Re: GPT-5 is behind schedule

#192

Post by kevingadd »

layer8 wrote: post #183
SpicyLemonZest wrote: post #129

Earlier quoted context omitted.

Laundry folding is an instructive example. Machines have been capable of home-scale laundry folding for over a decade, with two companies Foldimate and Laundroid building functional prototypes. The challenge is making it cost-competitive in a world where most people don't even purchase a $10 folding board. I would guess that most cooking and cleaning tasks are in basically the same space. You don't need fine motor co…

Counterexample: Everyone uses dishwashers. Yet I don’t think we’ll have a robot doing the dishes human-style, or even just filling up and clearing out a dishwasher, within the next decade or two, regardless of price.

Part of the tradeoff there is efficiency. I like my dishwasher because it's as good at getting things clean as I am but it does it using less water and less soap, and at scale, it takes less time too. It's just a great use case for machine automation because you can do clever stuff w/a dishwasher that's hard to replicate outside of that closed environment.

I struggle to imagine a scenario where a 1-2 person household would get the same benefits from something like a laundry-folding robot. I hate folding my laundry and I still can't imagine buying one since I simply don't do laundry that often. If I really wanted to spend less time doing laundry, I could spend the cost of that laundrybot on a larger collection of clothing to wear, for that matter.

Robot vacuums are a good comparison point since vacuuming is something you (ideally) do frequently that is time and labor intensive. I do own one of those, and if it got better at dealing with obstacles thanks to "AI" I would definitely like that.

Re: GPT-5 is behind schedule

#193

Post by stuartjohnson12 »

achierius wrote: post #176
bobxmax wrote: post #132

Earlier quoted context omitted.

All frontier labs are now employing LVMs or LLMs. But that's my point is you won't see the fruits of it this early.

That's the point being made. It's transformed robotics research, yes, but it both remains to see whether it will have a truly transformative effect on the field as experienced by people outside academia (I think this is quite probable) and more pointedly when .

I think this is an opinion borne out of weariness with constant promises that amazing robots are right around the corner (as they have been for 20 odd years now). For anyone who is close to the front line, I think the resounding consensus is clear - this time is different, unbelievably different, and capability development is going to accelerate dramatically.

Re: GPT-5 is behind schedule

#194

Post by CSMastermind »

mathieuh wrote: post #26
phito wrote: post #5

Earlier quoted context omitted.

I keep reading this on HN so I believe it has to be true in some ways, but I don't really feel like there is any difference in my limited use (programming questions or explaining some concepts). If anything I feel like it's all been worse compared to the first release of ChatGPT, but I might be wearing rose colored glasses.

It’s the same for me. I genuinely don’t understand how I can be having such a completely different experience from the people who rave about ChatGPT. Every time I’ve tried it’s been useless. How can some people think it’s amazing and has completely changed how they work, while for me it makes mistakes that a static analyser would catch? It’s not like I’m doing anything remarkable, for the past couple of months I’ve b…

I mean if you're getting no value out of ChatGPT I'd love to have a session seeing how you use it.

Re: GPT-5 is behind schedule

#195

Post by famouswaffles »

BoorishBears wrote: post #184
glenstein wrote: post #180

Earlier quoted context omitted.

>GPT-5 is called GPT-4o This is the first I have heard of this in particular. Do you know of any article or source for more on the efforts to train GPT 5 and the decision to call it GPT 4o?

I think my biggest pet peeve is when someone shares an insight which is unmistakably based on intuition, inference, critical thinking, etc (all mental faculties we are allowed to use to come to conclusions in the face of information asymmetry btw) ...and then gets hit deadpan with the good old "Source?", like it's some sort of gotcha. I think people have started to confuse "making logical conclusions without perfect…

It doesn't even look like 4o is scaled up parameter wise from 4 and was released closer in time than either 3 or 4 were from their predecessors at a time where the scaling required for these next gen iterations has only gotten more difficult.

Critical thinking ? Lol it's just blind speculation.

Re: GPT-5 is behind schedule

#196

Post by zapnuk »

vasco wrote: post #162
simonw wrote: post #146

Earlier quoted context omitted.

No, I'm complaining that just because GPT-4 is called GPT-4 doesn't mean it's the fourth LLM from OpenAI. Off the top of my head: GPT-2, Codex, GPT-3 in three different flavors (babbage, curie, davinci), GPT-3.5. Suggesting that GPT-4 was "fourth" simply isn't credible. Just the other day they announced a jump from o1 to o3, skipping o2 purely because it's already the name of a major telecommunications brand in Europ…

Imagine coming up with a naming scheme for the versioning of your product just for it to fail on the second time you want to use it.

Should have used chatGPT to ask for a name or at least check it

Re: GPT-5 is behind schedule

#197

Post by imiric »

BoorishBears wrote: post #184
glenstein wrote: post #180

Earlier quoted context omitted.

>GPT-5 is called GPT-4o This is the first I have heard of this in particular. Do you know of any article or source for more on the efforts to train GPT 5 and the decision to call it GPT 4o?

I think my biggest pet peeve is when someone shares an insight which is unmistakably based on intuition, inference, critical thinking, etc (all mental faculties we are allowed to use to come to conclusions in the face of information asymmetry btw) ...and then gets hit deadpan with the good old "Source?", like it's some sort of gotcha. I think people have started to confuse "making logical conclusions without perfect…

It's reasonable to ask for sources when an opinion is phrased as a fact, as GGP did. I don't see how you got that it was _unmistakably_ an opinion from that comment.

There is no way to deduce by intuition alone that GPT-5 == GPT-4o. So either that person has some information the rest of us aren't privy to, or it's an opinion phrased as a fact. In either case, it deserves clarification.

Re: GPT-5 is behind schedule

#199

Post by riku_iki »

norir wrote: post #158
peepeepoopoo98 wrote: post #145

Earlier quoted context omitted.

O3 has demonstrated that OpenAI needs 1,000,000% more inference time compute to score 50% higher on benchmarks. If O3-High costs about $350k an hour to operate, that would mean making O4 score 50% higher would cost $3.5B (!!!) an hour. That scaling wall.

I used to run a lot of monte carlo simulations where the error is proportional to the inverse square root. There was a huge advantage of running for an hour vs a few minutes, but you hit the diminishing returns depressingly quickly. It would not surprise me at all if llms end up having similar scaling properties.

And I suspect o3 is something like monte carlo: generates tons of CoTs, with most of them are junk, but some hit the answer.

Re: GPT-5 is behind schedule

#200

Post by riku_iki »

famouswaffles wrote: post #167
peepeepoopoo98 wrote: post #145

Earlier quoted context omitted.

O3 has demonstrated that OpenAI needs 1,000,000% more inference time compute to score 50% higher on benchmarks. If O3-High costs about $350k an hour to operate, that would mean making O4 score 50% higher would cost $3.5B (!!!) an hour. That scaling wall.

Not really. o3-low compute still stomps the benchmarks and isn't anywhere that expensive and o3-mini seems better than o1 while being cheaper. Combine that with the fact that LLM inference has reduced orders of magnitudes in cost the last few years and hampering over the inference costs of a new release seems a bit silly.

If you are talking about ARC benchmark, then o3-low doesn't look that special if you take into account there are plenty of finetuned models with much smaller resources achieved 40-50% results on private set (not semi-private like o3-low).
Post reply on HN