That’s not at all the topic of discussion. The article is talking about the fact that OpenAI solved a version of the problem that is niche and isn’t the one the math community cares about
At the time that the Millenium Prize problems were formulated, the force term was understood to make the problem more realistic, since real fluids are always going to have external forces applied to them. A blowup that happens under constant gravity, for example, would probably be no less interesting than an entirely unforced blowup. The strategy of constructing impossibly complex external forces to induce a blowup was pioneered by Córdoba and Martínez-Zoroa only over the past few years.
Today that point is moot because.. including this one, the Mill problems that are most likely to have been solved are least likely to have real world impact..
Extrapolating on that, I'll take on the biased hope that almost none of the 100 solutions to be released will have any applications for at least 30 years.. (besides PR wins for AI companies)
In order to counter the fear (my own lonely one) that the 9 big names will not be able to hold the execs to account or get openAI to act "more responsibly" (whatever that means).. in the form of direct hits to new subs or partnerships or funding
I plead guilty to any accusations of (vicarious) sour grapes or sympathy for the weak
Point to you, because.. the committee would make more sense if they also bring Ant to the table.. unfortunately nerds will be nerds, so perhaps, to you, and I'll reluctantly concede, mathematicians deserve to be serfs
Exactly! One effect of the committee might be to benefit Ant by its presence alone.. if only by striking fear in OpenAI.. Ant don't even need to thank them at once
Let's say they do not much on Ant's findings but edge on the NDA with OpenAI. That might even be PR victory for mathematicians
Who is this math community? Since when did they form the consensus that this option is not at all what they care about, before or after they knew about OpenAI’s solution?
What about Tristan Buckmaster and Levent Alpöge, did they also attempt to solve the same challenge? Didn’t they know it wasn’t interesting?
> What about Tristan Buckmaster and Levent Alpöge, did they also attempt to solve the same challenge? Didn’t they know it wasn’t interesting?
Yes, they were working on what is considered a niche case, the option C from the millennium statement for Navier-Stokes. The article explains that clearly, you can just read it and get the details
I wish there were some better way to see, before the IPO, how openAI's actions correlate with their bottomline.
(I'm proAI (for the masses, but green) BUT antiopenAI and a bit less antiAnthropic. For me it's all about the personalities.. the people in oAI are deeply uncool. not so in Ant. Dario is still dangerous, but I guess sexily dangerous lol
Dario is sexily dangerous until you read his essays in full then realize it’s just the same LessWrong sci-fi ideas at the core, all based on absurdly simplified thought experiments designed to justify their priors. Those people are deeply unserious and do not value existing humans. They already made their mind decades ago rereading the fact we need a machine god and that they should be the ones building it.
Just the idea of making a conscious digital being, to then get it to process excel files for its whole existence is such an immoral concept
The authors have some inconsistencies with training token length…
Most errors are probably responses that didn’t finish before their 3K token limit. They’ve measured how well RL is able to shorten the response to their limit.
They do quite a lot of distillation. As we've seen from the American open weight models from AI2 (OLMo series of models). They have a lot of incentive to distill beyond just copying, they're much more compute constrained, so open model companies distill, but also do really good architectural work to make their models run faster. Theres also technical challenges to distillation when all of the top models have their reasoning traces hidden, so we have to assume these open weight labs also have really great training pipelines as well.
A lot of distillation happens. E.g. OLMo models have a completely open dataset and they are heavily distilled. It only makes sense to try to absorb behaviors from the best models out there. That said, I think the open weight juggernaughts are doing really genuinely great work with RL, training environments, architectural innovations etc.
Thanks for the response. i had too many noodles tonight and forgot to check my writing. I’m a rare generalist and so it is so very hard to keep up with this without saying “better autocomplete” my one goal is to not get washed out like my parents did in the great username and password wars.
i used to have this theory about knowledge in society/silos and i likened it to condensation on a window. you have all this water so close to each other and yet not touching-then, something happens and a bead runs down the window and it all connects. i guess distillation reminds me of it but ai overall reminds me of it. because we all know there are silos and complementary info just waiting to run together and make something happen. I am undoubtedly a naive optimist and believe there are good things coming. it’s not a popular opinion and i think that’s mostly because people would rather spend their time guarding than defining their future.
oh baby, there are more noodles in the fridge and to think i almost left them at the restaurant.
There isn't really such a thing as 'too slow' as an objective fact though. It depends on how much patience and money for electricity you have. In AI image gen circles I see people complaining if a model takes more than 5s to generate an image, and other people on very limited hardware who happily wait half an hour per image. It's hard to make a judgement call about what 'too slow' means. It's quite subjective.
If it would take so long to train that the model will be obsolete before the training is finished that might be considered too long. With ML you can definitely hit a point where it is too slow for any practical purpose.
Obsolete because of what? Because with limited hardware you’re never aiming for state of the art, and for fine-tuning, you don’t steer for too long anyway.
That’s just playing semantics. Nobody is talking about, “objective facts” or need define them here. If the step time is measured in days, and your model takes years to train, then it will never get trained to completion on consumer hardware (the entire point).
So distribute copies of the model in RAM to multiple machines, have each machine update different parts of the model weights, and sync updates over the network
This would be my guess too. It can probably be generated synthetically or via agentic rollouts, but high quality long context examples where outputs meaningfully depend on long-range interactions probably remain scarce
There’s a section of I-15 in Utah’s Salt Lake County which reliably has a crash on weekdays at 6pm. It was unfortunately at a pinch point in the mountains with no good alternate route… very annoying.
In a similar way that Google Maps shows eco routes, it’d be fun for them to show “safest” routes which avoid areas with common crashes. (Not always possible, but valuable knowledge when it is.)
reply