>>109782053
https://archive.is/H6j2W (New York Times)

>>109782053
Who are they stealing this other one from?

>>109782053
Luddites are getting absolutely buckbroken, AI is the future.

Rumor is that it is the Hodge Conjecture (and that it is not substantial progress, but fully solved and is only awaiting verification). Also less substantiated rumors that significant progress had been made on Birch-Swinnerton-Dyer.

>>109782066
It's such a cruel faith, that the same year AI begins to be good enough to solve major math problems, is the exact same year actual human mathematicians solve ALL Millennium problems.

>>109782111
This is impossible, the Hodge conjecture is not something that you can just bruteforce and solve.

>>109782119
Rly makes you wonder how much stuff mathfags had lying around for ages, but didn't go to the effort of publishing until someone started tryna take it from them

>>109782053
>Altman wasn’t trolling
Superconductorbros…are we next?

>>109782119
whata coincidence
and everytime humans will be somehow first

>>109782119
Reminds me of a related Tao tweet (or whatever you call Mastodon posts:
>By sheer coincidence, another completely independent result on the Euler blowup question has just been released by Ganeshram, Duruisseaux, and Anandkumar
Yes, sheer coincidence, what are the odds that all these breakthroughs are happening about 3-6 months after AI became good enough to accelerate frontier math research by orders of magnitude? lmao

>>109782053
TOTAL SNAILCAT DEATH

>>109782053
They didn't share the last results thoughtfully why wouldn't they just release it what they have they obviously don't give a shit.

>the newest "AI stole a human's work" narrative wont even last a full week
How do you cope if you're anti-AI?

>>109782300
Look if anti ai people are wrong what do we lose? Literally nothing you won't get ahead of anti ai people since the skill ceiling on using AI is nothing.

>>109782111
Can someone explain what’s the Hodge Conjecture about?

File: 02E6F206-F1E9-4895-B764-F(...).jpg (1.23 MB, 2708x3464)

1.23 MB JPG

>>109782053
This is another vibeGOD victory

>>109782053
I wish you and these LLM PR scams wree over. Plagaising some old papers is not 'solving' anything. Its stealinng vredit for the work of others

>>109782053
These LLM PR scams are so shallow and corny I wonder if sam altman thought of it.

>>109782111
This exact post was made word for word within minutes on a reddit thread here reddit.com/r/mathematics/comments/1wcn9xy/openai_claims_to_have_made_substantial_progress/
The whole internet is botted as fuck, discussion is worthless.

>>109782085
/thread

>>109782300
>okay so it solved all the Millennium problems, but basically humans did 99% of the work inventing math over the past 4000 years, AI just did the last stretch of work.
>we totally could have solved those million dollar problems if we really tried btw, but unlike those soulless clankers we have a life
>now that AIs have solved all the problems we ever identified for them they can't come up with novel ideas, LLMs will never expand to the stars on their own
>okay yes, Alderson drives and Dyson spheres, very impressive, but ALDERSON and DYSON are the names of the humans who came up with those ideas even if they didn't quite figure out the details
>lmao, what the fuck is a "Recursive Hyperboloid Matter Nanocompactor", this dumb autocomplete is just spitting out technoba-

>>109782388
Go back

>>109782053
At this pace there will not be enough Millenium Prize problems for OpenAI's marketing campaign before its IPO

>>109782240
Kek

>>109782558
That's OK an LLM will hallucinate a prize for a problem some old paper solved and win the price and if it can;t it can just shit out five pages of math only a handful of people will look at and claim it solved it anyway but china may win the race or it might be AGI or it may escape or it may enslave humanity or something so you betrer give them money

>>109782696
>erm it will just le make up le correct answer to long standing open problem and if it can't it will just le dream up a proof that is correct but only le 5 people can understand it because it's actually stupid and not a long standing problem for any reason
I understand why the AI jeets call you trooners luddites now. What's next, will it just make up the correct code to replace you at your devshitter job? Oh wait...

>>109782696
Schizophrenic

>>109782558
fuck off to reddit and never return newfag

>>109782053
How funny would it be if openAI was just tricking anthropic into wasting millions of dollars in compute? Being so vague about which exact problem was partially solved also makes them waste 5x as much compute.

Either way this could be bad news also because it's probably a boring counterexample. It could be some self-referential bullshit that makes P≠NP

They should add collatz to the problems, why cant ia improve that

>>109782053
Why are you anons so impressed by the millennial prize? You do understand that what GPT was try absurd conditions with mathematical functions to find that it might not work with all cases right? It's not that impressive if you're not a mathematician or physicist. It's useful for those guys but for us it's useless. Just because it's called millennial prize it doesn't mean it's the biggest achievement of the millenium.

File: Trust Us.jpg (3.62 MB, 2363x3150)

3.62 MB JPG

>>109782388
Of course it is, if they try to drive the narrative on HN and Reddit, why wouldn't they try here?

>>109782993
>Just because it's called millennial prize it doesn't mean it's the biggest achievement of the millenium
It is though

>>109783022
You know how big a millenium is? We're just starting. Don't be ridiculous.

AI is solving millennium problems.

Meanwhile, I give Opus 5 a PDF of a homes floorplans with dimensions and ask it to calculate the sqft of the kitchen, and it spins its wheels and falls on its face (it tried flood filling pixels)

ok but does this one literally affect anything at all or is it yet another useless mental masturbatory like the previous one?

>>109783135
Useless

>>109783135
We don’t know which one is it

>>109783153
We do know, inbred

>>109783135
>>109783153
if it any commercial value whatsoever they wouldn't just give it out for free for anyone to see

>>109783165
Nope

>>109783170
>>>/x/

>>109783193
Oh, so you don’t know?

>>109782959
Collats already has a very close result. Even though it's miles away from actually proving it.

Assuming collatz is true, solving it probably involves inventing a lot of novel concepts and ideas. Not just new theorems in a field like number theory, but truly a new theory for dealing with systems like collats. Looking at the navier stokes solution, it seems unlikely AI is able to come up with something like that as of yet, though i might be mistaken.
Assuming collatz is false, i find it hard to imagine AI could come up with a counterexample by thinking hard enough about it.

I wouldn't be surprised if collatz turns out to be a lot more difficult to prove (or disprove) than most of the millenium prize problems.

I wish they would concentrate some of their computing power on collatz. I'm sure they will eventually do so, since its rather famous, but i highly doubt we will see a result soon.

>>109783104
Did you try spending hundreds of thousands of dollars on a thinking model and letting it spin for days? You should try that. Take out a loan if you have to. Trust Altman.

>>109783332
>Assuming collatz is false, i find it hard to imagine AI could come up with a counterexample by thinking hard enough about it
Why?