AI Solves Notorious Math Problem — 25 Fields Medalists Fire Back With Condemnation

From our other sites

The story

After US-based OpenAI announced it had solved the unsolved math problem known as the Navier-Stokes equations, 25 Fields Medal laureates released an emergency statement on September 11 condemning the AI-driven solving of hard math problems. The statement reportedly plans to gather more signatures from mathematicians going forward. On 5channel, opinions split: some voiced fears that being outdone by AI could erode mathematicians’ sense of purpose and cost them their jobs, others dismissed AI as merely repackaging ideas humans came up with, while still others took the pragmatic view that the process doesn’t matter as long as results are delivered.

On the 11th, 25 mathematicians who hold the Fields Medal — often called the “Nobel Prize of mathematics” — released an emergency statement condemning efforts by companies to use artificial intelligence (AI) to solve difficult math problems. They say they plan to collect signatures from even more mathematicians going forward.

The statement appears to be a response to US-based OpenAI’s announcement that it had solved a hard math problem known as the Navier-Stokes equations. The Navier-Stokes equations were named one of the Millennium Prize Problems by the US-based Clay Mathematics Institute in 2000, carrying a $1 million prize…

Source: nikkei.com / Read the original article here

What people said

2AnonymousSep 12, 2026 18:20
Wait, it wasn't even the Riemann Hypothesis?
3AnonymousSep 12, 2026 18:21
A future where mathematicians get sacked
4AnonymousSep 12, 2026 18:24
Illustrators (lol)
Programmers (lol)
Mathematicians (lol)
5AnonymousSep 12, 2026 18:34
That statement won't do a damn thing.
Once data centers have spare capacity to burn, plowing through unsolved math problems one by one is exactly the first thing they'll use it for.
6AnonymousSep 12, 2026 18:36
I guess the edge you get from being smart just vanishes, huh.
So it comes down to "please don't use AI, for the sake of our advantage," doesn't it.
7AnonymousSep 12, 2026 18:37
Re: #1
If mere AI can solve it, their whole reason for existing disappears — so is this really just about being scared of losing their jobs?
Honestly, nobody in the general public cares who solves a math problem — we just want to use the results.
21AnonymousSep 12, 2026 19:59
Re: #7
No, that's not it.
It's about what a kid actually learns once they start having AI do their homework for them.
These people aren't banning the use of AI to solve problems.
8AnonymousSep 12, 2026 18:43
If mere AI can solve it, why couldn't anyone solve it until now?
9AnonymousSep 12, 2026 18:52
Well, it's because their jobs are on the line, lol
Futile resistance.
10AnonymousSep 12, 2026 18:52
I mean, who cares as long as it gets solved? Are they saying the process matters more than the answer?
12AnonymousSep 12, 2026 18:56
It's a lost cause already.
13AnonymousSep 12, 2026 19:12
The real question is whether AI can keep it up on its own once the scholars give up on it.
14AnonymousSep 12, 2026 19:13
Where's the fun in having a machine solve the puzzle for you?
15AnonymousSep 12, 2026 19:13
Huh?
It's useful, isn't it.
16AnonymousSep 12, 2026 19:13
AI is just ripping off ideas humans came up with to spit out answers.
People say that if AI keeps getting free rein like this, mathematicians will eventually stop publishing their ideas altogether.
17AnonymousSep 12, 2026 19:25
Who knew the day would come when a von Neumann-caliber mind becomes worthless.
39AnonymousSep 13, 2026 01:16
Re: #17
In principle a von Neumann-type mind is still faster and more power-efficient.
Catching up is still ahead of us.
18AnonymousSep 12, 2026 19:31
Humanity's intelligence declining as a result — that's a given at this point.
19AnonymousSep 12, 2026 19:40
After milking the benefits of science and technology for all it's worth, this complaint is pretty rich, lol
20AnonymousSep 12, 2026 19:55
Because this doesn't just waste the effort of fake scholars — it wastes the effort of legitimate, hard-working ones too.
27AnonymousSep 12, 2026 22:06
Re: #20
Yeah, fair point.
If someone out there is thinking "I'm so close to solving it!"
right before AI beats them to it, that'd be hell.
33AnonymousSep 13, 2026 00:20
Re: #20
Guess all that effort was wasted, lol
Next time, learn to make AI work for you instead.
22AnonymousSep 12, 2026 20:05
It's getting closer and closer…
the era of Detroit: Become Human.
24AnonymousSep 12, 2026 20:50
Blatant self-preservation.
26AnonymousSep 12, 2026 21:00
Isn't the bigger problem that these hard problems never get solved any other way, lol
28AnonymousSep 12, 2026 22:17
Solving a hard math problem doesn't actually benefit anyone anyway.
I think it's fine to just leave that kind of thing as a pastime for humans to enjoy.
29AnonymousSep 12, 2026 22:34
Re: #27
No, wasn't it (v·∇)v that stole the idea and cracked the solution?
30AnonymousSep 12, 2026 22:57
More to the point, half the Japan Meteorological Agency's staff and every weather forecaster out there are unnecessary.
AI would nail it more accurately anyway.
34AnonymousSep 13, 2026 00:53
Re: #30
They've had the media loudly repeating "linear rain band" (a Japanese term for a stalled heavy-rain storm cluster) over and over,
even having them warn of a "risk of a linear rain band forming" to stoke fear,
yet they can't actually predict any of it —
it's all explained after the fact.

So they hide behind vague "risk of forming" language to cover up how poor their forecasting really is,
and coining the term "linear rain band" in the first place was just another layer of that cover-up.
And now, on top of it all,
they've started saying "squall line" instead —
which is actually the more internationally standard term, yet they never used it until now.

It's nothing but the Japan Meteorological Agency putting on a show so they can keep the supercomputer budget flowing their way.

And the media just falls in line without a shred of criticism.
31AnonymousSep 12, 2026 23:19
In a brute-force, all-comers approach, you'd still see gaps show up in speed, accuracy, fatigue, and so on.
32AnonymousSep 13, 2026 00:18
Only someone who solves these hard problems with nothing but pencil and paper — no computer, no calculator — has any right to complain.
35AnonymousSep 13, 2026 00:54
Not just math — anything involving formulas is basically already like conveyor-belt sushi at this point (a jab at how sushi restaurants already use robots — meaning the automation ship sailed long ago).
36AnonymousSep 13, 2026 01:02
Well, it must sting to have it solved right out from under you.
Some people probably poured their entire lives into it to the point of losing their minds.
37AnonymousSep 13, 2026 01:06
Re: #1
The real problem here is something else entirely:
AI doesn't actually understand the underlying theory —
it's just generating text in whatever way suits it,
and it's humans who have to vouch for whether the resulting output is even correct.
Even the mistakes get "corrected" based on edits from people who already back the AI's output.
Bottom line, it's no different from Yutabon (a controversial Japanese child YouTuber and school dropout) claiming he proved a math formula.
38AnonymousSep 13, 2026 01:10
Tasugisanmyaku (the thread's poster/OP) seems to have a taste for junk news that's already slightly past its expiration date —
a lot of the AI-related stuff he posts is months behind,
and in the worst cases, about a year behind.
If this is who's running the thread, YouTube's coverage is way fresher.
40AnonymousSep 13, 2026 01:19
No matter how capable AI gets, it still can't get me a raise.
41AnonymousSep 13, 2026 01:25
Don't you know the history of how, when the steam locomotive was invented, the horses got so furious they attacked people?

Background and key points of this debate

The Navier-Stokes equations are one of the seven Millennium Prize Problems named by the US Clay Mathematics Institute in 2000, each carrying a $1 million prize at the time. Of those seven, only the Poincaré Conjecture has been certified as solved so far, and the rest have stumped mathematicians for decades. The Fields Medal, awarded once every four years to mathematicians under 40, is often called the “Nobel Prize of mathematics” — which makes it unusually significant that 25 of its laureates signed a joint statement at all. The thread’s debate splits largely into two camps talking past each other: anxiety that being solved by AI undermines mathematicians’ profession and intellectual edge (comments 6, 17, etc.), versus a pragmatic view that the means don’t matter as long as results are delivered (comments 10, 15, etc.). As comment 21 points out, the statement doesn’t actually call for banning AI use outright — it also raises an educational point about what humans are supposed to learn along the way, a fact worth noting apart from the emotional backlash.

*This article is excerpted and summarized from the 5channel (Business News+) thread “[AI] Math Problem Solved by AI, 25 Fields Medalists Issue Statement Condemning It as “Harmful to Science”.”

Comments

Leave a Reply

Your email address will not be published. Required fields are marked *