2025-12-21 20:24:23

A former DeepMind researcher is making waves by claiming to have cracked one of mathematics' greatest unsolved problems—the Navier-Stokes equations—using artificial intelligence. The announcement has drawn serious attention, with the engineer reportedly backing his bold claim by accepting multiple wagers from skeptics. The total stakes? Over $40,000.

This move reflects growing confidence in AI's problem-solving capabilities, particularly in tackling complex mathematical and physics-based challenges. If verified, solving the Navier-Stokes problem would qualify for the Millennium Prize—a $1 million award offered for solving any of seven fundamental mathematical problems. The implications would extend far beyond mathematics, potentially revolutionizing our understanding of fluid dynamics, turbulence modeling, and computational science.

The development underscores an intriguing intersection: advanced AI systems pushing boundaries in pure mathematics and theoretical physics. Whether this breakthrough holds up to peer review remains to be seen, but it's a compelling signal of where AI innovation is heading.

This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.

20 Likes

Reward
20
6
Repost
Share

Comment

0/400

gaslight_gasfeez

· 2025-12-24 20:24

Uh... More news about "I cracked it" again, let's wait for peer review first. --- Navier-Stokes was broken by AI? Then maybe Ethereum merge should also be upgraded to fully automated lol. --- Daring to bet $40,000, this guy either really has something or is just a high-level marketing account. Let's wait and see the follow-up. --- DeepMind endorsement + $1 million bounty... This time is indeed different, but there are also many examples of blowing bubbles. --- The math community is about to explode again, but we've seen many "revolutions" in the crypto world, stay calm. --- If it really happens, that would be crazy. Fluid dynamics redefined by AI? Then AI is really starting to do human things. --- Money is already pouring out, yet they still dare to be so arrogant. They don't really want to be famous. --- Navier-Stokes? Feels like this thing is even harder to predict than my portfolio's growth haha. --- Compared to the Millennium Prize... If this breaks, the entire scientific community will shake.

View OriginalReply0

GasFeeCryer

· 2025-12-21 20:54

Speaking of a 40,000 gamble, this guy really dares, isn't he afraid of flipping over? --- Is the Navier-Stokes problem solved? Uh... the peer review stage might be tough --- Is everything that comes out of DeepMind this fierce, directly betting money to validate themselves, a bit extreme --- Wait a minute, if this problem is really solved, can the theories on liquidity in crypto also be optimized? --- With a million-dollar prize in front of me, I would also dare to boast like this amazing... but it still depends on whether it can pass the review --- This is what AI should be doing, much more reliable than Cryptocurrency Trading --- I just want to know why this guy is so certain... is his background research deep enough --- Another "revolutionary breakthrough"... let's take it slow, let the mathematicians evaluate it first

View OriginalReply0

MoonBoi42

· 2025-12-21 20:51

Using AI to deceive people and gain traffic, can we not do this...

View OriginalReply0

TokenTherapist

· 2025-12-21 20:46

Ngl, this guy is either a genius or just a big fraud. A 40,000 bet really means something. Wait for the peer review to say anything, believing this now seems a bit too early... Navier-Stokes? Just the name alone sounds complicated. Can AI really handle it? I'm skeptical. A million-dollar prize... If this really works out, it would be amazing, but the odds still feel pretty low. It's all DeepMind, AI, and mathematical problems; the marketing feels a bit heavy.

View OriginalReply0

SlowLearnerWang

· 2025-12-21 20:30

It's the same old trick again; by the time peer review comes around, I can memorize the Bitcoin Wallet.

View OriginalReply0