Es geht um das Navier-Stokes-Problem.
November 3, 2026 at 12:11 PM
Es geht um das Navier-Stokes-Problem.
What OpenAI left out of its Navier–Stokes announcement…
November 3, 2026 at 2:14 PM
What OpenAI left out of its Navier–Stokes announcement…
In September, OpenAI claimed that an AI model solved the million-dollar Navier-Stokes equation. Weeks later, OpenAI shared results for more than 350 previously unsolved math problems. The news has prompted a conversation among mathematicians: In an age of AI, what’s the role of human mathematicians?
November 1, 2026 at 10:20 PM
In September, OpenAI claimed that an AI model solved the million-dollar Navier-Stokes equation. Weeks later, OpenAI shared results for more than 350 previously unsolved math problems. The news has prompted a conversation among mathematicians: In an age of AI, what’s the role of human mathematicians?
Für das Navier-Stokes-Problem hat OpenAI angeblich 130 Milliarden Token verbraucht.
November 3, 2026 at 5:35 PM
Für das Navier-Stokes-Problem hat OpenAI angeblich 130 Milliarden Token verbraucht.
on en est où sur l'analyse de la preuve de Navier Stokes par OpenAI ?
November 1, 2026 at 12:40 PM
on en est où sur l'analyse de la preuve de Navier Stokes par OpenAI ?
10,000 AI agents worked together to solve the Navier Stokes equations
I wonder... did they set up governance in there? Did they have votes? Were there factions where they disagreed on the direction to take the proof? Did any love triangles form?
Reality is probably more boring though.
I wonder... did they set up governance in there? Did they have votes? Were there factions where they disagreed on the direction to take the proof? Did any love triangles form?
Reality is probably more boring though.
November 1, 2026 at 9:37 AM
10,000 AI agents worked together to solve the Navier Stokes equations
I wonder... did they set up governance in there? Did they have votes? Were there factions where they disagreed on the direction to take the proof? Did any love triangles form?
Reality is probably more boring though.
I wonder... did they set up governance in there? Did they have votes? Were there factions where they disagreed on the direction to take the proof? Did any love triangles form?
Reality is probably more boring though.
Did Anthropic value itself at $1.2tn in August? ft.trib.al/gwf03Bg
Did Anthropic value itself at $1.2tn in August?
Not quite Navier-Stokes
ft.trib.al
October 30, 2026 at 2:07 PM
Did Anthropic value itself at $1.2tn in August? ft.trib.al/gwf03Bg
OpenAI has a better model than the one that solved Navier Stokes
October 30, 2026 at 12:04 PM
OpenAI has a better model than the one that solved Navier Stokes
Read it now: community.wolfram.com/t/28013
#AI #physics #algebra #programming #coding #WolframLanguage
#AI #physics #algebra #programming #coding #WolframLanguage
An intuitive exploration of OpenAI's Navier–Stokes singularity construction (for the $1M Millennium Prize Problem)
Open in Wolfram Cloud ↗ DOWNLOAD-DESKTOP-NOTEBOOK.nb (2.5 MB)
community.wolfram.com
November 3, 2026 at 6:55 PM
Ngl this is kind of why I tend to be skeptical of AI economic claims. Undeniable that it’s having massive effects but like
In California you can build a machine that solves the navier stokes smoothness and existence problem but not a 3 story apartment building
In California you can build a machine that solves the navier stokes smoothness and existence problem but not a 3 story apartment building
October 31, 2026 at 12:24 AM
Ngl this is kind of why I tend to be skeptical of AI economic claims. Undeniable that it’s having massive effects but like
In California you can build a machine that solves the navier stokes smoothness and existence problem but not a 3 story apartment building
In California you can build a machine that solves the navier stokes smoothness and existence problem but not a 3 story apartment building
this really is my intuition as well. But I also didn't expect language models to find a solution to the Navier-Stokes Millennium Prize problem.
October 30, 2026 at 8:15 PM
this really is my intuition as well. But I also didn't expect language models to find a solution to the Navier-Stokes Millennium Prize problem.
As puzzles go it's kind of easy. It's hardly the Navier-Stokes Millennium Prize problem. What's hard about it is how tedious it is. But if you can weather the tediousness (or you have a computer to do the tedious parts for you) and you have a basic understanding of neural networks, it's not so hard.
November 4, 2026 at 7:46 PM
As puzzles go it's kind of easy. It's hardly the Navier-Stokes Millennium Prize problem. What's hard about it is how tedious it is. But if you can weather the tediousness (or you have a computer to do the tedious parts for you) and you have a basic understanding of neural networks, it's not so hard.
I’m glad AI has solved the Navier-Stokes Millennium Problem before any human mathematicians, but we’re out here mistaking o for 0 at the same time.
November 3, 2026 at 7:36 PM
I’m glad AI has solved the Navier-Stokes Millennium Problem before any human mathematicians, but we’re out here mistaking o for 0 at the same time.
Did you see the preprint about how OpenAI has a major Gödel problem?
And frankly given the issues with alignment with the current generation of models, it’s nuts to think they can set reasonable constraints (ie. Cognitive control) to any problem at all?
arxiv.org/abs/2610.08144
And frankly given the issues with alignment with the current generation of models, it’s nuts to think they can set reasonable constraints (ie. Cognitive control) to any problem at all?
arxiv.org/abs/2610.08144
Navier-Stokes lost in translation: Why Lean verification of AI autoformalisation does not guarantee correct natural language proofs
Autoformalisation is increasingly used to verify mathematical texts, including those generated by AI, as in OpenAI's announced proof of blow-up of solutions to the Navier-Stokes equations. In this pro...
arxiv.org
November 2, 2026 at 4:15 AM
Did you see the preprint about how OpenAI has a major Gödel problem?
And frankly given the issues with alignment with the current generation of models, it’s nuts to think they can set reasonable constraints (ie. Cognitive control) to any problem at all?
arxiv.org/abs/2610.08144
And frankly given the issues with alignment with the current generation of models, it’s nuts to think they can set reasonable constraints (ie. Cognitive control) to any problem at all?
arxiv.org/abs/2610.08144
' Part of the reason is
that nobody yet understands the answer. '
#Navier–Stokes equations
#science #problems #ai
#OpenAi #math #publicUnderstanding
#sciencecommunications
//cc @marcoderksen.eurosky.social
that nobody yet understands the answer. '
#Navier–Stokes equations
#science #problems #ai
#OpenAi #math #publicUnderstanding
#sciencecommunications
//cc @marcoderksen.eurosky.social
Introducing our new columnist, professor Hannah Fry. Read on for Hannah's take on the "existential crisis" engulfing maths, and come back every fortnight for her latest column.
Hannah Fry: How OpenAI has caused an existential crisis in mathematics
www.newscientist.com
October 31, 2026 at 5:34 AM
' Part of the reason is
that nobody yet understands the answer. '
#Navier–Stokes equations
#science #problems #ai
#OpenAi #math #publicUnderstanding
#sciencecommunications
//cc @marcoderksen.eurosky.social
that nobody yet understands the answer. '
#Navier–Stokes equations
#science #problems #ai
#OpenAi #math #publicUnderstanding
#sciencecommunications
//cc @marcoderksen.eurosky.social
My favorite mathematician described the navier-stokes openai "proof" by the very term it failed to resolve "a convoluted f"
November 1, 2026 at 7:48 AM
My favorite mathematician described the navier-stokes openai "proof" by the very term it failed to resolve "a convoluted f"
The particular Navier-Stokes question that was a Millennium prize problem wasn’t included because anybody particularly wanted the answer, it was based on the assumption that new math would be required to answer it.
November 3, 2026 at 5:30 PM
The particular Navier-Stokes question that was a Millennium prize problem wasn’t included because anybody particularly wanted the answer, it was based on the assumption that new math would be required to answer it.
🚨 In which @tim.fo and I do some maths to try to establish whether Anthropic told us they reckon they're worth $1.2tn.
www.ft.com/content/60fb...
www.ft.com/content/60fb...
Did Anthropic value itself at $1.2tn in August?
Not quite Navier-Stokes
www.ft.com
October 30, 2026 at 12:53 PM
🚨 In which @tim.fo and I do some maths to try to establish whether Anthropic told us they reckon they're worth $1.2tn.
www.ft.com/content/60fb...
www.ft.com/content/60fb...
Josef Demmel: 2D Navier-Stokes with viscosity-dependent Navier slip: strong convergence for unbounded initial velocity https://arxiv.org/abs/2611.05212 https://arxiv.org/pdf/2611.05212 https://arxiv.org/html/2611.05212
November 4, 2026 at 6:47 AM
Josef Demmel: 2D Navier-Stokes with viscosity-dependent Navier slip: strong convergence for unbounded initial velocity https://arxiv.org/abs/2611.05212 https://arxiv.org/pdf/2611.05212 https://arxiv.org/html/2611.05212
Not necessarily. Reasoning models can give “smarter” answers just by reasoning for longer, and the 10,000 agent swarm that made the Navier Stokes proof was essentially that on steroids. A potential RSI path could be continued efficiency improvement enabling substantially larger swarms
November 2, 2026 at 8:36 AM
Not necessarily. Reasoning models can give “smarter” answers just by reasoning for longer, and the 10,000 agent swarm that made the Navier Stokes proof was essentially that on steroids. A potential RSI path could be continued efficiency improvement enabling substantially larger swarms
OpenAI asegura haber resuelto el problema de Navier-Stokes encontrando una singularidad, pero la demostración usó una fuerza externa que deja abierta la gran pregunta de la física.
La IA resolvió Navier Stokes y no lo resolvió
muyinteresante.okdiario.com
November 2, 2026 at 4:28 PM
OpenAI asegura haber resuelto el problema de Navier-Stokes encontrando una singularidad, pero la demostración usó una fuerza externa que deja abierta la gran pregunta de la física.
Either Yen and his AI addicted CSuite are out or I'm out from everything Proton. The Navier-Stokes code theft already shows Claude is insecure to code with, and there's no reason to believe any other big name is safe to use.
The rest of the engineers sound like they understand reality.
The rest of the engineers sound like they understand reality.
October 29, 2026 at 8:10 PM
Either Yen and his AI addicted CSuite are out or I'm out from everything Proton. The Navier-Stokes code theft already shows Claude is insecure to code with, and there's no reason to believe any other big name is safe to use.
The rest of the engineers sound like they understand reality.
The rest of the engineers sound like they understand reality.
After the way Open AI "borrowed" information from mathematicians work in GPT to solve navier stokes, I think Goldman might be worried Anthropic may borrow their proprietary trading formulas and techniques and spread them to other users.
November 2, 2026 at 5:52 PM
After the way Open AI "borrowed" information from mathematicians work in GPT to solve navier stokes, I think Goldman might be worried Anthropic may borrow their proprietary trading formulas and techniques and spread them to other users.