Summary
According to a post on X by @tanayj, Sam Altman said on the Cheeky Pint podcast with @collision that the compute behind OpenAI's Navier-Stokes proof cost on the order of $10 million. He said that with better models and agent swarms, the work could be redone from scratch for about 1/100th of that within a year. In a separate clip, Altman said what he wants from AI agents is simple: they should just work. The post's snippet is partial, so the full context of these remarks is not shown here.
AI-assisted summary based on the listed source.
What happened
On the Cheeky Pint podcast with @collision, Sam Altman said the compute behind OpenAI's Navier-Stokes proof cost on the order of $10 million. He said that with better models and agent swarms it could be redone from scratch for about 1/100th of that within a year. In a separate clip, he says what he wants from AI...
Why it matters
The cost figure is an estimate from OpenAI's CEO, not a published cost breakdown in the source material. For builders and businesses weighing agent-based workflows, the cost trend Altman describes is the number to track. Watch for independent figures or an actual redo of the work that would test the one-year, 1/100th claim.
Signal Intelligence
Signal Strength 88%
Technical label WATCH
Public Interest 42
Category BIG MOVE
Reader Depth TECHNICAL
Signal Strength reflects source quality, relevance, freshness and evidence. Public Interest helps organize discovery; it is not proof of truth.
Public Interest components
Recognizable Entity Score 73
Practical Impact Score 0
Novelty Interest Score 70
Consequence Score 18
Curiosity Score 16
Shareability Score 57
Why this is here
VQV surfaced this signal because it is recent, relevant to AI Agents, connected to Scobleizer News Pulse (X).
