Neuroscientist says GPT-6 cracked a 38-year-old chaos problem, matching real simulations in 100 minutes
CatAstro_Piyush · x · 2026-10-10
A Flatiron Institute theoretical neuroscientist reports his most significant pure-theory breakthrough yet: the Lyapunov spectrum of a 1988 brain-activity model, unsolved for nearly 40 years. GPT-6 Astra worked 100 minutes and matched real computer simulations; he then iterated with the AI until he could explain the reasoning, crediting GPT-6 and Claude in the paper. Commentators call it one of the clearest signs AI can do real science.
Related event: GPT-6 Cracks 38-Year-Old Neuroscience Chaos Problem in 100 Minutes(2 posts)→
More from Research
- Researcher finds many eval harness functions exist only to patch dataset flaws — remilouf · 2026-10-11
- Bo Wang joins SF Tech Week 'Models to Medicine' panel on virtual cells and Xaira — BoWang87 · 2026-10-11
- AI on track to solve 1,000+ math problems in a month, a 100x jump in 2 months — Dr_Singularity · 2026-10-11
- "The product of math is proofs": a case for vibe mathing everything with AI — cgarciae88 · 2026-10-11
- OpenAI drops 372 math proofs, and mathematicians are grieving — genmon · 2026-10-11
- Was Claude's gibberish fixed by capping KL divergence in RL? One theory — burny_tech · 2026-10-11