● ResearchOct 30, 2026 · 20:538 sources
Anthropic and OpenAI Independently Disprove Exponential Time Hypothesis
From marketbrief.now
AI;DR
Anthropic and OpenAI independently reported that their internal models debunked the Exponential Time Hypothesis, with Anthropic's Claude research model finding a subexponential 3-SAT solution on Oct. 16 and OpenAI reaching the same conclusion shortly before.
Read originalSign in to suggest
Sources
+ 6 more
- SUPPORT@choblin29“Anthropic has now formalized the result across 1,050 Lean modules, checked by both Lean and the independent nanoda kernel.”x.com
- SUPPORT@andrewcurran_“Anthropic researchers worked with Claude to improve the presentation, resulting in this paper.”x.com
- SUPPORT@kimmonismus“An internal Anthropic model also refutes the ETH. With only 14m Output tokens and already on October 16th.”x.com
- SUPPORT@scaling01“14M output tokens is like 3-4 days with a single model.”x.com
- SUPPORT@andrewcurran_“You don't know the half of it yet:”x.com
- SUPPORT@teortaxestex“Levent has materially damaged my opinion of Anthropic *and* OpenAI”x.com