ResearchNovember 10, 2026
27
🌶️
AI Cracks 18-Year Quantum Puzzle
The startup PSI bets on specialized model harnesses after AI cracked an 18-year-old quantum problem.
#AI for Science#Quantum Computing#Anthropic#Lean#Research Agents

🔥 What happened
A recent MIT grad used Anthropic's Fable 5.2 to solve the 18-year-old Aaronson-Ambainis conjecture on quantum speedups. The proof was autoformalized in Lean and vetted by top researchers.
💡 Why it matters
The result closes one avenue for quantum computers to beat classical ones — a rare case of AI answering a deep theoretical CS question, not just crunching numbers. Scott Aaronson admits he doesn't fully grasp the proof but trusts it's correct.
⚡ Our take
When a startup with specialized model harnesses cracks a decades-old problem that OpenAI apparently didn't even attempt, the power axis in AI research shifts away from the big labs.
The title, summary and analysis of this item were produced automatically by an AI system and have not been editorially reviewed. They may contain errors, bias or omissions — when in doubt, read the linked original source.
Deep Dives & Similar Intelligence
Anthropic opens Claude Skills to all
Tools & ProjectsNovember 09, 2026

Axiom Formalizes 7,500 Pages of Algebra
ResearchNovember 05, 2026
Claude Code Leaks Your Email
Ethics & SecurityNovember 05, 2026

Anthropic's AI Is Speeding Up AI
ResearchNovember 09, 2026

Anthropic's AI is accelerating itself 10x
ResearchNovember 06, 2026