Two models of progress: AI excels at checklists, but science is open-ended
Dr_Atoosa · x · 2026-11-21
DrAtoosa distinguishes two models of progress:
- Checklist model: progress means incrementally solving a fixed list of problems humans designated as "major" (Millennium Prize Problems, formalization benchmarks). AI is rapidly clearing these in verifiable domains — and if that's your only view of progress, it breeds dread.
- Generative encounter model: science and math are open-ended; real progress comes from encountering messy real-world problems on no list, using judgment and taste to decide what matters, abstracting reality into new formal structures, and opening new domains where every solution generates fresh questions.
Related event: Scientist Proposes Two Models of Progress to Explain AI Anxiety(3 posts)→
More from AGI Musings
- Stanford Study: Deactivating AI for Two Weeks Reduces Loneliness and Boosts Social Interaction — stanfordnlp · 2026-11-21
- 'Just Gonna Try 57M Things': The Agent Brute-Force Meme and the Formal Verification Question — josephdviviano · 2026-11-21
- ML Professor Guestrin: Daughter Built a Game with an AI Agent, Echoing His 1998 First Neural Net — guestrin · 2026-11-21
- Peter Diamandis: The Right Analogy for AI Isn't Flight—It's Fire — PeterDiamandis · 2026-11-21
- Debate on X: most frontier-lab safety roles net negative, but Perez and Nanda are exceptions — austinc3301 · 2026-11-21
- Mathematician predicts superhuman AI math in 2-3 years after LLM disproves unit-distance conjecture — soumitrashukla9 · 2026-11-21