Physicist: AI now beats most junior theorists, and researchers should ride the dragon
IgorCarron · x · 2026-11-17
Physicist Antoine Tilloy's essay "The short age of dragon riders" argues researchers should ride the AI dragon while they still can.
- Since August, public AI models in a proper harness outperform most theoretical physicists (himself included) on most technical tasks: complex analytical calculations, lengthy proofs, turning problems into code, iterating, simulating, summarizing.
- He has repeatedly seen "the impossible" done on his own problems, starting with Fable 5 in May; deniers haven't tried hard enough or are blinded by ego.
- Limits: best public models still lack research taste — they execute beautifully but fall short of top researchers at asking good questions and sparking creativity out of dead ends.
- Trend: human data is no longer the bottleneck; RLVR lets models improve by solving new tasks, unbound by human intelligence. OpenAI and Anthropic internal models have reportedly solved countless open problems, a generation ahead of public ones.
Igor Carron's quote-reshare adds a terse "And they fall..."
More from AGI Musings
- a16z Introduces the B2A Economy: Agents Are Now the Internet's Primary Users and Buyers — venturetwins · 2026-11-17
- Ben Bajarin: enterprise AI adoption is the only conviction that matters for the decade — BenBajarin · 2026-11-17
- LangChain founder: agents are shifting from executing tasks to owning goals — hwchase17 · 2026-11-17
- Paul Christiano: 1-in-3 chance of government-toppling superintelligence within a year — Dwarkesh Patel · 2026-11-17
- Sam Altman on stage: "We didn't have AGI. Now we do" — and OpenAI is withholding self-improving AI — imjustnewatai · 2026-11-17
- Vik Paruchuri: Even Experts Struggle to Tell Good Benchmarks from Bad Ones — VikParuchuri · 2026-11-17