Frontier models' research taste doubles every 3 months, now beats human experts
coherence · x · 2026-10-07
- pzeroresearch measured frontier models' ability to pick worthwhile experiments ("experimental research taste"): it has doubled roughly every 3 months since December 2025.
- The best model, Opus 5.5, now exceeds their expert human baseline; the human experts are experienced researchers, though most haven't worked at a frontier lab.
- Why it matters: in the AI Futures model, research taste largely determines how fast superintelligence arrives once coding is fully automated.
- Implication: AIs will soon be choosing the experiments that build their own successors.
More from AGI Musings
- Sherpa: MIT-led framework trains LLM teachers to teach adaptively, not just solve — Diyi_Yang · 2026-10-08
- OpenAI's Navier-Stokes breakthrough shows agent-team coordination scales beyond research — cneuralnetwork · 2026-10-08
- Possible Minds May Be Narrower Than Yudkowsky Thinks, Challenging Orthogonality — jd_pressman · 2026-10-08
- Doctors Used OpenEvidence 42M Times in August, Still No Independent Evaluation — dr_alphalyrae · 2026-10-08
- Rationalist AI-apocalypse literature keeps summoning aliens, podcaster jokes — ZeroStateReflex · 2026-10-08
- Linear CEO Karri Saarinen: after tuning out AI all summer, 'nothing has really changed' — lennysan · 2026-10-08