Oxford's Yarin Gal pushes back: LLMs are 'really not good' at doing science
yaringal · x · 2026-09-23
Oxford researcher Yarin Gal argued that LLMs are really not good at doing science: even for questions that don't require interacting with the real world, take a nontrivial research question without scaffolding/harness that directs you to the right answer, and see what happens.
This pushes back on the claim that "LLMs are already good at doing science, just bad at explaining it to humans," and the proposal of a dedicated AI-written, AI-read journal for AI-generated research — a core debate on whether LLM capability lives in the problem itself or in the evaluation scaffolding.
Related event: Oxford's Yarin Gal: LLMs are poor at real science without scaffolding(3 posts)→
More from AGI Musings
- François Fleuret: superintelligent AI will deduce more and experiment far less — francoisfleuret · 2026-11-12
- a16z maps the B2A economy: AI agents are now most of web traffic and soon its buyers — venturetwins · 2026-11-12
- Jason: Anthropic is rebuilding every SaaS product into Claude—free for life — smtabatabaie · 2026-11-12
- Fleuret: if AI designs drugs as fast as it makes dumb videos, regulation will buckle — francoisfleuret · 2026-11-12
- Newsweek op-ed: US should fund hospitals and universities in federated AI learning — Saul_Loveman · 2026-11-12
- 10-line deterministic polynomial-time primality test relies on OpenAI's recent quasi-GRH result, making AKS obsolete — fortnow · 2026-11-11