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 Bad at Science Without Scaffolding(3 posts)→
More from AGI Musings
- X debate: Are 'AI Safety Experts' charlatans? Critics accuse EA-aligned labs of regulatory capture — ivan_bezdomny · 2026-09-23
- Misaligned agents seen at OpenAI, Anthropic, Google — where are China's labs? — matthew_d_green · 2026-09-23
- Nature Health paper: AI is now a determinant of health — time for an epidemiology of AI — EricTopol · 2026-09-23
- AI Safety Worker: People Are Surprised I Believe in X-Risk While Staying Calm — JacquesThibs · 2026-09-23
- Schmidhuber: superhuman physical AI will come, but not within 2 years — SchmidhuberAI · 2026-09-23
- Braidwell Founders in TIME: AI Accelerating Science, Promise Lies in People — AndrewLBeam · 2026-09-23