Stanford professor: LLMs are great brainstorm partners but confidently spew plausible nonsense
anshulkundaje · x · 2026-09-26
Stanford professor Anshul Kundaje weighs in on using LLMs for research brainstorming: they're genuinely useful as cross-domain brainstorm partners, but far from reliable — models confidently produce legitimate-sounding nonsense that doesn't match their actual accuracy.
A commenter adds that having a partner who can even pretend to follow you across many fields is a great boon, with a known rigor penalty at the end — which applies to human experts too. The thread also suggests this gap may be a natural safeguard against misuse: non-experts struggle to avoid rabbit holes, while the models only become useful partners if you invest enough hours and bring real domain knowledge.
More from Models
- One prompt gets Claude Opus to build a third-person League of Legends action game — TAbrodi · 2026-09-26
- Claude's Sweet Music Videos Creep Out a Self-described Non-doomer — ctjlewis · 2026-09-26
- Unified Cowork/Projects View Across Machines Has One User Saying Anthropic Is Back — toptickcrypto · 2026-09-26
- Open System 1 model runs at 45ms per query on laptop CPU, 70% on support tickets — ghumare64 · 2026-09-26
- Researcher says Claude solved his months-old multimode spectrum problem in hours — jwt0625 · 2026-09-26
- ARC Prize once insisted plain deep learning couldn't crack ARC — then o3 happened — teortaxesTex · 2026-09-26