Researchers debate whether observation-trained AI can guide hands-on biology experiments
suragnair · x · 2026-09-25
Stanford researcher Anshul Kundaje doubts AI can direct novel humans through bio experiments requiring deft, hard-to-verbalize manual skills without explicit motion control. suragnair counters that an AI trained by observing everything will internalize these abstract skills, eventually codifying them into instructions and verifying human execution by watching.
The exchange highlights a core obstacle for AI research agents in wet-lab settings.
More from AGI Musings
- Stanford Lecture: AI Slashes Build Costs, Entry Barriers and Switching Costs at Once — le_james94 · 2026-09-26
- Stanford Lecture Maps Every AI Era's Bottleneck; Today's Is High-Quality Tasks — le_james94 · 2026-09-26
- As LLMs Grow Context Windows, Redditors Debate the Future of Mathematics — Lolsebca · 2026-09-26
- AI Now at House Hearing: AI Companies Aren't Just Grading Their Own Homework, They're Writing the Curriculum — AINowInstitute · 2026-09-26
- With 60k ICLR abstracts, a researcher mourns LLM-generated figures homogenizing papers — mariyaivasileva · 2026-09-26
- AI-generated French novel shortlisted for literary prizes, professor weighs in to WSJ — TuhinChakr · 2026-09-26