Bioinformatician Anshul Kundaje doubts AI can talk novices through wet-lab experiments
anshulkundaje · x · 2026-09-25
Anshul Kundaje (co-lead of Stanford's AI Index) debated suragnair on X about the idea of AI instructing humans through wet-lab experiments. Suragnair's framing: the AI doesn't execute steps itself — it learns how to best instruct a human, and recommends course corrections during or after execution.
Kundaje's pushback: many bio experiments rely on deft movements and touch-based, hard-to-verbalize skills. Without explicit motion control, he doesn't see how a novice could follow instructions from an AI that knows exactly what to do. He concedes the approach might work for classical chemistry experiments that are more exact, but doubts it scales to serious biology.
More from AGI Musings
- If open source wins, AI value accrues to whoever serves inference cheapest — markjeffrey · 2026-09-26
- EU's top AI adviser: stop chasing US and China frontier models, fix productivity instead — alexvoica · 2026-09-26
- At OpenAI's 2026 AGI Economy Conference, researchers call for economic-task evals — soumitrashukla9 · 2026-09-26
- AI could codify tacit organizational knowledge, worth up to one US GDP, paper says — soumitrashukla9 · 2026-09-26
- Blaise Agüera y Arcas marks one year of 'What Is Intelligence?' — blaiseaguera · 2026-09-26
- Tech emits 4% of global carbon, more than aviation, and data centers eat 2% of power — ingliguori · 2026-09-26