DeepMind's Co-Scientist Runs Real Lab Experiments, Grows 3 2D Semiconductors on First Try
jiqizhixin · x · 2026-09-07
Google DeepMind published an 83-page paper on Co-Scientist, a Gemini-driven AI scientist that designs experiments, writes executable code, reads instrument feedback, and connects directly to real lab equipment — closing the loop between computation and physical experimentation.
Key results:
- In materials science, it generated 272 candidate synthesis protocols for 2D semiconductors; after 25 rounds of iterative experimentation, MoS₂, MoSe₂, and WS₂ all successfully grew single-layer crystals on the first experiment.
- In computer science, it went further by designing and executing AI experiments autonomously.
The system moves AI from in-silico hypothesis generation to fully autonomous real-world experimental execution.
More from AGI Musings
- OpenAI computer-use agents near superhuman speed, say the team building them — a16z · 2026-10-26
- Kai-Fu Lee publishes AI Native, a roadmap for building AI-native companies — kaifulee · 2026-10-26
- RCT: ChatGPT temporarily boosts consultants' data science skills, gains vanish without it — EricTopol · 2026-10-26
- Dev on AI games backlash: reject AI slop, not the tool itself — ezshine · 2026-10-26
- Worry less about AI than about the bad actors wielding it — zakkohane · 2026-10-26
- If an agent can rewrite itself, does your consent still mean anything? — tallmetommy · 2026-10-26