Google's Co-Scientist autonomously discovers molecules and algorithms
iScienceLuvr · x · 2026-08-28
- Tool Overview: Google DeepMind (GDM) applied the Co-Scientist system to three real-world scientific domains for validation.
- Materials Science: Interfaced with a semi-automated CVD instrument to design a non-hazardous precursor route (C2Cl6) for Ti3C2T𝑥 MXene growth. Physical execution yielded 2D layered structures with similar characteristics.
- Biology: Built a system predicting emergent swarming phenotypes of engineered E. coli across inducer gradients from sparse imaging data, validated quantitatively against unpublished wet-lab data.
- Computer Science: Operated fully autonomously to discover an inference-time scaling architecture outperforming six frontier models on HealthBench Hard and Professional benchmarks.
Related event: Google DeepMind's Co-Scientist Extension Brings AI into Real-World Labs(30 posts)→
More from Apps
- Grok Bot can now read your X bookmarks and turn 2,000 saved posts into a second brain — 4KTV · 2026-10-16
- Full course: build a playable Unity fishing game with Claude Opus 5.5 in six steps — Roger_M_Taylor · 2026-10-16
- HyperFrames launches Reference Any Video and auto-applied Brand Kits for AI video — aftahi_ai · 2026-10-16
- Marc Lou praises PostShiba, a cute indie alternative to AWS SES for email infra — marclou · 2026-10-16
- Man with cerebral palsy uses HeyGen avatar to tell his life story in his own voice — HeyGen · 2026-10-16
- The one prompt that makes AI reveal its hidden assumptions — TheMoonMidas · 2026-10-16