Verifiable Environments for AI Agents in Biology: Why Frontier Models Can't Be Trusted Yet
kenbwork · x · 2026-08-02
This is an in-depth talk on building verifiable environments for AI agents in biology. It covers measurement methods in modern bio research, utilizing data analysis as an executable substrate for science, and five years of lessons from applying coding models to biological workflows.
The presentation also explores why current frontier models cannot be fully trusted (yet) in these scenarios, detailing the design principles of SpatialBench, the anatomy of evaluations, human verification, and new work in multi-omics, therapeutics, and biosecurity.
Related event: LatchBio Warns Frontier AI Models Remain Unreliable in Biology(2 posts)→
More from coding & agent
- DeepMind turns coding agent Antigravity into a science tool, surfacing novel antimicrobial peptide in one session — pushmeet · 2026-10-16
- Cross-review coding workflow: Claude writes the plan, Codex audits it — theteknosaur · 2026-10-16
- Simon Willison: code mode is the future of tool calling, if it can be sandboxed — springrod · 2026-10-16
- Normalizing Claude's schema-free MCP output before it hits downstream n8n nodes — snowingbol · 2026-10-16
- Agent named 'Ilya' reportedly scores 1.1pt higher on benchmarks — daniel_mac8 · 2026-10-16
- A plain notes.md fixed my agent's cross-session amnesia better than chat history — rahul_watertech · 2026-10-16