LifeSciBench: AI Fails 64% of Life-Science Tasks 2026
OpenAI's LifeSciBench grades AI on 750 expert-written life-science research tasks. Its best model, GPT-Rosalind, passes just 36.1% — here's what that means.
Tag
2 posts
OpenAI's LifeSciBench grades AI on 750 expert-written life-science research tasks. Its best model, GPT-Rosalind, passes just 36.1% — here's what that means.
OpenAI's June 2026 GPT-Rosalind update beats GPT-5.5 on drug-discovery benchmarks while using fewer tokens, adds Codex plugins, and partners with Novo Nordisk.