SOURCE-LINKED INTELLIGENCE
Measuring how AI agents handle everyday wet-lab science
OpenAI introduces BioBasicsBench, a 100-task benchmark measuring how well AI agents interpret raw experimental data and handle everyday wet-lab science.
Read original source ↗ Open in workspace
- recordType
- article
- region
- Global
Evidence & attribution
- OpenAI News · 2026-11-06T00:00:00.000Z
First collected: 2026-11-07T04:23:34.316Z. This is not the publication date.