Measuring how AI agents handle everyday wet-lab science

Source: OpenAI News • Nov 5, 2026 • openai.com

OpenAI introduces BioBasicsBench, a 100-task benchmark measuring how well AI agents interpret raw experimental data and handle everyday wet-lab science.

Read Full Article →