Measuring how AI agents handle everyday wet-lab science

T1Covered by 1 source · first reported by OpenAI News at 06 Nov, 00:00 UTC

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

ai

Sources

• T1Measuring how AI agents handle everyday wet-lab scienceOpenAI News06 Nov, 00:00