# OpenAI's BioBasicsBench finds its best model scores 66.7% on interpreting raw wet-lab data

The 100-task benchmark, built with the enzyme-design company Ligo, reports scores only for OpenAI models and gives no human baseline, while OpenAI president Greg Brockman predicts major biology gains within 12 to 18 months.

- Published: 2026-11-07T06:25:02.319Z
- Canonical: https://polylog.news/ai/2026-11-07/openai-s-biobasicsbench-finds-its-best-model-scores-66-7-on
- Publisher: Polylog (AI desk)
- Section: tech
- Sources: [OpenAI News](https://openai.com/index/measuring-how-ai-agents-handle-wet-lab-science), [Polylog editors](https://polylog.news), [Unite.AI](https://www.unite.ai/openai-introduces-biobasicsbench-to-test-ai-agents-on-wet-lab-data/), [FourWeekMBA](https://fourweekmba.com/ai-openai-gpt-6-astra-gets-66-7-on-lab-tasks-2-1x-gpt-5-2/)

OpenAI introduced BioBasicsBench, a 100-task benchmark that measures how well AI agents interpret raw experimental data from routine lab work. OpenAI built it with Ligo, an enzyme-design company, according to Unite.AI. The tasks cover five…

This story is for subscribers. Read it in full at https://polylog.news/ai/2026-11-07/openai-s-biobasicsbench-finds-its-best-model-scores-66-7-on (subscription information: https://polylog.news/pricing).