# Using thousands of AI agents to uncover new leads in biological data

The system used GPT-Rosalind agents to analyze experimental data from millions of mouse brain cells in 32 minutes, generating a literature-linked reasoning atlas.

- Human experts preferred distributed-agent reports over single-agent reports 75% of the time in blinded evaluations
- Distributed agents cited more than 700 references at peak workload, compared to 11 cited by a single agent with the same dataset
- The reasoning atlas and tooling have been made publicly available on GitHub

## Why it matters

Biomedical researchers can rapidly synthesize massive cellular perturbation datasets with published literature to generate and prioritize testable laboratory hypotheses.

## Sources

- [OpenAI: Using thousands of AI agents to uncover new leads in biological data](https://openai.com/index/scaling-ai-research-for-complex-biological-data)

---

Summarized by dstilled on 2026-10-21. https://dstilled.ai/story/138c966e-0bcf-496b-bfb0-18ed4cd02b04
