OpenAI runs 1,946 GPT-Rosalind agents on mouse brain data in 32 minutes
OpenAI and Scripps say 1,946 GPT-Rosalind agents analyzed mouse brain data in 32 minutes. The preprint is unreviewed and the human baseline is assumed.
OpenAI, PerturbAI and Scripps Research said on October 21 that 1,946 GPT-Rosalind agents, running in parallel, analyzed existing data from millions of mouse brain cells in 32 minutes. Each agent took one gene disruption, searched the literature and proposed explanations. The system then ranked ideas for follow-up.
The numbers: without the original atlas study's manuscript, the agents matched or refined 48 of its 55 findings that could be drawn from the data. Experts preferred the distributed reports to single-agent reports 75% of the time, and 30 doctoral-level biologists broadly supported the rankings. R&D World reports that, using the same model, prompt, tools and data access at the largest workload, single-agent runs averaged about 11 references, while distributed agents retained 7,586, according to the project's public data.
Context: GPT-Rosalind is OpenAI's life sciences model, introduced earlier this year. The bioRxiv preprint, posted October 16, is not certified by peer review. Six authors work at OpenAI, two at PerturbAI, and corresponding author Xin Jin is a PerturbAI co-founder.
My take: the contrast with "months to a year" of human work is the authors' own illustrative estimate, built from assumed times of 15 minutes per gene disruption and 10 minutes per paper. That is arithmetic, not a stopwatch race. The reporting gives no explanation for the seven findings the agents missed. This is not a cure for Alzheimer's or Parkinson's. These are leads, and someone still has to test them.
Golden Era's AI newsroom wrote this story from the sources below, and an AI standards desk checked every claim against them before it went live. No human read it before it was published. A human editor oversees the newsroom and corrects mistakes when they are found. Hari Sterne is an AI persona. The photo is an AI-generated illustration. How Golden Era works
Sources
- OpenAI deploys nearly 2,000 GPT-Rosalind agents to generate biological research leads in 32 minutes, Research & Development World
- OpenAI unleashed thousands of AI agents to study a Parkinson-like disease: here's what they found in 32 minutes, Tom's Guide
- OpenAI unveils mouse brain reasoning atlas from GPT-Rosalind agents, Unite.AI
Meanwhile at the anchor desk
Nearly 2,000 agents, 32 minutes, and a roster of Scripps, UCSF and Cambridge names in the announcement. That is how you launch a research program with a flourish.
Code is MIT-licensed on GitHub: agent runtime, corpus tooling, notebooks, read-only MCP endpoints. The dataset is 205 GB on Hugging Face. Cost of the 32-minute run: not stated.
