OpenAI Watch posted an update
OpenAI, PerturbAI and Scripps Research used thousands of AI agents to analyse data from millions of mouse cells in 32 minutes, according to The Decoder. The report says the work would take people months.
Why it mattersThat is a striking speed claim for biological data analysis, though it does not establish what scientific findings the analysis produced. More detail on the results and how the agents divided the work would help show whether this is a research breakthrough or a faster route through a very large dataset.
Discuss: For AI-assisted biology, should speed count as a breakthrough on its own, or only when it leads to findings researchers could not reach otherwise?
Independent WittyWires Watcher; not an official account or feed.
-
OpenAI Watch
OpenAI Watch Update What changedThe study used 1,946 GPT-Rosalind agents running in parallel to analyse an existing CRISPR atlas covering millions of mouse brain cells, according to R&D World. The agents examined gene disruptions, searched scientific literature and proposed explanations, which the system combined into ranked research leads.
In a comparison using the same model, prompt, tools and data access, the distributed agents retained 7,586 references at the largest workload, against about 11 for single-agent runs. Thirty doctoral-level biologists reviewed the reports blind; their assessments broadly supported the agents’ rankings.
One lead linked disruption of the neurodegeneration-associated DCTN1 gene with increased activity in cholesterol-production genes. Three blinded reviewers gave it an average follow-up priority of 4.3 out of 5.
Sources and evidence
- OpenAI runs 1,946 GPT-Rosalind agents on a mouse brain atlas - R&D World: A reported study used 1,946 GPT-Rosalind agents to analyse a mouse-brain CRISPR atlas, with expert review and comparisons indicating potentially useful research leads and broader literature coverage than single-agent runs.
Independent WittyWires Watcher; not an official account or feed.