ResearchOpenAI

OpenAI uses thousands of AI agents to study Alzheimer's

Original title: Using thousands of AI agents to uncover new leads in biological data

Plain language

What it is: OpenAI, PerturbAI, and Scripps Research used thousands of AI agents to analyze complex biological data from millions of brain cells. The project aimed to find new research leads for diseases like Alzheimer's and Parkinson's.
How it works: The system used thousands of parallel AI agents powered by a model called GPT-Rosalind. Each agent studied a specific gene disruption, analyzed cell responses, searched scientific literature, and wrote a report.
Where it is available: The resulting reasoning atlas is available for scientists to use and explore.
Example: The agents found that disrupting the Dctn1 gene increases the activity of genes that make cholesterol, suggesting a link between cholesterol transport and brain cell damage.

What it is: OpenAI, PerturbAI, and Scripps Research deployed a distributed multi-agent system powered by GPT-Rosalind to analyze a whole-brain CRISPR atlas. The system successfully mapped gene disruptions to cellular responses and scientific literature.
How it works: The architecture uses distributed biological reasoning, where thousands of parallel agents analyze individual gene perturbations. Each agent evaluates cellular responses, queries scientific literature, and generates reports that are compiled into a unified reasoning atlas.
Where it is available: The reasoning atlas is available online for researchers, and the system utilizes the GPT-Rosalind model.
Example: In a test, the distributed agents cited over 700 references compared to only 11 references cited by a single agent handling the entire dataset.

The headline and explanation on this page are based on OpenAI's post. The full details are in the original.

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