AI & Tech··1 sources verified
DiscoveryAgent Pairs Autonomous Research Agents With Scientist-Directed Oversight
Summarised by Relevant News AI · Read time: 3 min
Researchers introduced DiscoveryAgent, a domain-agnostic system in which scientists select, refine, or replace hypotheses between research rounds while audit agents check checkpointed records. Scientists working with audit agents found 14 times more valid errors than scientists doing manual review, and the system produced the best variable-star classifier among the baselines tested. Case studies in astronomy, social science, and genomics included a photometric-redshift estimator with 0.971 AUROC for low-redshift classification.
Why it matters: It offers a concrete, measured design pattern for keeping humans in control of agentic AI research workflows, including scalable auditing of agent-generated claims.
All sources