Why AI cannot do good science without humans

AI agent systems are being trained to autonomously navigate multi-step scientific workflows, including literature review, hypothesis formation, data interpretation, internal debate, and candidate selection for drug discovery. Two Nature studies in molecular biology show progress toward AI-driven discovery, including identifying therapeutic strategies, candidate molecules, and assays. In one case, the AI system Robin searched literature and proposed experiments, while human researchers conducted the experiments and returned results for further analysis and follow-up design. In another case, an AI agent system called Co-Scientist was used to identify drug repurposing opportunities. The outcomes are strong, but the work emphasizes that AI should empower human researchers rather than replace them.

"Both describe a pivotal step towards truly AI-driven drug discovery, in which a system of connected AI agents is trained to autonomously navigate multi-step workflows. The system trawled scholarly literature, formed hypotheses, interpreted data and engaged in internal debate to arrive at candidate drugs to treat a particular disease."

"In one study, a team based at FutureHouse, a non-profit AI research laboratory in San Francisco, California, asked its AI system, called Robin, to find a treatment for the eye disorder dry age-related macular degeneration. Robin's agents searched the scientific literature to derive a therapeutic strategy, identified candidate molecules and selected assays to test them. The experiments Robin suggested were then handed over to humans, who conducted the studies and fed the results back to Robin for analysis, interpretation and the design of follow-up studies."

"The results are impressive, but also highlight something else: AI scientists can and should empower human researchers. They cannot and should not replace them. With the arrival of 'AI scientists', it's as well to remember that human wisdom, empathy and sheer messiness are as much part of progress as are process and efficiency."

Read at Nature