AI Lab Teams Cut Drug Research Time From Months to Hours

Scientists just got a powerful new partner in the lab, and it works at lightning speed.

Two groundbreaking AI systems described in Nature this week can team up with human researchers to develop hypotheses, propose experiments, and analyze data. What once took months now happens in hours.

Google DeepMind's system, called Co-Scientist, recently tackled a challenge in blood cancer research. The team asked it to find existing drugs that might treat acute myeloid leukemia in new ways. Within hours, Co-Scientist identified promising candidates. Human researchers selected five for further testing, and three showed real potential in early lab studies.

"It almost seems like an agentic, in silico implementation of the thought process in a scientist's head," says Vivek Natarajan, a researcher at Google DeepMind. "The goal is to give scientists superpowers."

A second system called Robin, developed by nonprofit AI lab FutureHouse in San Francisco, took on an eye condition called dry age-related macular degeneration. Robin consulted AI agents trained to review scientific literature, then suggested lab experiments to test various drug candidates. Humans carried out those experiments and fed the results back to Robin, which analyzed the data and proposed next steps.

The system identified ripasudil, a glaucoma drug, as a potential treatment and suggested specific tests to confirm its effectiveness. The work that typically requires weeks of reading papers and designing experiments happened in a fraction of the time.

These AI scientists still need human guidance and hands to carry out physical experiments. But they excel at processing massive amounts of information and spotting connections that might take researchers months to find on their own.

None of the drugs identified have been fully evaluated yet, and many promising candidates fail in later testing stages. But experts say these systems prove AI can arrive at plausible scientific hypotheses remarkably fast.

The Ripple Effect

The technology could transform how quickly researchers respond to health crises and discover new treatments. Labs worldwide could tackle more questions simultaneously, testing ideas that might otherwise sit on the back burner for lack of time.

Karandeep Singh, who oversees AI initiatives at UC San Diego Health, sees enormous potential as these tools become available to more scientists. The real test will come when researchers across different fields start using them daily.

For now, the message is clear: AI isn't replacing scientists, it's amplifying what they can accomplish. The partnership between human creativity and machine speed could accelerate discoveries that save lives and improve health for millions.

Science just got a lot faster, and that's very good news for anyone waiting for breakthrough treatments.