Two MIT professors who showed that artificial intelligence could find a brand-new antibiotic in weeks rather than years have won one of the newest prizes in science. Regina Barzilay and James Collins are the 2026 recipients of the Pritzker Prize for AI in Science Research Excellence, MIT News reported Monday.
The award recognizes their joint work in “pioneering AI-enabled antibiotic discovery, integrating deep learning, bioengineering, and experimental validation to identify novel antibacterial candidates against drug-resistant pathogens,” according to MIT.
About the prize
The University of Chicago set up the prize in 2024 in partnership with Caltech to recognize work that advances both AI and the natural sciences or engineering. Each award carries $50,000, and the 2026 prizes were presented at the third University of Chicago and Caltech Conference on AI+Science, the university’s Data Science Institute said.
This year is the first time the prize has gone to a team. Kristin Persson of the University of California, Berkeley, was named this year’s individual laureate for data-driven materials design and for founding The Materials Project, an open database of computed materials data.
From halicin to abaucin
Barzilay is MIT’s School of Engineering Distinguished Professor of AI and Health, and Collins is the Termeer Professor of Medical Engineering and Science. The two co-lead MIT’s Abdul Latif Jameel Clinic for Machine Learning in Health.
In 2020 they published a paper in the journal Cell, with then-MIT postdoc Jonathan Stokes as first author, describing halicin, the first antibiotic discovered by AI. It was named for HAL 9000, the computer in Stanley Kubrick’s 1968 film “2001: A Space Odyssey.” Halicin had originally been studied as a possible diabetes drug. A deep learning model trained on 2,500 molecules flagged it, and in testing it killed drug-resistant bacteria including Clostridium difficile, Acinetobacter baumannii and Mycobacterium tuberculosis, MIT said.
In 2023 the team, again working with Stokes, now at McMaster University, used machine learning to identify abaucin, a narrow-spectrum antibiotic that targets A. baumannii. Because it goes after a single species, it is less likely to drive resistance or harm helpful gut bacteria, according to MIT.
Why it matters
Antimicrobial resistance is projected to cause 10 million deaths a year by 2050, the University of Chicago said. Barzilay and Collins’ Antibiotics-AI Project, in partnership with the nonprofit Phare Bio, aims to develop seven new classes of antibiotics targeting seven of the world’s deadliest bacterial pathogens.
“Historically, antibiotic discovery has depended on laborious, trial-and-error strategies,” Collins said in MIT’s announcement. “With generative AI, we’re now seeing a shift from AI being used as a tool for discovery to AI being used as a design tool.”
Barzilay struck a more cautious note. “With AI, we can only automate parts of the pipeline, rather than totally transforming the discovery process,” she said. “To make it happen, we need to work really hard to build realistic AI models of human biology.”