MIT's Barzilay and Collins win Pritzker Prize for AI-discovered antibiotics

What the prize recognizes

Regina Barzilay is MIT's School of Engineering Distinguished Professor of AI and Health. James Collins is MIT's Termeer Professor of Medical Engineering and Science. According to MIT News, the award honors 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."

The Pritzker Prize for AI in Science Research Excellence was established in 2024 by the University of Chicago in partnership with Caltech. It recognizes contributions that jointly advance AI and the natural sciences or engineering. Barzilay and Collins are faculty co-leads of the Abdul Latif Jameel Clinic for Machine Learning in Health, and affiliates of the MIT Schwarzman College of Computing and the MIT School of Engineering.

From halicin to abaucin

The two published their first breakthrough paper together in 2020 in Cell, with then-MIT postdoc Jonathan Stokes as first author. Stokes worked in the Collins lab from 2017 to 2021. The paper described halicin, named after the fictional AI system HAL 9000 from Stanley Kubrick's 1968 film 2001: A Space Odyssey.

Halicin is a broad-spectrum antibiotic that was originally studied as a potential diabetes drug. A deep learning model trained on 2,500 molecules identified it. In testing, the research team found that halicin could kill antibiotic-resistant bacteria such as Clostridium difficile, Acinetobacter baumannii and Mycobacterium tuberculosis. MIT says a discovery process that once took years was completed in just a few weeks.

In 2023, Barzilay and Collins worked with Stokes, now an assistant professor at McMaster University, to use machine learning to identify abaucin. It is a narrow-spectrum antibiotic that works against A. baumannii, a resistant bacterium commonly found on hospital doorknobs and in the wounds of U.S. soldiers stationed in Iraq and Afghanistan. According to MIT, the narrow spectrum is its main advantage, because it minimizes the likelihood of bacteria developing resistance and avoids harming beneficial bacteria in the gut.

The researchers on where the field stands

Collins said that "historically, antibiotic discovery has depended on laborious, trial-and-error strategies." He described the process as "slow, costly, and unpredictable, often yielding few viable antibiotic candidates." With generative AI, he said, "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. "Despite significant progress in molecular modeling, we are still in early stages of drug discovery," she said. In her view, AI can only automate parts of the pipeline rather than totally transforming the discovery process. She added that realizing more will require hard work "to build realistic AI models of human biology."

Why it matters

MIT frames the halicin results as a demonstration, amid growing alarm over the global antimicrobial resistance crisis, that AI could be an effective tool for rapidly identifying new and effective antibiotics. The prize adds recognition for that line of work, which MIT says is widely seen as a milestone in AI-driven drug discovery. The researchers' own comments suggest the next step is moving from using AI to find candidates toward using it to design drugs, while acknowledging that much of the drug discovery process remains beyond what AI can currently automate.