MIT scientists who used AI to find new antibiotics win the Pritzker Prize for AI in science
Regina Barzilay and James Collins helped discover halicin, the first antibiotic found by artificial intelligence, as drug-resistant germs spread.
Two Massachusetts Institute of Technology professors who used artificial intelligence to find new antibiotics have won the 2026 Pritzker Prize for AI in Science Research Excellence, MIT announced Monday. Regina Barzilay, a computer scientist, and James Collins, a bioengineer, were honored for "pioneering AI-enabled antibiotic discovery."
The prize was created in 2024 by the University of Chicago with the California Institute of Technology, known as Caltech, to reward work that advances both AI and the natural sciences or engineering.
Why their work matters
Many germs have evolved to survive the antibiotics doctors rely on. These drug-resistant bacteria, often called superbugs, make common infections harder to treat.
Finding a new drug has traditionally meant testing chemicals one by one, a slow and costly process. In 2020, Barzilay, Collins and then-postdoctoral researcher Jonathan Stokes instead trained a deep learning model, a type of AI that learns patterns from examples, on about 2,500 molecules and asked it to predict which other chemicals might kill bacteria.
It flagged halicin, a compound first studied as a possible diabetes drug. In lab tests, halicin killed several drug-resistant bacteria. MIT said a search that once took years was done in weeks. The team named it after HAL 9000, the computer in the film "2001: A Space Odyssey."
In 2023 the group used AI to find abaucin, which targets only Acinetobacter baumannii, a hospital germ. Hitting just one species lowers the odds of bacteria developing resistance and spares helpful gut bacteria.
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 cautioned that AI can so far automate only parts of drug discovery, and that more realistic AI models of human biology are needed.
Source: MIT News, Nov. 2, 2026.