Barzilay, Collins Win 2026 Pritzker Prize for AI Antibiotic Discovery
Updated
Updated · MIT News · Nov 2
Barzilay, Collins Win 2026 Pritzker Prize for AI Antibiotic Discovery
2 articles · Updated · MIT News · Nov 2
Summary
Regina Barzilay and James Collins won the 2026 Pritzker Prize for AI in Science Research Excellence for joint work using AI to find antibiotics against drug-resistant pathogens.
Their breakthrough came in 2020 with halicin, the first AI-discovered antibiotic, which a deep-learning model trained on 2,500 molecules identified in weeks rather than years.
Tests showed halicin killed resistant bacteria including C. difficile, Acinetobacter baumannii and Mycobacterium tuberculosis, helping establish AI as a practical antibiotic-discovery tool.
The pair later helped identify abaucin in 2023, a narrow-spectrum antibiotic targeting A. baumannii that could limit collateral damage to beneficial gut bacteria and slow resistance.
The award, established in 2024 by the University of Chicago and Caltech, comes as researchers push generative AI from screening compounds toward designing new drugs amid a global antimicrobial-resistance crisis.
If AI can design antibiotics in weeks, what is stopping these superbug killers from reaching our hospitals today?
As AI shifts from finding to inventing drugs, who is ensuring these synthetic molecules will not cause unforeseen biological harm?
Can generative AI outpace the rapid mutations of deadly bacteria, or are we just accelerating the evolutionary arms race?
The 2026 Pritzker Prize for AI in Science: Transforming Antibiotic Discovery and Addressing the Global AMR Crisis
Overview
The 2026 Pritzker Prize marked a historic shift by honoring a collaborative team—MIT’s Regina Barzilay and James Collins, and UC Berkeley’s Kristin Persson—for pioneering AI-driven antibiotic discovery. Their work began with deep learning models that identified halicin, the first AI-discovered antibiotic, and evolved into the Antibiotics-AI Project with Phare Bio, targeting deadly drug-resistant pathogens. This approach replaced slow, manual drug discovery with automated, closed-loop AI systems that rapidly design and test new molecules. By integrating explainable AI and meeting strict regulatory standards, their efforts not only accelerate drug development but also set a new standard for interdisciplinary, team-based science against the global antimicrobial resistance crisis.