The MIT-led Artificial Intelligence and Fundamental Interactions Institute (IAIFI) has received new support from the National Science Foundation (NSF) for an additional five years, increasing annual funding from $4 million to $4.98 million. This renewal marks a new phase for IAIFI. IAIFI has spent its first five years building a research model and interdisciplinary community around the central premise that while AI can open up new ways of doing physics, physics can help shape better AI systems.
Launched in 2020 as part of the National Institute for Artificial Intelligence program, IAIFI brings together researchers from MIT as well as Harvard University, Northeastern University, Tufts University, and Boston University. The research shows that while machine learning can accelerate discoveries in physics, insights from physics can make AI systems more principled and easier to interpret.
“From the beginning, IAIFI has been built around a two-way street: AI enables better physics, and physics enables better AI,” said Jesse Saylor, IAIFI director and MIT physics professor. “Over the past five years, we have seen this virtuous cycle unfold across multiple fields of physics and AI. This interaction is not only producing new results, but creating truly new ways of doing science.”
Research that crosses physics and AI
IAIFI’s research spans particle physics, nuclear physics, astrophysics, and fundamental AI, and many advances result from collaboration across these fields.
In particle physics, IAIFI researchers have developed AI techniques to process huge data rates from the Large Hadron Collider in real time, helping to turn the firehose of collision data into actionable physics. In nuclear physics, IAIFI researchers use AI-based generative techniques to model quark-gluon interactions in lattice quantum chromodynamics, creating new ways to study the structure of matter from first principles. In astrophysics, machine learning is being used to uncover new cosmic phenomena and improve the sensitivity of the MIT-led LIGO gravitational wave experiment.
At the same time, ideas from physics are influencing the development of new AI methods. IAIFI researchers develop learning algorithms and new model architectures that embed physics knowledge and best practices, such as symmetry, geometry, accuracy guarantees, and statistical methodologies, directly into neural networks, resulting in more reliable, interpretable, and data-efficient systems.
“AI is beginning to transform the way physicists approach the field’s most difficult problems,” said Mike Williams, Interim Director of IAIFI and Professor of Physics at MIT. “More importantly, the range of problems that we can realistically address has begun to expand, allowing us to pursue problems that were once completely out of our reach.”
nurturing the next generation
IAIFI is characterized by its investment in human resources. The IAIFI Postdoctoral Fellows Program supports young scientists pursuing research at the intersection of physics and AI, pairing each fellow with a mentor in both disciplines and fostering cross-institutional collaboration.
To date, eight fellows have completed the program. The three secured teaching positions. Some take research positions at large AI companies or join startups. This reflects how widely applied the skills developed at IAIFI are.
“The IAIFI Fellowship shows what can happen when young scientists are given the freedom and support to work beyond traditional boundaries,” says Fiala Shanahan, IAIFI interim vice president and MIT physics professor. “Our colleagues are not just making individual contributions to physics and AI, they are helping to shape fields that are growing at the intersection.”
IAIFI’s annual PhD Summer School is at the heart of a growing community of “centaur scientists” with expertise in both physics and AI. For the 2026 edition, the program received nearly 600 applications for approximately 100 in-person spots, with approximately 300 additional participants expected to participate virtually. Previous participants highly recommend this school to their colleagues for its combination of lectures, hands-on tutorials, coding sprints, and networking events.
At MIT, IAIFI is supporting the formation of new educational pathways, including an interdisciplinary doctoral program in physics, statistics, and data science (a collaboration between the Department of Physics and the Center for Statistics and Data Science), which has awarded 20 Ph.D. degrees since 2021. IAIFI members Phil Harris and Isaac Chuan have also developed a course in computational data science in physics, which is offered both on campus (course 8.16) and at the university. Free online courses from MITx.
growing community
In addition to its core research and training programs, IAIFI convenes researchers through its annual summer workshop, which this year will be held in the MIT Schwarzman College of Computing building. The Institute also aims to engage a broader audience through collaborations with the MIT Museum and Boston Museum of Science, hackathons, and widely viewed online content exploring AI and physics.
“IAIFI shows what is possible when researchers in physics, computation, statistics, and data science organize around a common scientific question,” says Nergis Mavalbara, dean of the MIT School of Science and the Curtis and Kathleen Marble Professor of Astrophysics. “Ongoing, cross-disciplinary collaborations like this are essential to the future of scientific discovery.”
IAIFI is hosted by MIT’s Nuclear Science Institute and is led by Director Jesse Saylor (currently on sabbatical), Interim Director Mike Williams, Interim Deputy Director Fiala Shanahan, Managing Director Marisa LaFleur, and steering committee members Lisa Barsotti, Isaac Chuan, Will Detmold, Bill Freeman, Phil Harris, Lina Necib, Tess Smit, and Marin Soljačić. (and steering committee members from other IAIFI universities).
Looking to the future
As a member of the National Institute for Artificial Intelligence program, IAIFI is part of a national effort to advance AI-driven discovery and innovation.
“The connections between NSF AI Institutes are as valuable and continue to grow as the work they do within them,” said Marisa Lafleur, Managing Director of IAIFI. “We are sharing management strategies and resources for training, community building, and collaboration that strengthen the entire network.”
For IAIFI, the new funding is an opportunity to go deeper into what the institute calls the “physics of AI”: using physical reasoning, physical challenges, and physical tools to not only apply AI, but to understand and improve it. This challenge will drive the Institute’s next phase, along with a growing community of researchers trained to work across disciplines.
“The first phase of IAIFI established a model of interdisciplinary research, early career talent, and a dynamic community organized around the idea that AI and physics reinforce each other,” Thaler said. “Now we have the foundation and the entrepreneurial spirit of centaur scientists to push that model into new territory and increase our ambitions.”
