 has received renewed support from the National Science Foundation for an additional five years.){kind=link}
The Institute for Artificial Intelligence and Fundamental Interactions (IAIFI), led by MIT, has received renewed support from the National Science Foundation (NSF) for an additional five years, increasing annual funding from $4 million to $4.98 million. The renewal marks a up-to-date phase for IAIFI, which has spent the first five years of its existence building a research model and interdisciplinary community around a central premise: that artificial intelligence can open up up-to-date ways of doing physics, while physics can assist shape better artificial intelligence systems.
Launched in 2020 as part of the National Institutes for Research on Artificial Intelligence program, IAIFI brings together researchers from MIT and Harvard, Northeastern, Tufts and Boston universities. His work has shown that machine learning can accelerate discoveries in physics, and that insights from physics can make artificial intelligence systems more rule-based and easier to interpret.
“From the beginning, IAIFI has been about a two-pronged effort: AI enabling better physics and physics enabling better AI,” says Jesse Thaler, director of IAIFI and professor of physics at MIT. “Over the past five years, we have witnessed this virtuous cycle in many areas of physics and artificial intelligence. The exchange brings not only new results, but also truly new ways of doing science.”
Research in the field of physics and artificial intelligence
IAIFI’s research spans particle physics, nuclear physics, astrophysics and basic artificial intelligence, and many advances result from collaboration in these areas.
In particle physics, IAIFI researchers have developed artificial intelligence techniques to handle massive amounts of data from the Enormous Hadron Collider in real time, helping to transform the firehose of collision data into actionable physics. In nuclear physics, IAIFI researchers operate artificial intelligence-based generative methods to model the interactions of quarks and gluons in lattice quantum chromodynamics, creating up-to-date ways to study the structure of matter from first principles. In astrophysics, machine learning is being used to discover up-to-date cosmic phenomena and escalate the sensitivity of MIT’s LIGO gravitational wave experiment.
At the same time, ideas from physics influence the development of up-to-date artificial intelligence methods. IAIFI researchers are developing learning algorithms and up-to-date model architectures that embed physical knowledge and best practices – including symmetries, geometric structures, accuracy guarantees and statistical methodologies – directly into neural networks, creating systems that are more reliable, interpretable and data proficient.
“Artificial intelligence has begun to change the way physicists tackle some of the most difficult problems in the field,” says Mike Williams, interim director of IAIFI and professor of physics at MIT. “More importantly, the frontier of problems we can realistically address is beginning to expand, making it possible to address issues that were once completely beyond our reach.”
Training the next generations
A characteristic feature of IAIFI is investing in people. The IAIFI Postdoctoral Fellowship Program supports early-career scientists conducting research at the intersection of physics and artificial intelligence, connecting each fellow with mentors in both fields and fostering collaboration between institutions.
To date, eight scholarship recipients have completed the program. Three have secured teaching positions; others have taken up research roles at leading artificial intelligence companies or joined start-ups, reflecting how widely the skills cultivated at IAIFI translate.
“The IAIFI fellowship shows what can happen when early-career scientists are given the freedom and support to work across traditional boundaries,” says Phiala Shanahan, interim deputy director of IAIFI and professor of physics at MIT. “Our fellows don’t just make separate contributions to physics or artificial intelligence — they help shape a growing field at the intersection.”
The annual IAIFI Summer Doctoral School has become a focal point for a growing community of “centaur scientists” with expertise in both physics and artificial intelligence. The 2026 edition of the program received nearly 600 applications for approximately 100 spots, with approximately 300 additional participants expected to join virtually. Previous participants have highly recommended this school to their colleagues due to its combination of lectures, hands-on tutorials, coding sprints, and networking events.
At MIT, IAIFI has helped shape up-to-date educational pathways, including an interdisciplinary Ph.D. program in physics, statistics, and data science—a collaboration between the Department of Physics and the Center for Statistics and Data Science—that will award 20 Ph.D.s as of 2021. IAIFI members Phil Harris and Isaac Chuang also developed a course in computational data analytics in physics, offered both on campus (Course 8.16) and within the program free online course through MITx.
A growing community
In addition to its core research and training programs, IAIFI organizes annual summer workshops for researchers, which will be held this year at the MIT Schwarzman College of Computing. The Institute also engages the broader public through partnerships with the MIT Museum, the Boston Science Museum, hackathons, and widely viewed online content on artificial intelligence and physics.
“IAIFI shows what becomes possible when researchers in physics, computation, statistics, and data science organize around shared science questions,” says Nergis Mavalvala, dean of the MIT School of Science and the Curtis and Kathleen Marble Professor of Astrophysics. “This type of sustained, interdisciplinary collaboration is essential to the future of scientific discovery.”
IAIFI is hosted by the MIT Nuclear Science Laboratory, chaired by Director Jesse Thaler (currently on sabbatical), interim director Mike Williams, interim deputy director Phiala Shanahan, and managing director Marisa LaFleur, as well as steering committee members Lisa Barsotti, Isaac Chuang, Will Detmold, Bill Freeman, Phil Harris, Lina Necib, Tess Smidt, and Marin Soljacic (and steering committee members from other IAIFI universities).
Looking to the future
As a member of the National Institutes for Artificial Intelligence Research program, IAIFI participates in nationwide efforts to advance artificial intelligence-based discoveries and innovations.
“The connections between NSF AI Institutes are as valuable as the work performed within them and are constantly evolving,” says Marisa LaFleur, managing director of IAIFI. “We provide management strategies and resources for training, community building and collaboration that strengthen the entire network.”
For IAIFI, the renewed funding is an opportunity to delve deeper into what the institute calls the “physics of artificial intelligence” – the operate of physical reasoning, physical challenges and physical tools not only to apply artificial intelligence, but also to understand and improve it. This program, along with a growing community of researchers trained to work in a variety of disciplines, is driving the next phase of the institute’s activities.
“The first phase of IAIFI established a model: interdisciplinary research, early-career talent, and a dynamic community organized around the idea that AI and physics are mutually reinforcing,” says Thaler. “Now we have the foundation – and the entrepreneurial spirit of our centaur scientists – to push this model into new territory and raise our ambitions.”