AI – Math Transformation

The rapid developments of frontier models and their successes in solving mathematical problems will have a tremendous impact on the field of mathematics, and on many aspects of the career of mathematicians. This naturally raises worries, but deeply understanding mathematics will always be crucial, and frontier models can help in that endeavor.

On this page, we collect information, links, opinions, and activities either by MIT mathematicians, or of interest to our MIT math community.

  • (August 13, 2026.) Report of the Ad Hoc Committee on AI Use in Teaching, Learning, and Research Training. One of the report's many guiding principles is to put humanity front and center.
  • In August, in response to all the developments and announcements of AI-derived results, a survey of MIT math graduate students on AI in mathematics was initiated and conducted by them. Here is an Open Letter to Faculty on AI they provided to the whole department at the end of August. A Town Hall with the entire mathematics department ensued. These graduate students have also created an extended survey for junior researchers in the wider community. If you would like to help distribute it to graduate students and postdocs at your university, please contact abrande@mit.edu.
  • (Updated September 14, 2026.) "Use of AI in Mathematical Research: A Guide for Young Mathematicians" by Prof. Pavel Etingof. This was originally written for our PRIMES high-school students, but its audience was later expanded (and the guidance revised) given the tremendous successes achieved by frontier models.
  • (Posted September 15, 2026.) In this essay, Prof. Henry Cohn stresses the role of understanding and draws a parallel with technical debt in software development.
  • Prof. Shafi Goldwasser is one of the founders of the Institute for Responsible Superintelligence (RESI). RESI is a nonprofit research institute building the scientific foundations needed to make superintelligence safe by design.
  • AI for the Working Mathematician, a seminar series at Harvard's Center of Mathematical Sciences and Applications (CMSA).