-----

This talk presents a case study in AI-assisted mathematical experimentation, beginning with a specific problem in computational number theory and extending to larger, more open-ended problems. McLain's goal was to test how much of the research process could be delegated to frontier AI systems. She will describe what worked, what failed, and what these experiments suggest about evaluating AI tools for real mathematical research. For the LPS community, this offers a concrete setting in which to examine how new AI tools reshape scientific practice, including the division of labor between humans, machines, and formal methods. This work was undertaken as part of the DARPA expMath program.

-----

 

Connect with us: