During the first part of this talk, we will provide some background on how modern generative AI chatbot systems work, focusing on the setting of answering math questions. We then focus on the current state of “AI and math.” While it is clear that AI systems are at least helpful assistants for some parts of research mathematics, their ability to answer research-level math questions without an expert in the loop is less clear. To assess this, we are running a community experiment called “First Proof”, where we have shared a set of ten math questions which have arisen naturally in the research process of the authors but which had not appeared publicly until February 7, 2026. Answers to the questions are known to the authors of the questions, but will remain encrypted for one week, while the experiment is running. We will discuss the set-up and initial outcomes from the experiment, and finally discuss next steps for further assessments.

About the speaker: Rachel Ward is a professor of mathematics and holds a distinguished professorship in Data Science at the Oden Institute for Computational Engineering and Sciences at UT Austin. From 2023-2025, she was on leave as Principal Researcher at Microsoft Research. From 2017-2018 she was a visiting Researcher at Facebook AI Research. Her research interests include optimization, randomized numerical linear algebra, theoretical machine learning, and AI + Math.