AI is changing math hiring in academia, and one prominent professor argues that the shift demands a more personal, communication-focused approach. Frank Calegari, a mathematician at the University of Chicago, contends that as artificial intelligence lowers the barriers to mathematical discovery, the traditional reliance on published papers to judge candidates is becoming insufficient.
In a recent essay, Calegari outlined what he now seeks in a researcher: a clear vision for where mathematics should go and the technical ability to explain why that vision might be achievable. His concern is that low-level research positions could be filled by students who have primarily produced AI-generated proofs. A 30-minute conversation, he suggests, would distinguish those candidates from promising ones who possess a clear sense of direction and the skill to communicate it.
The Paper File Problem
Calegari criticized what he called “lazy committees” that hired students on the strength of a single paper or publication—credentials that could now be generated by AI. He argued that the hiring process must become far more about personal interactions than about a paper file. Once that approach is adopted, he added, the responsibility falls on departments to train their own graduate students to communicate mathematics more effectively, a task he believes should be taken much more seriously.
The stakes are high for a field that has long prized solitary brilliance. AI companies have pushed aggressively into mathematics, flooding the discipline with hundreds of proposed solutions to complex problems. That surge has left many mathematicians unsettled, pondering how to remain relevant while the tools of their trade evolve at a rapid clip.
Charisma as a Differentiator
The idea that personal qualities matter more than ever is gaining traction. Last month, Cornell professor Steven Strogatz remarked that mathematicians with “rizz”—slang for charisma—had newfound stock because they could explain AI-generated results and turn them into new questions. In other words, the ability to interpret and communicate machine output is becoming a valued asset.
Responses within the academic community have been varied. Some professors have sworn off AI altogether, while most are trying to find ways to incorporate the technology into their work. Students, too, are struggling to adapt, with many demanding that their professors publish clear guidance on how to use AI—if at all. One student, in a recent MIT survey, wrote that high-level leadership from across the academic community is lacking.
For Calegari, the solution lies in human interaction. A conversation can reveal a candidate’s depth, their ability to articulate ideas, and their genuine passion for the field—qualities that a stack of AI-assisted papers cannot convey. As AI continues to reshape what it takes to get hired as a mathematician, the emphasis may shift from what you have published to how you think and speak about mathematics.