A former applied mathematics researcher reflects on how artificial intelligence is reshaping the discipline, drawing a sharp distinction between applied and pure mathematical work. The author, who spent years simulating light wave interactions in liquid crystals for a Ph.D., now estimates AI could complete such applied tasks in days or even hours. This technological shift is forcing a reevaluation of what it means to be a mathematician.
The contrast is most stark when considering pure mathematics. The author recalls colleagues who toiled for years on abstract problems, often without publishing a single paper, leaving a seemingly slow and frustrating output. In hindsight, this struggle is now understood not as a failure but as a fundamentally different kind of intellectual labor, one that currently resists AI automation.
Applied mathematics, focused on modeling and simulation for real-world problems, appears increasingly vulnerable to AI-driven efficiency gains. The author's own Ph.D. work on liquid crystals, once cutting-edge, is now considered outdated and easily replicated by machine learning models. This suggests a growing divide between problems solvable by pattern recognition and those requiring deep conceptual leaps.
For pure mathematics, the path forward is less clear. While AI can assist with calculations and even suggest conjectures, the core act of creating entirely new mathematical frameworks remains a deeply human endeavor. The field may see a shift in value, rewarding the ability to pose the right questions rather than merely computing answers.
This evolution does not spell the end of all mathematical work but rather a redefinition of its roles. The most profound challenges—building new logical structures from scratch—still appear to require the uniquely human capacity for sustained, creative struggle.