Terence Tao: New mathematical workflows | Future of Mathematics

Mathematics is entering an era of proof abundance, where AI-driven tools rapidly generate and verify theorems, yet the field faces a critical "impedance mismatch." While AI excels at solving explicit problems, it often fails at "proof digestion"—the essential human process of understanding, contextualizing, and communicating mathematical insights. This disconnect creates a backlog of technically correct but unusable proofs that lack narrative, pedagogical value, and connection to existing literature. To adapt, the mathematical community must shift its incentive structures to prioritize exposition and clarity over mere output volume. By establishing new infrastructure that distinguishes between automated "freeway" tasks and human-centric "pedestrian" research, mathematicians can harness AI's power without sacrificing the deep understanding required for long-term scientific progress.