Tyler Cowen Responds to Mathematicians’ Concerns About AI

Summary

Tyler Cowen discusses a statement on artificial intelligence signed by Terry Tao and other prominent mathematicians. He rejects cynical readings of the declaration and says he shares its concern that AI could make established intellectual methods less central, comparing the situation to a future in which economists can obtain answers without internalizing marginalist ideas. Cowen disputes the statement’s suggestion that the outcome will be determined largely by humans controlling the technology, arguing instead that mathematicians’ own adaptation may be the main variable. He says the community could reward people who supply missing mathematical understanding, even when an AI has already proved or disproved the relevant theorems. He expects mathematics funding and public interest to increase in the medium term because AI makes mathematical knowledge more consequential, while acknowledging that mathematicians may lose status relative to AI systems. Cowen does not expect companies to manage this transition for the field and argues that mathematicians must develop more AI-compatible intellectual and professional approaches. He also suggests that humans could guide AI toward producing, explaining, and teaching deeper mathematical understanding, and concludes that the current complaints may be premature because the technological shift is still at an early stage.