Two dire warnings, one from Terence Tao, the other from someone who just quit Anthropic

Terence Tao has issued a warning regarding the speed at which artificial intelligence is solving the most challenging problems in mathematics. According to a report from Yahoo Tech, Tao suggests that AI is solving these best problems faster than they can be replaced. This development is occurring alongside the International Congress of Mathematicians (ICM) 2026, where discussions are centering on the fundamental purpose of mathematics in an age dominated by AI, as detailed in coverage from Quanta Magazine. The situation highlights a growing tension between automated problem-solving capabilities and traditional mathematical inquiry. Coverage from New Scientist emphasizes Tao's specific claim that AI companies are actively harming mathematics.

This sentiment is echoed in FlowingData, which explores what it describes as the paradox of solving mathematical problems through the lens of Tao's perspective. The reports indicate a systemic concern that the rapid automation of proofs and solutions may disrupt the intellectual ecosystem of the field. These warnings are appearing across a diverse range of technical and scientific outlets, signaling that the concern is shared by both academic specialists and general science journalists. Context for these warnings is further expanded by Marcus on AI, which links Tao's concerns with another dire warning coming from an individual who has recently quit Anthropic. While the specific nature of the Anthropic departure's warning is not detailed in the provided headlines, the pairing of these two events suggests a broader trend of alarm coming from high-level experts within both the mathematical and AI development communities.

The timing coincides with the 2026 ICM, a premier gathering for the global mathematics community, making the discourse particularly timely as the field evaluates its future. Looking forward, observers will likely monitor the ongoing discussions at the ICM 2026 to see how the mathematical community intends to respond to the influence of AI. Further details regarding the warnings from the former Anthropic employee may emerge to clarify the connection between corporate AI development and the academic harms cited by Tao. Based on the coverage, the primary point of tension remains the pace of AI discovery versus the human capacity to generate and frame new mathematical challenges.