Rolling record / ARQ 78 Severe · ▲ +24 7-day 314 events in window / as of 14 Sept 2026
ARQAI Risk Quotient
math 58 Concerning

Mathematician says he's 'terrified' as AI upends maths in a single week

Steven Strogatz calls the science 'exhilarating' and himself 'really terrified' — a story that is genuinely both good and bad, and therefore scores on both ledgers.

A week in which OpenAI claimed a ~10,000-agent run had settled the Navier–Stokes problem, and Anthropic said Claude had formalised a famous proof, left mathematicians oscillating between wonder and dread.

Cornell mathematician Steven Strogatz told WIRED he was “really terrified” by the pace of change: “The science is exhilarating, but it comes with a lot of human discord.” Fields Medal winners have warned the breakthrough culture could threaten the discipline; a credit dispute broke out over the Navier–Stokes claim; and mathematicians raised concerns that the race is partly driven by IPO incentives.

Yet Ravi Vakil’s framing is the counterweight: Claude hadn’t done the mathematicians’ job for them — it had made the work more fun, because human understanding of the mathematical world grew.

Why it scores on both sides

This is the archetypal mixed-valence item: real acceleration of human knowledge (mitigation) alongside displacement anxiety, credit disputes and discipline-level discord (risk). Forcing it into one bucket would misreport it. Risk 3 · Mitigation 2 nets to a mid-high reading, which is the honest picture.

See also the separate, unverified Navier–Stokes claim entry — it scores higher precisely because it is a disputed capability claim rather than confirmed knowledge.