Math spent 40 years justifying funding via societal benefit — now it's suddenly about mathematicians

nifferin · reddit · 2026-10-08

The author traces the math community's reversal on AI, calling out a double standard: for decades, math justified taxpayer funding by promising societal benefit —

Then in 2026, as AI knocks off famous open problems, the response changed: the IMU-endorsed Leiden Declaration says AI threatens math's core values and hits early-career mathematicians hardest; 25 Fields Medalists say solving problems is "only a tool and proxy" for conceptual understanding; AHM urged a boycott of OpenAI.

The author's favorite example is Fields Medalist Cédric Villani: in 2012 he wrote that since "most of our research is paid for by the taxpayer," citizens must see math benefiting society; in 2018 he authored France's national AI strategy; in 2026 he signed the Fields letter claiming AI labs' race to solve problems harms the science. The author sees naked hypocrisy: how did math being solved become bad for society?

Original post →

More from AGI Musings

AGI Musings channel →