377 AI solutions: who defines progress in mathematics?
FAZ reports on OpenAI’s results for 377 mathematical problems. The Association for Human Mathematics (AHM) criticises the mass release as “not a demonstration of scholarship, but a demonstration of power”.
According to FAZ, OpenAI released 722 manuscripts containing AI-generated results on 377 previously open mathematical questions. Reactions range from enthusiasm to concern about the future of human research. [1]
Verification is a central issue: a machine-checked formal proof must correspond exactly to the original claim. The article describes doubts about this correspondence in the previously released Navier–Stokes proof. [1] The AHM also criticises the company’s power and its publication model. [2]
Our interpretation: this example shows scientific work changing. Alongside finding solutions, shared understanding, exposition and the rules for recognising AI results become crucial.
Sources
This indication by Martin Ebner is licensed under
CC BY 4.0
Comments 0
Comments are moderated. New comments appear after review by the editors.
No comments yet.