AI challenges math by solving the unsolvable
Announcements over recent months that artificial intelligence models have cracked previously unsolvable problems are shaking top-level mathematicians to the core, and they are now asking themselves: Where do we come in?
Recent AI developments have caused a stir among top mathematicians, who are questioning the future of mathematical research. Hugo Duminil-Copin, a Fields Medal winner, recently expressed his concerns at the French Mathematical Society. Since April, AI companies have been regularly publishing solutions to previously unsolvable problems, demonstrating the power of their latest models.
OpenAI recently released solutions to 719 previously unsolved problems. While acknowledging the emotional impact, Stephane Mallat, a Gold Medal recipient, maintains that mathematicians are still essential. Mathematics is not just about calculation but also about clear thinking. The Simons Foundation announced a $15 million program co-funded by OpenAI and Anthropic to support mathematicians in dealing with AI.
Mallat noted that mathematicians are needed for teaching, as young people may be less interested in mathematics due to AI's rapid advancement. Francois Charles, a mathematician and department head, acknowledges the challenges posed by AI but remains optimistic about overcoming obstacles.
Written by urgent.news from Phys.org's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
