Open AI has released solutions to 372 well-known mathematical problems. The company’s internal AI models produced the answers in a matter of hours. That speed contrasts with the time many research teams would need to tackle the same problems. For mathematicians around the world, the release is expected to have major consequences. The results may change how researchers use AI in their work and how they evaluate machine-generated solutions. The development raises questions about verification and credit, because published answers still require expert scrutiny. Beyond academic research, the technology could affect society more broadly as mathematics continues to underpin many fields.