Mathematician James Maynard, a Fields Medal winner from the University of Oxford, has been reflecting on the future of his field, as AI begins to take over. OpenAI recently revealed it had produced solutions to long-standing mathematical problems using its advanced model Astra, prompting mixed feelings among mathematicians.

The solutions spanned various fields, including data encoding and quantum game theory, raising questions about how AI will change the landscape of mathematical discovery. The excitement is tempered by concerns over how much credit should be given to AI versus human contributors.

One particularly contentious issue arose when OpenAI’s announcement initially downplayed the contributions of researchers Andreas Thom and Gábor Kun, whose recent work formed a crucial part of the solution. After some back-and-forth, OpenAI revised its language to better acknowledge these contributions.

The complexity of verifying such solutions means that mathematicians are still divided on their validity. However, there is broad agreement that AI’s involvement marks a significant shift in how mathematical problems are tackled.