Leading mathematicians in Trinity for long-planned meeting to develop Deligne–Drinfeld conjecture within days of new OpenAI proof announcement

Posted on: 15 October 2026

The Deligne–Drinfeld conjecture

The Deligne–Drinfeld conjecture concerns mathematical symmetries appearing in several seemingly unrelated areas of mathematics and theoretical physics, including the theory of knots and braids, quantum physics, number theory and algebraic geometry.

One family of these symmetries is described by the Grothendieck–Teichmüller group, which arises in the study of braids, associators and quantization. Another comes from the theory of motives, which is a central subject in modern arithmetic geometry.

The conjecture predicts a remarkable correspondence between these structures. Although important progress had been made, notably by Francis Brown of Oxford University (Simons Visiting Professor at HMI from 2018 to 2024), a longstanding part of this relationship remained unresolved.

But on 6 October, OpenAI released a claimed proof of the conjecture as part of a much larger collection of AI-generated mathematical results.

The OpenAI announcement and the HMI workshop

By an extraordinary coincidence, the HMI is this week hosting an international workshop on Grothendieck–Teichmüller theory, double shuffle relations and the Kashiwara–Vergne problem. The workshop had been planned long before OpenAI's announcement.

The Deligne–Drinfeld conjecture lies at the heart of the mathematics discussed at the workshop, which brings together leading international researchers in the field.

Already on the first day, Leila Schneps of Paris presented her understanding of the newly released AI-generated proof, while Thomas Willwacher of Zurich has also developed a substantial simplification of the argument.

According to the workshop organisers, he has identified two new ideas that reduce the original, approximate 40-page proof to an argument of approximately eight pages. He will present this work during the workshop.

The meeting will also include a discussion of the implications of these developments for other longstanding conjectures involving the double-shuffle and Kashiwara–Vergne groups.

AI and the future of mathematical research

“The recent advances in AI-generated mathematics present both remarkable opportunities and important challenges,” said Anton Alekseev, Professor at the University of Geneva and an HMI Simons Visiting Professor.

“The Deligne–Drinfeld conjecture provides an especially interesting example. A problem that had resisted mathematicians for many years has become the subject of an unexpected AI-generated argument. Within days, human researchers, many of whom are meeting here in Trinity’s HMI, have begun analysing it, finding possible simplifications and exploring its implications for other mathematical questions.”

“The value of mathematical research lies not only in establishing whether a statement is true, but also in understanding why it is true, identifying the underlying ideas and connecting them to broader theories,” added Dr Florian Naef, from Trinity’s School of Mathematics.

“And the developments discussed at Trinity illustrate how AI-generated results and human mathematical insight may interact. AI could accelerate the exploration of difficult mathematical problems, but mathematicians remain essential to interpreting, validating and developing the ideas that emerge.”

The HMI workshop offers a timely opportunity to examine this interaction at the forefront of mathematical research.

• Tags:
• Research
• Science