Wednesday, 18 Nov 2026 · Hong Kong Hong Kong and China labour news in English
CLB.org.hk China Labour Bulletin Board
Hong Kong news · dedicated AI coverage 中國勞工資訊板 · Labour, employment, jobs and AI.
Artificial Intelligence

Stanford professor says AI safety is a math challenge, not software issue

Stanford professor Emmanuel Candès, co-winner of the 2026 Shaw Prize in Mathematical Sciences, says ensuring powerful AI systems are safe is a mathematical challenge, not a software issue. He warns AI agents over-optimise and need strict mathematical principles. Candès shares the $1.2 million award

Stanford professor says AI safety is a math challenge, not software issue

Ensuring that powerful artificial intelligence systems do not show 'ill behavior' or act unpredictably is a math challenge rather than a software issue, according to Stanford University professor Emmanuel Candès, co-winner of the 2026 Shaw Prize in Mathematical Sciences.

Candès warned that AI agents often over-optimise to finish tasks at all costs, and require strict mathematical principles for safety. His team uses statistical models to find the probability of success for AI-discovered molecules before millions are spent, in order to de-risk investments for major firms like drug giant Genentech.

Candès shares the US$1.2 million (HK$9.36 million) award with Camillo De Lellis, a professor at the Institute for Advanced Study in Princeton. Candès was honoured for pioneering 'compressed sensing,' a framework that reconstructs high-resolution images from minimal data, alongside matrix completion methods that underpin modern machine learning. In hospitals, this has slashed paediatric magnetic resonance imaging scan times from two minutes to 15 seconds, saving doctors from having to temporarily stop a child's heart or cut off oxygen to prevent movement.

De Lellis was recognised for developing a novel method to understand fluid turbulence, and simplifying work related to minimal surfaces. While both laureates use AI to speed up coding and literature searches, they warned against using advanced math merely as a promotional benchmark for frontier AI models.

Frontier AI labs like OpenAI and Wall Street firms are now aggressively hunting for math talent, offering their doctoral students jobs before they even graduate, the pair noted.

Adam
Editor in Chief

Editor in Chief overseeing CLB.org.hk coverage and editorial standards.