Stuart Burrell

Director of AI Research and Innovation VISA

Stuart Burrell is Director of AI Research and Innovation at Visa, where he leads the Frontier AI Unit within the Risk and Security AI Lab. His team develops next-generation machine learning systems for Visa's global payment network, focusing on large-scale payments foundation models, multimodal reasoning, fairness, robustness, machine-generated text detection, and adaptive AI under distribution shift. His team's work spans theoretical AI foundations to practical deployment, with research publications at venues such as AISTATS and ICLR and production models that protect billions of transactions. Prior to joining Visa, Stuart was an AI Research Scientist and Engineer at Featurespace, a machine learning start-up specializing in fraud detection and financial crime prevention, which was acquired by Visa in 2024. Stuart holds a PhD in Pure Mathematics from the University of St Andrews and an MPhil in Machine Learning from the University of Cambridge.

Responsible AI Summit Main Conference Day 2 - Tuesday 22 September

2:00 PM Presentation / Case Study – Fairness and Safety: Building Trusted AI Systems for High-Stakes Domains

AI systems in high-stakes domains such as finance must remain reliable and trustworthy as new capabilities emerge at scale. In this session, Dr Stuart Burrell, Director of AI Research & Innovation, and Dr Maeve Madigan, Senior Research Scientist at Visa, share two research advances that help build safe and responsible AI systems at global payments scale. The first looks at research published at AISTATS 2025 and ICLR 2026 on detecting machine-generated text and what it takes to make such methods robust in real-world adversarial settings. Dr Madigan then turns to fairness, focusing on how bias can arise not only from individual models but as an emergent, collective property of complex multi-agent AI systems. Drawing on research published as a Spotlight paper at the 2026 AAAI Workshop on Agentic AI in Financial Services, she examines how these dynamics unfold, and what we can learn from large scale simulation studies in this area.

• Reliable detection of machine-generated text enhances safety and trust in many high-stakes domains.

• Bias can emerge as a collective property of complex multi-agent AI systems, beyond any single model.

• Principled safety and fairness research enables trustworthy AI at production scale.

Check out the incredible speaker line-up to see who will be joining Stuart.

Download The Latest Agenda

In Partnership With:

QUESTIONS?

Address: IQPC Ltd, 129 Wilton Road,
London SW1V 1JZ
Telephone: 44 (0) 207 368 9300
Fax: 44 (0) 207 368 9301
Email: enquire@iqpc.co.uk