Speaking in 2026

He is a recognised expert in the international governance of emerging technologies, diplomatic practice, institutional design, and technology regulation. He regularly advises government and industry leaders on these topics, with much of his recent research focusing on international governance regimes for AI.

Her research focuses on the technical tools and standards needed to support effective AI governance — post-deployment monitoring, incident tracking, and evaluation frameworks for GPAI systems — and has been published at ICML and NeurIPS. Previously Acting Head of AI Governance at interface, a European tech policy think tank, with research fellowships at Stanford’s Intelligent Systems Laboratory, the RAND Corporation, and the Centre for the Governance of AI.

Building off his July 2026 paper “Can AI agents conduct open-ended AI research? Early evidence from two case studies,” he is currently working on better tools and methods for transcript analysis of very large multi-agent runs, and estimating research skills and taste, due to be published in September 2026.

Previously, he was an Associate Director at RAND focusing on field-building, where he helped expand the organisation to over 100 people. Prior to RAND, he was a Senior Research Manager within the AI Governance and Strategy team at Rethink Priorities and a Research Scholar at the University of Oxford’s Future of Humanity Institute.

Sarah’s previous work at SPAR includes clarifying definitions of AI systems and AI models, as well as responsibility allocation for upstream versus downstream GPAI developers across jurisdictions. She was also a technical expert at Mila, and the co-founder of AI Governance Exchange, whose aim is to bridge Chinese and Western AI safety efforts to advance responsible development and cross-border standards.

Alongside the Oxford Martin AIGI, he also co-founded and does research at the Oxford Hardware AI Governance Lab and was a summer research fellow at the Centre for the Governance of AI (GovAI). Prior to this, he specialised in full-stack software engineering and semiconductor physics.

Before FLI, Oliver was a Senior AI Policy Fellow at the Institute for AI Policy and Strategy (IAPS) and a Research Scholar at the Centre for the Governance of AI (GovAI), working on UK AI policy. Previously, he came from the UK Civil Service, leading the HM Treasury team where he advised the Chancellor on economic and scientific Covid-19 developments.

Previously, she designed ISO 27001-aligned security policies for public sector clients and validated IT-security assessments for telecommunications firms at Accenture. She has a background in computer science.

Alongside this, he is a Principal Scientist at Martian, working on model routing, interpretability and agent evaluation. His research focuses on understanding what happens inside neural networks and using those insights to make AI systems safer and more governable.