Ethos Workshop Series: Responsible Research Workshop | Dr Christopher Burr

Register for the event here: https://www.eventbrite.co.uk/e/responsible-research-and-the-trustworthy-and-ethical-assurance-platform-tickets-1992811617192 

Bio

Dr Christopher Burr is a Principal Research Scientist and Theme Lead for Trustworthy and Assured AI at the Alan Turing Institute. He is also a Senior Visiting Fellow at the University of Sheffield and a Visiting Fellow at the University of York's Centre for Assuring Autonomy. 

His work sits at the intersection of philosophy, computer science, and policy, and focuses on a central question of AI governance: how do we establish justified trust in increasingly autonomous systems? He leads the Trustworthy and Ethical Assurance (TEA) platform, open-source tooling for building structured, evidence-based assurance cases, which is included in the UK Government's Introduction to AI Assurance as a primary case study. He has also designed and delivered responsible research and innovation training for researchers, policymakers, and practitioners, including through the Turing Commons platform.

His current research addresses the assurance of agentic AI and AI-enabled digital twins across safety-critical domains, from healthcare to critical national infrastructure. From April 2026 he will be PI for Project DARTER (Digital twin assurance via runtime trust and evidence reporting) — funded by the UK’s Digital Twin Centre (Digital Catapult).

Workshop Description

Title: Responsible Research and the Trustworthy and Ethical Assurance (TEA) Platform

Responsible research is easy to endorse and hard to evidence. How do you show, to a funder, an ethics board, or an affected community, that a claim like "this system is fair" or "this system protects privacy" is actually justified, and not simply asserted?

This hands-on workshop introduces assurance cases as a practical method for responsible research and innovation, using the open-source Trustworthy and Ethical Assurance (TEA) platform, developed with support from BRAID UK. An assurance case makes a structured argument: a top-level goal, the property claims that must hold for that goal to be true, and the evidence that grounds each claim.

After a short presentation and a clarificatory Q&A, you will work in small groups to build an assurance case for a single, demanding case study: an AI-driven facial recognition system designed to identify individuals from aerially captured drone imagery, with applications in surveillance, crowd monitoring at large public events, and locating missing persons. The system pairs generative AI for face reconstruction with energy-efficient neuromorphic classifiers, and it raises exactly the responsible-research questions that are hard to answer well (e.g. privacy and the protection of sensitive biometric data, demographic bias and fairness, transparency and explainability, and accountability in a national-security setting).

Each group chooses one of these as its assurance goal, agrees a strategy for decomposing it, and develops a few well-formed claims along with the evidence that would support them, working directly on the TEA platform. Although you will not finish a complete case, you will leave with a working grasp of how to turn a responsible-research principle into an argument you could actually defend.

No prior experience with assurance cases or the TEA platform is required.