Library Science Talk - Safeguarding Testimony Archives: An Ethical Impact Assessment Framework with Preliminary Red Team Validation

by Nathan Heath, Broderick James McDonald, Susan Nesbitt, Dr Jennifer Victoria Scurrell

Thursday 22 Oct 2026, 16:00 → 17:00 Europe/Zurich

This Talk will take place in English exclusively on Zoom. Registration is not required to attend.

Abstract:

Testimony archives preserving survivor and witness accounts of genocide, mass violence, and crimes against humanity form an evidentiary backbone for historical memory, education, and international justice. Generative AI threatens this record on two fronts: fabricated testimony, voice cloning, and altered audiovisual material can contaminate the archive, while the "Liar's Dividend" enables authentic testimony to be dismissed as synthetic. Existing archival governance, which relies largely on static preservation standards and institutional reputation, was not designed for adversarial manipulation at scale. We propose the Testimony Archive Integrity Assessment Framework (TAIAF), an operational trust infrastructure that translates ethical commitments into five institutional safeguards: transparency through cryptographic provenance, consent as ongoing, open access with real-time monitoring, early detection, and documented accountability. 

Speakers:

Nathan Heath
Nathan Heath is an AI safety executive and decision scientist with over 14 years of experience across risk assessment, geopolitics, and AI safety. He currently serves as the Founder & CEO of Syntony, an AI safety startup translating red teaming evaluations into good governance. He is also the Co-Founder of The AIHL Project, a research initiative focused on assessing generative AI threats to protections under international humanitarian law. Nathan consults as a red teamer for OpenAI and Anthropic, and previously spent 6 years supporting the U.S. Department of Defense in areas such as emerging technology risk, security cooperation, and countering violent extremism. He received his M.A. in Law and Diplomacy from The Fletcher School.

Broderick James McDonald

Broderick James McDonald is a behavioural scientist and senior AI researcher at Oxford, King’s College London, and The Alan Turing Institute. Outside this, he is a Senior Advisor to Tech Against Terrorism and a Senior Fellow at Mila and FIG. His work examines AI safety and catastrophic risk from online harms and national-security threats, including disinformation, hybrid warfare, terrorism, CBRN, harmful manipulation, deepfakes, scams & fraud, child safety, and election interference. He has advised frontier AI labs, social media platforms, AI safety institutes (AISI), intelligence agencies, law enforcement, and prosecutors on emerging technology & global security threats. He previously held fellowships at the United Nations, University of Cambridge, and the Aspen Institute. His research & commentary have been featured in The New York Times, The Washington Post, and The Wall Street Journal, among others. He frequently appears on television broadcasters, including BBC, PBS, GMA, CBC, ABC News, France24, and Al Jazeera News.

Susan Nesbitt

Susan Nesbitt is Managing Director at Cantellus Group, where she advises governments, technology companies, foundations, and international organizations on the strategy, governance, and responsible deployment of agentic AI systems, helping leaders move from experimentation to enterprise adoption. She serves on the International Association for Safe and Ethical AI (IASEAI) Policy Committee, is a member of OpenAI’s External Red Teaming Network, advises AI Safety Connect, and contributes to international initiatives with UNESCO on AI governance and the ethical use of AI. Previously, Susan led Global Social Impact Products & Partnerships at WhatsApp and held leadership roles at the World Economic Forum, building global partnerships at the intersection of technology, policy, and societal impact.

Dr. Jennifer Victoria Scurrell
Dr. Jennifer Victoria Scurrell is a scientist working at the intersection of human-AI interaction, responsible AI, and pluralistic AI alignment. She is a postdoctoral researcher at the University of Zurich and serves as co-founder and CTO of an AI safety startup, while also conducting research on public preferences for AI alignment in Switzerland through a Public AI Fellowship. Her work spans AI governance and red teaming, including collaborations with the Geneva Human Rights Hub, UN’s OHCHR, Microsoft, and OpenAI's Red Teaming Network. Jennifer also bridges science and the arts through performance and conceptual art within various collectives and collaborations with the Schauspielhaus Zürich, bringing AI topics into public conversation. She received her PhD from ETH Zurich, where her research examined how AI chatbots influence human opinions and behavior.

The Zentralbibliothek Zurich, the CERN Scientific Information Service, and AILIS (Association of International Librarians and Information Specialists, Geneva) jointly organize the Library Science Talks. A programme of talks for 2025 can be found on the AILIS website.