Data serves as the foundational resource for artificial intelligence (AI). While demand for vast, high-quality datasets grows in tandem with AI’s integration across domains, issues over data availability and scarcity have come to the fore, limiting the potential of “data-centric AI”. Synthetic data – artificially generated data that mimics real-world data – has been touted as a potential solution to address these solutions.
Synthetic data holds particular promise in the military domain, where data scarcity is acute, but where data is in increasing demand as interest in AI-enabled military capabilities grow. However, uncertainty remains over the security and governance implications of the use of synthetic data. As AI governance in the civilian and military domains takes shape, the question of data should be central. Given the potential security implications of synthetic data and the any governance and regulatory questions surrounding its use, it is paramount that we achieve a clearer understanding of how synthetic data fits into ongoing discussions and the wider data governance toolbox.
In this context, UNIDIR’s Security and Technology Programme, with the support of the European Union, organized a Technology and Security Seminar on synthetic data to explore the implications of this important enabling technology for international security and governance. This half-day event consisted of a Technology Breakfast, serving as an introduction to the technology for policy makers, as well as a Multistakeholder Dialogue on synthetic data where experts from industry, government, international organizations and academia convened to share views on the technology’s governance challenges and potential policy actions.
Event recording
Agenda
9:00-9:10: Introductory remarks
9:10-10:10: Technology breakfast on synthetic data and international security
- Dr Eleonore Fournier-Tombs, Head of Anticipatory Action and Innovation, Centre for Policy Research, United Nations University
- Calum Inverarity, Senior Researcher, Open Data Institute
Moderated by Wenting He, Associate Researcher, Security and Technology Programme, UNIDIR
10:10-10:30: Break
10:30-12:00: Multistakeholder dialogue on synthetic data: What opportunities and challenges for international governance
- Dr Ana Beduschi, Full Professor of Law with a Personal Chair at the University of Exeter, Director, Research Centre for Science, Culture and the Law at the University of Exeter Law School
- Aldo Lamberti, Founder and CEO, Syntheticus, Subject Matter Expert, European Commission, Vice-Chair, Standard for Security and Trustworthiness Requirements in Generative Pretrained AI Models, IEEE
- Dr Jane Pinelis, Chief AI Engineer of the Applied Information Sciences Branch, Johns Hopkins University’s Applied Physics Laboratory
- Yasmin Afina, Researcher, Security and Technology Programme, UNIDIR, Expert, Global Commission on Responsible AI in the Military Domain
Moderated by Federico Mantellassi, Researcher, Security and Technology Programme, UNIDIR
When and where
Tuesday, 29 October, 9:00-12:00 EDT (14:00-17:00 CET), online.
In partnership with
The European Union
