AI Safety Summit 2023
- November 1-2 2023
AI Safety Summit
The AI Safety Summits are bi-annual (approximately every six months) international meetings aimed at discussing the safety, governance and regulation of advanced AI systems — especially so-called “frontier” AI.
Agenda & Discussion Streams
The summit was structured around several roundtables and sessions focusing on different aspects of AI safety. Detailed summaries for each roundtable were published.
Purpose:
To bring together governments, leading AI companies, civil society and research institutes to address the risks of “frontier AI” (i.e., the most advanced AI systems) and discuss international cooperation on AI safety.
Key Documents & Declarations
- Bletchley Declaration – The main joint declaration signed by participating countries. It affirms commitment to human-centric, trustworthy AI, international cooperation, evaluation and testing of frontier AI, and inclusive benefits.
- Chair’s Statement / Chair’s Summary – Published 2 Nov 2023 by the UK as host, summarising discussions and next steps.
- Roundtable Chairs’ Summaries – Detailed breakdowns of what was discussed in each roundtable (Day 1 and Day 2).
Hosted by
The UK government (via the Foreign, Commonwealth & Development Office, the Department for Science, Innovation and Technology and others).
Summit Information
About the Summit
Key roundtables included:
- Risks to Global Safety from Frontier AI Misuse
- Risks from Loss of Control over Frontier AI
- Risks from the Integration of Frontier AI into Society
- What should frontier AI developers do to scale capability responsibly
- Focused on what the international community and the scientific community should do respectively.