AI Seoul Summit Issues Global Declaration on Frontier AI Safety — May 21, 2024
Leaders from 16 governments endorsed the Seoul Declaration on safe, inclusive, and innovative AI at the AI Seoul Summit, committing to coordinated safety evaluations and international standards for advanced models.
The declaration follows the 2023 Bletchley Summit and outlines shared principles for managing frontier AI risks, promoting inclusive benefits, and accelerating science-based safety collaboration. Governments also issued a Statement of Intent to create an international network of AI safety institutes that will exchange methodologies and jointly test high-risk models.
- Frontier AI safeguards. Signatories agreed to advance evaluations, watermarking research, and incident reporting.
- Inclusive innovation. The declaration emphasises supporting developing countries through capacity building and equitable access to compute and data.
- Institutional cooperation. Safety institutes will coordinate research agendas, share findings, and explore joint experiments.
AI teams should align global safety narratives and anticipate interoperability requirements emerging from this new international network.