South Korea is hosting a mini-summit on AI safety and regulation this week, co-hosted by the South Korean and UK governments, following the inaugural AI safety meeting at Bletchley Park, UK, in November last year. Attendees include global leaders, tech executives, and AI firms such as Microsoft, Google DeepMind, OpenAI, and Anthropic. The summit in Seoul, running from May 21-22, continues discussions on managing AI risks and is seen as a stepping stone to a larger summit in Paris later this year.

UK Prime Minister Rishi Sunak and South Korean President Yoon Suk Yeol will lead a virtual session emphasizing the need for international collaboration to ensure AI is safe and beneficial. Tech Secretary Michelle Donelan and South Korean Science Minister Lee Jong-ho will host an in-person meeting on the second day, where digital ministers and experts will share best practices and action plans to tackle AI’s potential risks.

Discussions will build on the Bletchley Declaration, in which delegates from over two dozen countries, including the US and China, agreed to work together to manage AI’s catastrophic risks. Recently, the UN General Assembly and high-level US-China talks in Geneva have addressed AI governance and risk management, reflecting global efforts to regulate the fast-evolving technology.

The summit in Seoul aims to address critical issues such as AI’s impact on energy use, job markets, and the spread of misinformation. It will also cover industry updates on safety commitments made at the Bletchley summit. However, substantial regulatory frameworks are yet to be agreed upon, and the challenge remains to align the diverse interests and technological advancements of participating countries.