SEOUL, South Korea — The AI Safety Summit Seoul brought world leaders, technology executives, and digital ministers together on May 21–22, 2024, as South Korea and the United Kingdom co-chaired a two-day conference aimed at advancing the global dialogue on AI risks and regulation. The event built directly on the first-ever AI Safety Summit held at Bletchley Park in the UK the previous November.
AI Safety Initiatives and Actions Around the World
The Seoul summit is one of a growing series of international efforts to develop AI regulation that addresses risks such as biased algorithms, labour market disruption, and the misuse of advanced AI systems. In November 2023, Bletchley Park in the United Kingdom brought together technology giants and government representatives to tackle AI safety for the first time at this scale. Leaders of more than 20 nations — including the United States and China — endorsed the Bletchley Declaration, pledging to regulate and mitigate risks posed by frontier AI.
That momentum continued in March 2024, when the United Nations General Assembly passed a resolution endorsing efforts to promote and regulate AI in a way that benefits all member states and upholds human rights. Late in 2023, the US and China also convened a senior-officials meeting in Geneva to discuss AI risks and identify common management approaches.
Seoul Summit: The Plan
The two-day event was co-chaired by the Republic of Korea and the United Kingdom on May 21–22. On the first day, South Korean President Yoon Suk-yeol and UK Prime Minister Rishi Sunak met with AI company leaders to discuss the development of safety measures. Day two brought together digital ministers from the United States, China, Germany, France, Spain, and other nations alongside AI industry leaders from OpenAI, Google, Microsoft, and Anthropic. Discussions centred on concrete measures to address the negative social impacts of AI — from energy consumption and employment to information integrity.
Key Outcomes and Agreements
The summit produced three major international documents: the Seoul Declaration, the Frontier AI Safety Commitments, and the Seoul Ministerial Statement. Ten countries and the European Union signed the Seoul Declaration on May 21, while 27 countries and the EU signed the broader Seoul Ministerial Statement on May 22.
Sixteen AI tech companies — including OpenAI, Google, Microsoft, and Anthropic, as well as China's Zhipu.ai and the UAE's Technology Innovation Institute — signed commitments to develop AI safely. Japan, South Korea, and Canada also announced plans to establish their own national AI safety institutes, with the European Commission AI Office expected to serve a similar function for the EU.
Korean and UK organisers also secured a statement of intent signed by 10 countries plus the EU for AI safety institutes to cooperate as a formal international network — marking a significant moment of multilateral coordination on frontier AI risk.
AI Safety: Current Challenges and Future Direction
Critics noted that the first Bletchley agreement lacked concrete regulatory measures. Lee Seong-yeob of Korea University pointed to the difficulty of uniting nations with competing national interests and differing stages of AI development. Nevertheless, policymakers and technical experts at the summit expressed determination to agree on common safety standards.
An expert panel's interim report highlighted key threats: the misuse of AI for fraud and propaganda, and systemic risks to labour markets. South Korea sees the summit as a springboard to assume a leading role in global AI governance — even as analysts note that its regulatory capabilities are still being developed.
The Seoul Ministerial Statement emphasised transparency, accountability, robust risk management, and ensuring equitable AI benefits while enhancing digital literacy worldwide.
