AI Companies Pledge Fresh Safety Commitments at Seoul Summit

Leading artificial intelligence (AI) companies made a significant pledge during a mini-summit in Seoul on Tuesday. Their commitment? To develop AI technology safely. Simultaneously, world leaders agreed to establish a network of publicly backed safety institutes to advance research and testing of AI.

Google, Meta, and OpenAI were among the companies that voluntarily committed to safety measures at the AI Seoul Summit. These measures include a willingness to shut down their cutting-edge systems if they cannot effectively manage extreme risks. The two-day summit follows up on last November’s AI Safety Summit at Bletchley Park in the United Kingdom.

Leaders from 10 countries and the European Union aim to “forge a common understanding of AI safety and align their work on AI research,” according to a statement from the British government, which co-hosted the event. The network of safety institutes will include those already established by the U.K., U.S., Japan, and Singapore since the Bletchley meeting.

U.N. Secretary-General Antonio Guterres emphasized the urgency of addressing AI risks. He stated, “We cannot sleepwalk into a dystopian future where the power of AI is controlled by a few people — or worse, by algorithms beyond human understanding.” Guterres called for universal guardrails and regular dialogue on AI safety.

The 16 AI companies that signed the safety commitments also include Amazon, Microsoft, Samsung, IBM, xAI, France’s Mistral AI, China’s Zhipu.ai, and G42 of the United Arab Emirates. They pledged to ensure the safety of their most advanced AI models through accountable governance and public transparency.

While this isn’t the first time AI companies have made safety commitments, the Seoul meeting underscores the industry’s ongoing efforts to address risks associated with AI deployment. The safety pledge includes publishing frameworks that outline how companies will assess the risks of their models. In extreme cases, where risks are severe and “intolerable,” these companies will halt the development or deployment of their models.