During the virtual AI Seoul Summit, 16 AI companies from around the world signed the Frontier AI Safety Commitments, pledging to publish safety frameworks to measure risks of AI models, coordinate with stakeholders to define risk thresholds, and halt development if risks are deemed intolerable, marking progress towards a global AI safety governance regime involving major tech giants like Amazon, Google, and Microsoft, with these commitments building on previous agreements to promote transparency and responsible AI development while also drawing some criticism for being seen as potentially mere 'empty promises' lacking clear restrictions and regulations.
https://www.infosecurity-magazine.com/news/ai-seoul-summit-safety-commitments/