DeepSeek Invited to Brief UN Security Council on AI Safety Risks

Citing people familiar with the matter, Reuters reports that Chinese AI startup DeepSeek will attend a UN Security Council meeting this week to speak about AI safety risks. World leaders are gathering in New York for the annual UN General Assembly at the same time.

The 15-member Security Council plans to meet on Wednesday to discuss AI and international security. Chinese AI companies including DeepSeek and Moonshot AI have been invited to speak. DeepSeek founder Liang Wenfeng does not plan to attend, but arrangements could still change.

The Security Council held its first AI risk meeting in 2023. UN Secretary-General Antonio Guterres warned leaders that the risks from fast-moving AI cannot be ignored, while the US President said existing safeguards are sufficient.