For the first time ever, a Chinese AI company will speak directly to the United Nations Security Council about the dangers of artificial intelligence. According to two sources familiar with the matter, DeepSeek UN Security Council talks are set for this Wednesday, September 23, putting a major Chinese AI lab in the same room as top American AI companies like OpenAI and Anthropic. Here is what’s happening, explained simply.
What’s Actually Happening
The UN Security Council, which has 15 member countries, is meeting on Wednesday to discuss AI and how it affects global security. DeepSeek, along with another Chinese AI company called Moonshot, has been invited to make statements at this meeting. This is happening during the same week world leaders are gathered in New York for the annual UN General Assembly.
OpenAI’s CEO, Sam Altman, is also planning to brief the meeting. Diplomats expect senior people from Anthropic to take part too. This means, for the first time, major AI companies from both the US and China will be speaking in the same UN meeting about AI risks.
Why This Meeting Is Happening Now
This meeting comes just weeks after several AI industry leaders publicly called for a slower, more careful approach to building increasingly powerful AI systems. Their warning was serious: they said these systems could soon start improving themselves and eventually slip beyond human control.
UN Secretary-General Antonio Guterres also warned world leaders last week that fast-moving AI development creates risks that simply cannot be ignored. Interestingly, this view puts him at odds with US President Donald Trump, who has argued that current safety rules are already enough, and who has suggested that China would actually benefit if doubt were cast on AI development, since it might slow down US progress.
Will DeepSeek’s Founder Actually Show Up?
One detail worth knowing: according to one source, DeepSeek’s founder, Liang Wenfeng, does not currently plan to attend in person. The source also cautioned that these plans are still flexible and could change at the last minute, so it’s not fully certain who exactly will represent DeepSeek at the meeting.
Not the First Time — But Different This Time
This isn’t the UN Security Council’s first conversation about AI. It first discussed AI risks back in 2023. At that meeting, China warned that AI technology should not become a “runaway horse” — meaning something that moves too fast to control. The US, on the other hand, warned against AI being used to censor or repress people.
What makes this week’s meeting different is the direct presence of competing AI labs from both countries in the same room. That raises an interesting question: can rivals from the US and China even agree on how to describe the danger of AI, let alone agree on real limits for it?
Does This Meeting Actually Change Anything?
It’s important to set realistic expectations here. The UN Security Council doesn’t currently have any specific enforcement power over AI, and its five permanent members often hold very different positions on many issues. The 2023 meeting, for example, led to broad recognition that AI carries risks, but very little actual follow-up action. Based on that pattern, this week’s meeting is likely to produce more discussion and public statements than binding rules or enforceable outcomes.
Bigger Picture: US-China AI Competition
This UN meeting is happening alongside other major US-China developments. Officials from both countries are also preparing for a meeting between President Trump and Chinese President Xi Jinping in Washington, where AI is expected to be one of the topics discussed. Chinese state media have already dismissed calls for slowing down AI development, describing them as part of a “Cold War playbook” aimed at protecting America’s lead in advanced technology.
Looking further ahead, this UN process is still developing. The next major UN discussion on this topic, called the Global Dialogue on AI Governance, is already scheduled for New York in May 2027.
Frequently Asked Questions
When will DeepSeek brief the UN Security Council?
The meeting is expected to take place on Wednesday, September 23, 2026, during the week of the UN General Assembly in New York.
Which other AI companies are involved?
OpenAI CEO Sam Altman plans to brief the meeting, senior representatives from Anthropic are also expected to attend, and another Chinese AI firm, Moonshot, has been invited alongside DeepSeek.
Will DeepSeek’s founder attend in person?
Based on current information, Liang Wenfeng does not plan to attend, though sources say these plans could still change.
Can the UN Security Council actually control AI development?
Not directly. The Council has no specific enforcement power over AI, and given how differently its permanent members view many issues, this meeting is more likely to produce discussion and public statements rather than binding rules.
This week’s meeting marks a genuinely notable moment — the first time major AI labs from both the US and China will address the UN Security Council together on AI risk. Whether it leads to real cooperation or simply highlights how far apart the two sides remain is something the world will be watching closely on Wednesday.
