US and China Prepare AI Safety Talks
The United States and China are preparing to hold talks in mid-September on artificial intelligence safety risks. On the US side, the dialogue is expected to be led by US Treasury Secretary Scott Bessent, Reuters sources familiar with the planning said, reports Asharq Al-Awsat.
At the same time, a White House representative said that no meeting related to artificial intelligence is currently scheduled for mid-September. Reuters sources also cautioned that the final agenda, list of participants and venue of the potential talks could still change.
Likely topics of the talks
Washington wants to discuss cooperation in monitoring AI-enabled cyberattacks, one source said. The US side also proposed that AI laboratories in the United States and China independently track risks and share information to prevent cyberattacks related to artificial intelligence.
According to the source, the United States plans to raise the issue of possible Chinese Mythos-level models capable of carrying out cyberattacks, as well as claims concerning the distillation of US proprietary AI models. In June, White House science and technology adviser Michael Kratsios accused China’s Moonshot AI of stealing data from Anthropic’s Fable model to create the K3 release. Distillation means training smaller models on the outputs of larger and more expensive models.
More current news is available on the UA.News Telegram channel Telegram.
Possible participants
According to a second Reuters source, the Chinese side may be led by Vice Premier He Lifeng, who is Bessent’s protocol counterpart. An alternative could be Politburo Standing Committee member Ding Xuexiang, who coordinates China’s technology policy, including in the fields of AI and semiconductors.
Kratsios and Chinese Minister of Science and Technology Yin Hejun were also named among the possible participants. US President Donald Trump and Chinese President Xi Jinping are due to meet on September 24 at a summit in Washington. Chinese officials, Reuters sources said, view dialogue on AI as one of the key outcomes of the summit.
Last week, the sides also discussed safeguards for artificial intelligence during a Track 1.5 dialogue in Beijing. Former Microsoft executive Craig Mundie, co-chair of the unofficial US-China Track II dialogue on AI, is facilitating informal contacts on the US proposals, according to a source.
