Listen to the audio version of the article
US artificial intelligence companies OpenAI, Anthropic and Cohere have activated a secret channel of “diplomacy” with Chinese AI experts to discuss common concerns regarding the powerful technology. This was revealed by the Financial Times, which refers to several people familiar with the facts. In July and October last year, two meetings were held in Geneva in which scientists and political experts from American AI groups participated, together with representatives of Tsinghua University and other Chinese state institutions. Participants said the talks allowed both sides to discuss the risks of emerging technology and encourage investment in AI safety research. They added that the ultimate goal is to find a scientific path to safely develop more sophisticated Ai technology.
“There is no way to establish international standards on AI security and alignment without an agreement between this group of actors,” said a source interviewed by the Ft. “And if they agree, it is much easier to involve the others”. The talks, so far unreported, are a rare sign of Sino-US cooperation in a race for supremacy between the two great powers in cutting-edge technologies such as AI and quantum computing. Currently, Washington has blocked US exports of chips produced by companies such as Nvidia, necessary to develop sophisticated artificial intelligence software.
But the topic of AI safety has become a point of common interest between technology developers in both countries, given the potential existential risks to humanity. According to a negotiator present, who declined to be named, the Geneva meetings were organized after briefing the White House and British and Chinese government officials. “China supports efforts to discuss AI governance and develop necessary frameworks, norms and standards based on broad consensus,” the Chinese embassy in the UK said when asked by Ft.