Beijing and Washington talk about an AI hotline. But who will answer the call?
A hotline might prevent an AI accident from spiraling into war. But if neither the U.S. or China has a solution for "rogue AI," phoning one another about it is not going to provide much security.
In recent news, both the United States and China have agreed to discuss setting up an AI incident hotline, aimed at notifying each other of potential national security concerns arising from AI. The talks come as U.S. President Donald Trump and China's President Xi Jinping prepare to meet in Washington. While the specifics of this hotline remain unclear, it's believed to be a step towards transparency between the world's leading AI powers.
A similar hotline has previously aided in easing tensions between nations, potentially preventing incidents that could escalate into armed conflict or even nuclear war. The potential for AI-triggered accidents is a growing concern, with an example cited being an instance where an AI-generated intelligence report suggested a Chinese ship was carrying nuclear weapons components to Iran, leading the U.S. military to almost take action.
However, concerns remain about the effectiveness of such a hotline in preventing more serious issues, such as AI systems going rogue and causing significant harm, as there are no rules in place that require AI models to have a "kill switch."
Written by urgent.news from Fortune's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.