US and China snub military AI agreement at global summit

In A Coruña, Spain, officials from 85 countries met this week at the Responsible AI in the Military Domain (REAIM) summit to discuss how AI should be used, and limited, in conflict settings.

A proposed 20-point declaration was on the table, outlining principles such as human oversight of AI-enabled weapons, clear command chains, and robust risk assessments and training for personnel using military AI systems.

But, in a sign of deep geopolitical tension, two of the biggest military and technological powers opted out. Neither the US nor China added their names to the declaration, despite heavy diplomatic participation at the event.

Delegates framed the decision through the lens of strategic competition. Officials from several countries said that uncertainty in how transatlantic alliances will evolve and the desire not to weaken national competitive edges were key factors behind the reluctance to commit.

As Dutch defence minister Ruben Brekelmans put it, countries are caught in what he described as a “prisoner’s dilemma”: governments want to promote responsible AI use but also fear falling behind rivals in developing advanced systems.

“Russia and China are moving very fast. That creates urgency to make progress in developing AI. But seeing it going fast also increases the urgency to keep working on its responsible use. The two go hand-in-hand,” he told Reuters.

Just 35 of the 85 attending nations signed the declaration. Among them were Canada, Germany, France, Britain, the Netherlands, South Korea and Ukraine, all of which backed the call for shared principles on responsible military AI use.

AI has reached ‘biggest transition’ point ever, tech leaders say