UN Talks on AI Weapons End Without Binding Rules

Published: November 21, 2026, 9:02 am

Efforts to establish firm international regulations on AI-powered lethal autonomous weapons systems concluded in Geneva this Saturday with a softened agreement. Rather than committing to the binding rules sought by U.N. Secretary-General António Guterres and various humanitarian groups, the participating nations agreed to extend their discussions for another two years.

The week-long U.N. conference on weapons ended with a commitment to explore a potential accord on such technology by 2028. This outcome highlighted the significant challenges of achieving diplomatic consensus in an environment marked by global conflicts and competition between major military powers. Activist organizations, including Stop Killer Robots, criticized the result as a "watered-down mandate," while Beatrice Fihn, founder of the advocacy group Lex International, described the decision as merely "pushing the can down the road for two more years."

Prior to the conference, seven countries—Brazil, Austria, Belgium, Ireland, Mexico, New Zealand, and Panama—proposed the launch of formal negotiations for a rulebook intended to limit or prevent machines from executing lethal force without human accountability. This proposal did not gain sufficient traction. Ambassador Frederico Meyer of Brazil noted that the pace of technological advancement is outstripping diplomacy, remarking that "diplomats are very slow."

Advocacy groups pointed to obstructionism by key military powers, including Russia and the United States, which they suggest are seeking to maintain a freer hand in the development of high-tech weaponry. Max Tegmark, a machine learning specialist at the Massachusetts Institute of Technology and founder of the Future of Life Institute, criticized the outcome, stating that states have agreed it is sufficient to continue talking about the issue for a 14th year while these weapons continue to enable atrocities.

The push for regulation had been framed as an existential debate by many humanitarian organizations, who urged for immediate action to prevent AI weapons from becoming too widespread to regulate effectively. While activists report that armed drones capable of independent target strikes have already been deployed in select conflict zones, a widespread global rollout has not yet occurred. Secretary-General Guterres, who is set to leave office at the end of this year, had previously pushed for a 2026 deadline to initiate formal negotiations.

An armed robot developed by Turkish defense companies is demonstrated at a facility in Duzce, Turkey, on April 29. Omer Urer/Anadolu via Getty Images hide caption

Photo: Collected