Estimated read time: 3-4 minutes
- U.N. talks on AI killer robots extended two years without binding commitments.
- Activists blame military powers like Russia and the U.S. for blocking progress.
- U.N.'s Nakamitsu urges immediate work on regulating lethal autonomous weapons systems.
GENEVA — Efforts to set firm rules on AI-powered killer robots and drones ran aground Saturday as U.N.-backed talks in Geneva produced a softened agreement to extend discussions for two more years.
Scores of countries ended the weeklong U.N. conference on weapons by agreeing to "elaborate" a possible accord on such lethal autonomous weapons systems by 2028. But they didn't commit to binding rules sought by U.N. Secretary-General António Guterres, many countries and activists.
The outcome underscored the difficulty of achieving concrete results in diplomacy amid global conflicts as major military powers jockey for dominance. It also marked the latest failure to set guardrails for AI, at a time when some experts have warned about such systems breaking out of human control and going rogue.
Izumi Nakamitsu, the U.N. high representative for disarmament affairs, hailed a "significant achievement" but acknowledged a "glass half full or half empty" scenario, citing Guterres' hopes for a legally binding deal this year.
Activists accused military powers like Russia and the United States of blocking stronger language.
Stop Killer Robots lamented a "watered-down mandate," and Beatrice Fihn, founder of international law advocacy group Lex International, said the decision amounted to "pushing the can down the road for two more years."
They noted that most countries supported work toward a new treaty.
Max Tegmark, a machine learning specialist at the Massachusetts Institute of Technology and founder of the Future of Life Institute, said: "States have agreed it is sufficient to just talk about the issue, for the 14th year, while these weapons enable atrocities."
Before the meeting, seven countries proposed launching negotiations on a rulebook that could prevent or limit the ability of machines to kill without human control or accountability. That effort fell short.
"Technology is moving faster than diplomacy," said Ambassador Frederico Meyer of Brazil, which joined Austria, Belgium, Ireland, Mexico, New Zealand and Panama in proposing such negotiations. "Diplomats are very slow."
Guterres, who leaves office at the end of the year, had set a 2026 deadline for launching negotiations toward an agreement on autonomous weapons systems.
The U.N.'s Nakamitsu called for countries to begin work as early as Monday and "start thinking about how to get as strong as possible instrument to prohibit and regulate lethal autonomous weapon systems."
The International Committee of the Red Cross highlighted how autonomous weapons can be "unpredictable" — unlike, say, a bullet fired from a gun with a distinct and predictable trajectory.
"For years, the ICRC and many others have urged states to ensure that humans retain control over the use of force and that life-and-death decisions are never handed over to machines," it said in a statement.
Advocacy groups bemoaned obstructionism at this week's talks largely by Russia, which, like other leading developers of AI weapons, including the United States, wants a freer hand to develop such high-tech weapons.
Pope Leo XIV and humanitarian groups have framed the issue as an existential debate, urging quick action to rein in AI weapons before they become too commonplace to regulate easily.
Activists say armed drones that can independently strike targets have already been deployed in some select instances in conflict zones, but widespread rollout hasn't happened.
Talks this week centered on the Convention on Certain Conventional Weapons, ratified by 128 countries, where decisions are made by consensus. That means any single country can thwart an agreement, and any deal is likely to be based on the lowest common denominator.








