Skip to main content
Live
Main content

UN talks on AI killer robots end without binding treaty mandate, set 2028 deadline

128 CCW states agreed to elaborate an instrument on lethal autonomous weapons within two years, but almost 100 countries wanted binding-ban talks now.

Jaeden Schafer
Editor in Chief · · 5 min read
UN talks on AI killer robots end without binding treaty mandate, set 2028 deadline

UN talks on lethal autonomous weapons ended early Saturday in Geneva with the 128 states party to the Convention on Certain Conventional Weapons agreeing to launch a two-year process to elaborate an international instrument on so-called killer robots. The conference stopped short of a mandate to negotiate a legally binding treaty, the outcome most participating states had pushed for.

The five-day Seventh Review Conference of the CCW wrapped up at around 4:45 a.m. Saturday after several hours of overnight negotiations. Under the deal, a group of governmental experts will recommend the legal nature and form of the instrument, including whether it should become a binding protocol to the convention, by 2028.

Lethal autonomous weapons systems are weapons that, once activated, can select and attack targets without human intervention. The Stop Killer Robots coalition says such systems have already been tested or deployed, including in the war in Ukraine.

Key facts

  • 01The 128 states party to the UN Convention on Certain Conventional Weapons agreed on Nov. 21 to elaborate an instrument on lethal autonomous weapons.
  • 02Governmental experts will recommend by 2028 whether the instrument should be a legally binding protocol.
  • 03Almost 100 countries sought negotiations on a binding ban; the US, Israel, Belarus and Britain opposed a specific mandate, per dpa.
  • 04The ICRC cited a First Committee resolution backed by 163 states calling for action on autonomous weapons.
  • 05The five-day Seventh Review Conference ended around 4:45 a.m. Saturday after overnight talks in Geneva.

The Swiss ambassador who presided over the conference told reporters that states had "agreed to launch the elaboration of an international instrument on lethal autonomous weapons systems" after more than 10 years of discussion. He said the process will determine which weapons can be forbidden under international humanitarian law and which can be regulated, after which the parties will decide whether the instrument should be a treaty and how binding it should be.

Izumi Nakamitsu, the UN High Representative for Disarmament Affairs, called the agreement a "significant achievement" and urged countries to pursue the strongest possible instrument to prohibit and regulate the weapons. She said procedural disagreements also kept the conference from reviewing how the convention itself is being implemented.

“We at least have a mandate for high contracting parties to elaborate within two years.”
— Izumi Nakamitsu, UN High Representative for Disarmament Affairs

The split was over how fast to move toward binding rules. Almost 100 countries called for negotiations on a legally binding ban on fully autonomous weapons, according to the dpa news agency, while Stop Killer Robots put the number of states backing a treaty at more than 95. The United States, which dpa described as a leading military power and a pioneer in AI development, was among the countries opposing a specific negotiating mandate, along with Israel, Belarus and Britain. Brazilian Ambassador Frederico Meyer summed up the dynamic: "some countries have the power, some countries have the majority."

Stop Killer Robots, which says it represents more than 300 organizations, said states "adopted a watered-down mandate that fails to chart a clear path towards a legally binding instrument," and that a small number of states weakened the text in the final hours. Verity Coyle, deputy director for arms at Human Rights Watch, said the outcome does not provide the urgent action needed to agree new rules while the technology keeps advancing.

The International Committee of the Red Cross said it regrets that the conference failed to agree to a clear and immediate negotiation mandate. The ICRC pointed to a resolution backed by 163 states in the UN General Assembly's First Committee last month and urged states to adopt a binding instrument that prohibits unpredictable autonomous weapons and systems that directly target humans, while restricting other autonomous weapons.

“It remains possible -- likely, even -- that there will be no treaty by the end of 2028.”
— Rain Liivoja, Director, Asia-Pacific Institute for Law and Security
Related · from this week
UN panel urges AI safeguards now, citing precautionary principle on agent risks
Jaeden Schafer · 5 min read →

UN Secretary-General Antonio Guterres had sought a legally binding instrument negotiated and concluded by the end of 2026. The 2028 timeline for expert recommendations moves any decision on a treaty well past that target.

Nakamitsu described the two-year window as a "very compressed timeframe" and said the possibility of a legally binding instrument remains open.

ShareXLinkedInEmail
AI Box

Every AI model. One chat.

The latest models from ChatGPT, Claude, Gemini, Sora, ElevenLabs — 80+ models in a single chat. Compare answers side by side. Pick the best one every time.

  • ChatGPT, Claude, Gemini, Grok, DeepSeek — in one chat
  • Generate images & video with Sora, Veo, Ideogram
  • Compare any two models side by side
  • From $8.99/mo · 80+ models, all included
Try AI Boxaibox.ai
Trusted by 3,000+ teams
Got a tip?

Working on something we should cover, or seeing a story we missed? Send leads, documents, or feedback to hello@aichatdaily.com. For sensitive tips, see our secure tips page for Signal and PGP options.

Spotted an error? Email hello@aichatdaily.com with the URL and the issue, or read our full corrections policy.

AI Box Daily briefingFree · Daily · No fluff

Stay ahead of everyone in AI.

The tightly edited AI news email engineers, founders, and investors actually open. One email. Every weekday. Five minutes to finish.

Loved by 10,000+ AI professionals
Free forever. Unsubscribe with one click.

The briefing read inside teams at

Keep reading

More from Security

UN panel urges AI safeguards now, citing precautionary principle on agent risks
Security

UN panel urges AI safeguards now, citing precautionary principle on agent risks

The UN's first scientific brief on AI says governments shouldn't wait for certainty before reining in autonomous agents from OpenAI, Anthropic, Google and Meta.

Jaeden Schafer5 min read
UN chief Guterres warns AI is outpacing oversight, calls for global rules to protect children
Security

UN chief Guterres warns AI is outpacing oversight, calls for global rules to protect children

The Secretary-General says AI governance is failing to keep pace with deployment and wants binding international rules focused on child safety.

Jaeden Schafer4 min read
Google logo
Models

UN partners with Google to make its statistics AI-agent ready

A UNICEF benchmark found six top models averaged 21.2% accuracy on development data; the new Data Commons aims to fix that.

Jaeden Schafer5 min read