Government experts are convening in Geneva this week to advance talks on a possible negotiating mandate for autonomous weapons systems (AWS)—military technologies capable of selecting and engaging targets without human intervention. The meeting follows a joint appeal by the United Nations and the International Committee of the Red Cross (ICRC) for urgent, legally binding controls on these systems, often described in media as “killer robots.”
Recent events in Ukraine have sharpened concerns that the slide toward fully autonomous weapons—operating without meaningful human control—is already underway. According to the New York Times, a Russian drone attacked a petrol station in Zaporizhzhia on July 6, killing three civilians. What distinguished the attack was the drone’s use of artificial intelligence to decide where and when to strike. The aircraft was programmed to fly toward the station, but on approach, an AI system autonomously selected the precise target—likely propane tanks—based on its ability to recognize them.
On August 25, Mirjana Spoljaric, president of the ICRC, and António Guterres, UN secretary-general, issued a joint appeal for the international community to agree on binding rules governing the development and use of AWS. New technologies are making it cheaper and easier for nations to assemble such weapons, lending urgency to their call. The international community is well-positioned to begin negotiations soon, building on years of diplomatic groundwork under the Convention on Certain Conventional Weapons (CCW), a UN arms control framework.
Since 2017, the Group of Governmental Experts (GGE) on Lethal Autonomous Weapons Systems (LAWS) has met typically twice a year under the CCW. This week’s session aims to agree on a governance framework and language for a rolling text that could serve as the foundation for negotiations, with further discussions scheduled for November. While the process has been marked by political disagreements, most experts concur on the need to retain some level of meaningful human control over autonomous systems.
Moral and legal stakes
The shared sense among participating states that human judgment is pivotal reflects the moral significance of warfare and the importance of human responsibility. Human judgment is also embedded in international humanitarian law (IHL). Targeting processes require context-specific assessments about the target, potential civilian harm, and other precautions—tasks that cannot be delegated to machines.
AI-enabled systems already shape human perceptions of targets and risk. With autonomous weapons, the system shifts from influencing human judgment to exercising the targeting function itself. Any AI system is only as good as the data it has been trained on or works with. In armed conflicts, such data is scarce and often insufficient for dynamic, variable human contexts, meaning many systems are likely to be technically inadequate.
This problem becomes visible in rules around proportionality, which require a context-specific judgment about anticipated civilian harm versus military advantage. Militaries may try to quantify targeting decisions to create an appearance of precision, but the underlying legal judgment remains qualitative. Technical accuracy is not the same as lawfulness or moral consideration. Only humans can understand the significance of respecting the inherent rights and dignity of other humans. Decisions to end human life must therefore be made qualitatively, by humans, to be morally and legally justifiable.
Without such provisions and clear guidelines, governments risk falling into a legal vacuum and a steep moral abyss. AWS, especially those that prioritize speed and scale through AI, carry a significant risk of enabling uncontrolled and rapid escalation. Escalation from crisis to war, or from conflict to higher levels of violence, could occur due to erroneous indications of attack or simple sensor or computer errors. Unpredictable systems could give leaders false impressions of their capabilities, leading to overconfidence or encouraging preemptive strikes.
Inevitable or necessary?
Some academic researchers argue that AWS become necessary because they are inevitable. The risk is that more countries opt for full autonomy as components become cheaper and more readily available, exacerbating an arms race. However, an adversary’s conduct does not release a state from its own obligations under IHL. Following those rules is also part of what makes military force legitimate, including through civilian protection, precautions, and accountability.
We might turn the argument around: legally binding rules become inevitable because they are absolutely necessary. Governments and advocacy groups have worked hard over the past decade to craft a lucid proposal for regulating AWS. Some efforts aim to prohibit systems that target humans directly or that are unpredictable and thus cannot be meaningfully controlled by a human. All other AWS must be strictly regulated.
There are signs of political momentum. In November 2025, 156 states voted in favor of a UN resolution calling for negotiations on a legally binding instrument. The Geneva talks this week are a critical step toward turning that momentum into concrete rules. As the use of AI in conflict grows—from drones in Ukraine to directed-energy weapons on Chinese corvettes—the need for clear, enforceable standards has never been more urgent. The international community must act decisively to ensure that human judgment remains at the heart of warfare, not as a luxury but as a legal and moral necessity.
