Meaningful human control

Meaningful human control is the framing the Stop Killer Robots coalition has built its decade-plus treaty advocacy around: that any use of force must remain under human control sufficient to preserve responsibility and accountability, and that weapons systems which select and engage targets without such control should be prohibited under new international law. The phrase has become the corpus's clearest example of an organising term that travelled from a single NGO memorandum into UN documents, state position papers, and the working vocabulary of the coalition that now has a UN Secretary-General report on record calling for treaty negotiations to conclude by 2026.

Origin

The term was coined by Richard Moyes, head and Managing Partner of the UK humanitarian-disarmament NGO Article 36, in 2013. Moyes set out the underlying argument — that human control over individual attacks is the appropriate frame for autonomous-weapons advocacy — in Article 36's April 2013 policy paper Killer Robots: UK Government Policy on Fully Autonomous Weapons, published in direct response to the UK Ministry of Defence's "human-in-the-loop" / "human-machine interaction" position. The phrase itself was deployed as a campaigning term in a November 2013 memorandum to delegates at the Convention on Certain Conventional Weapons (CCW) in Geneva, titled "Structuring Debate on Autonomous Weapon Systems". The Stop Killer Robots coalition, which Article 36 co-founded the previous year, was the immediate vehicle for its propagation.

The framing was deliberately conceived as what Moyes himself called a political concept — a term sufficiently open to give states a shared vocabulary to negotiate around, rather than a tight legal definition that would foreclose discussion. Article 36's own April 2016 briefing Key Elements of Meaningful Human Control named the substantive content the campaign was filling the term with: that "meaningful" required adequate contextual information about the target area, a positive action initiated by a human operator, and a clear line of accountability for the human individuals involved in the planning and execution of the attack.

Propagation

The term moved fast. Human Rights Watch's 11 April 2016 briefing recorded that the count of states addressing the concept of human control in CCW statements had grown from approximately nine in 2014, to twenty-seven in 2015, to almost thirty by April 2016, with the International Committee of the Red Cross concluding that there was "broad agreement among States on the need to retain human control over the critical functions of weapon systems." Stop Killer Robots' own 11 April 2016 press release opened the third CCW meeting on killer robots with the framing as its public-facing focus, and explicitly attributed the term to Article 36's 2013 memo.

From 2017 onward, Article 36's own follow-on work extended the framing into the question of how meaningful human control could be operationalised within national weapon-review processes — and argued that those processes alone would be insufficient, that a new international legal instrument was needed to anchor the requirement. Moyes' sustained publication record — including the 2016 CCW briefings, a chapter in Robin Geiß's edited volume on lethal autonomous weapons systems, an ICRC expert-meeting contribution titled "Meaningful human control over individual attacks", and a 2017 weapon-review discussion paper — kept the term anchored in concrete substantive proposals throughout the GGE on LAWS years.

By the time the coalition's 2023 push to move the autonomous-weapons file from the consensus-blocked CCW track to the UN General Assembly succeeded, the framing had become the campaign's central demand. Stop Killer Robots' About page now states the coalition's core ask as "to demand meaningful human control, which ensures responsibility and accountability, in any use of force"; its Vision and Values page frames the campaign as working "for an international legal instrument that prohibits machines that determine whom to kill and requires meaningful human control over the use of force". The framing now structures the coalition's two-tier treaty model: prohibitions on weapons that target people or that inherently lack meaningful human control, and positive regulations on all other autonomous weapons to ensure that such control is preserved.

Adoption at the UN

The framing's most consequential travel has been into UN-system documents. The UN Secretary-General's August 2024 report — produced in response to the November 2023 UN General Assembly First Committee Resolution L.56 — called on states to conclude treaty negotiations by 2026 that would prohibit weapons that "operate without human control" or that cannot be used in compliance with international humanitarian law, and regulate all other autonomous weapons to preserve such control. Both halves of the meaningful-human-control framing — the prohibition tier and the regulation tier — are reproduced in the SG's text. At the first UN General Assembly informal consultations on autonomous weapons in May 2025, Secretary-General António Guterres put the framing in his own voice: "Human control over the use of force is essential. We cannot delegate life-or-death decisions to machines." Mary Wareham, the campaign's founding global coordinator, and Nicole van Rooijen, the coalition's current Executive Director, have both used the phrase as their working framing in coalition statements and press across the 2024–2025 UN cycle.

Why it has carried

Three features explain the framing's travel. First, the deliberate openness Moyes built into the original term — its status as a political concept rather than a tight legal definition — gave states a shared vocabulary to gather around without committing in advance to any particular substantive content, while leaving the campaign room to build that content out across thirteen years of CCW, GGE, and UNGA work. Second, the framing is structurally compatible with the two-tier treaty model the campaign has settled on: it gives the prohibition tier a target ("weapons that inherently lack meaningful human control") and the regulation tier a positive obligation ("preserving meaningful human control across all other autonomous weapons"), so the same term can do both jobs. Third, the framing names a human requirement rather than a technical capability, which has let it travel across the campaign's faith-community, scientific, survivor, and parliamentary engagement tracks — and across the position papers of newer AI-policy organisations across the corpus, where the question of how much human authority the deployment of an automated system requires has emerged as a cross-cutting concern beyond the lethal-weapons context where the term originated.