Killer Robots (a.k.a. Lethal Autonomous Weapons): a Chance for Regulation
By Lillian Mauldin, AIUSA’s Military, Security and Police Transfers Coordination Group
Next week, diplomats will convene in Geneva to decide whether to begin negotiating new international law on autonomy in weapons systems.
Lethal Autonomous Weapons (LAWs), colloquially known by many in the arms control community as “Killer Robots”, are “weapons systems which, once activated, can select, attack, kill and injure human targets without a person in control.” Because they’re robots, lethal autonomous weapons can’t exercise the judgement in life-or-death situations that is required by international law — and no amount of fancy technology will change that. Killer robots also perpetuate and extend human biases into the artificial intelligence realm, making historically marginalized individuals even more likely to experience state violence. At the same time, by distancing humans from the consequences of violence and detaching them from conflict and casualties, they risk lowering the threshold to war.
Killer Robots are not just science fiction. In March this year, a UN Panel of Experts informed the Security Council that in 2020 ‘lethal autonomous weapon systems’ had been used in attacks in Libya, with fighters ‘hunted down and remotely engaged.’ This was the first time a UN report had noted the use of autonomous weapons, and it shows that this is not some futuristic speculative technology — it’s a real problem, now. In fact, we’re already in the early stages of a killer robots arms race, with major arms-exporting countries — including the US, China, Israel, South Korea, Russia, Australia, India, Turkey and the UK — investing heavily in the development of autonomous weapons. The UK is developing an unmanned autonomous drone to identify a target in a programmed area without human intervention. China is creating small drone “swarms” which could be programmed to attack anything that emits a body temperature, and Russia has built a robot tank which can be fitted with a machine gun or grenade launcher. According to the Congressional Research Service, the Pentagon has informed Congress that it does not currently have LAWS in its inventory, but some senior military and defense leaders have indicated that if competitor states develop them the US might alter that position. Currently there is no prohibition in US law or policy against the development or employment of LAWS.
The prospect of a world filled with marauding killer robots is grim, but several organizations have been working together since 2012 to regulate and/or ban killer robots. To keep human beings firmly and permanently in control of life-or-death decisions during war and civil unrest, and as the most recent initiative in these coalition efforts, Amnesty International and its partners in the Killer Robot campaign are calling on government leaders around the world to launch negotiations for new international law limiting autonomy in weapons systems — to ensure human control in the use of force and to prohibit machines that target people, reducing us to objects, stereotypes and data points.
Amnesty International USA has been concerned about these developments for quite a while. But we have come to a new juncture. At the Sixth Review Conference of the Convention on Conventional Weapons (CCW), scheduled at the United Nations in Geneva for December 13–17, states will agree whether to proceed to a negotiation mandate for killer robots or continue with discussions.
In advance of the December discussions, Amnesty and the Stop Killer Robots campaign unveiled a social media filter that provides a glimpse into the future of war, policing and border control. Escape the Scan, a filter for Instagram and Facebook, uses augmented reality (AR) technology to depict aspects of weapons systems that are already in development, such as facial recognition, movement sensors, and the ability to launch attacks on ‘target’ without meaningful human control.
People must hold their governments accountable to protect those who are most vulnerable to violence as a result of killer robots. As international law currently stands, there is no clear prohibition on the use of autonomous weapons. The international community must agree to a robust set of controls to ensure that LAWS do not make the most important decision of all: the one of life or death.