A parliamentary report highlights a new arms race based on artificial intelligence. The United States, Russia and China are interested in acquiring technological superiority over the adversary. France prohibits the use of weapons systems beyond human control.
Can robots be delegated the function of killing in a military operation? How can these new weapons be ethically and legally framed? How should French armies prepare for the arrival of killer robots?
These were the questions to be answered by the Information Mission on Autonomous Lethal Weapons Systems (SALA). Its conclusions and recommendations were made public on 22 July by its authors, Fabien Gouttefarde, Member of Parliament for Eure (La République en Marche) and Claude de Ganay, MP for the Loiret (Les Républicains).
Autonomous weapons, but to what extent?
The issue of killer robots is becoming increasingly debated with the increasing integration of artificial intelligence technologies into weapons systems. But do they really exist already? After seven months of study and thirty hearings of military, researchers, lawyers, philosophers… parliamentarians acknowledged that they had received diametrically opposed opinions. “Some people told us that we were interested in a subject that didn’t exist and others said it was the third technological revolution in the world of defense after the invention of gunpowder and nuclear weapons,” says Fabien Goutteffarde.
This is finally normal, since there is no universal definition of autonomous lethal weapons (SALA) systems! “If autonomy is defined in a pure and perfect way, there is no weapons system capable of selecting and engaging a target on its own without human intervention in a changing environment, “he said. Automated weapons systems or remotely operated drones, which are widely deployed in conflicts, do not involve such a level of autonomy and are therefore not considered killer robots.
The risk of gearing
Beyond this semantic battle over killer robots, all the armies of the world are interested and take a stand. MEPs even describe a risk of over-bidding in this area, especially on the part of the major military powers: to minimise the loss of life, the Russian generals would have relied on the replacement of their soldiers by robotic units, leading the Americans to do the same if they were faced with such an army, while the Chinese military is preparing to integrate high-dose AI into their new weapons systems.
“There is a form of gearing,” the rapporteurs warn. Their paper evokes technological advances made by these great powers. The United States is developing the Sea Hunter, a 60-metre transoceaan autonomous vessel dedicated to anti-submarine warfare and capable of navigating international waters independently of current navigation rules. In the naval field, China has designed an HSU 001 autonomous submarine, of relatively small size, to mesh the oceans. As for Russia, it focuses on robotics. She developed a small tank capable of following a soldier and firing at the same target.
“Terminator won’t be marching on July 14th”
What about France? The French position was summed up by a shocking formula by the Minister of the Armed Forces Florence Parly during a speech given in April 2019 on the use of artificial intelligence by the armed forces: “Terminator will not march on 14 July”. It thus recalled France’s categorical refusal to entrust the decision of life or death to a machine that would act in a fully autonomous manner beyond human control. “Such systems are fundamentally contrary to all our principles. They have no operational interest in a state whose armies respect international law, and we will not deploy any,” she said.
Faced with these different military strategies, however, the authors of the report believe that the development of ever more autonomous weapons systems is inevitable and that their development must even be accompanied. “It is important that France and Europe more generally do not remain on the margins of the artificial intelligence revolution. This is not an option. At the risk of a strategic downgrade, we need to do research, to make developments in artificial intelligence defence, “advocate MEPs. According to them, the autonomy of arms systems is at the heart of a new arms race and Europe must be part of it, while remaining respectful of ethical principles and international humanitarian law.