It is very common in science fiction films for autonomous armed robots to make life-or-death decisions — often to the detriment of the hero. But these days, lethal machines with an ability to pull the trigger are increasingly becoming more science than fiction.
In September, Russian President Vladimir Putin warned that “the one who becomes the leader in this sphere will be the ruler of the world,” referring to artificial intelligence in general. In the same speech, Putin also appeared to suggest that future wars would consist of battles between autonomous drones but then reassured his audience that Russia would naturally share such technology if it were to develop it first.
Some systems already available come extremely close. The security surveillance robots used by South Korea in the demilitarized zone which separates it from North Korea could theoretically detect and kill targets without human interference, for instance.
But so far, no weapons system operates with real artificial intelligence and is able to adapt to changing circumstances by rewriting or modifying the algorithms written by human coders. All existing mechanisms still rely on human intervention and their decisions.
The rapid advances in the field have nevertheless triggered concerns among human rights critics and lawyers about the possible implications of the rise of autonomous weapons systems commonly known as killer robots. Who would take responsibility for incidents which are so far classified as war crimes? Could robots decide to turn against their own operators? And would wars fought between autonomous weapons systems be less brutal than conventional conflicts, or would they provoke more collateral damage?
One of the most vocal groups in opposition of such systems has been the Campaign to Stop Killer Robots, which calls for a pre-emptive ban. So far, more than 100 CEOs and founders of artificial intelligence and robotics companies have signed the campaign’s open letter to the United Nations, urging the world community “to find a way to protect us all from these dangers.”
“Lethal autonomous weapons threaten to become the third revolution in warfare. Once developed, they will permit armed conflict to be fought at a scale greater than ever, and at time scales faster than humans can comprehend,” read its open letter.
Critics fear that criminals or rogue states could also eventually get control of these systems. “(Autonomous systems) can be weapons of terror, weapons that despots and terrorists use against innocent populations, and weapons hacked to behave in undesirable ways,” the open letter added.
Such concerns have existed for years and were also shared by several Nobel laureates, including former Polish president Lech Walesa, who signed a joint letter in 2014, as well: “It is unconscionable that human beings are expanding research and development of lethal machines that would be able to kill people without human intervention,” the 2014 statement read.
So far, a proposed ban on autonomous weapons systems has triggered little enthusiasm among U.N. member states. Some of the world’s leading militaries, including the United States and Russia, are researching and experimenting on how to make existing weapons more autonomous. Some researchers have welcomed efforts to expand artificial intelligence use in warfare.
Defense analyst Joshua Foust has cautioned against condemning outright such systems, writing already in 2012 that humans, too, “are imperfect — targets may be misidentified, vital intelligence can be discounted because of cognitive biases, and outside information just might not be available to make a decision.”
“Autonomous systems can dramatically improve that process so that civilians are actually much better protected than by human inputs alone,” Foust wrote.
If that vision becomes reality, perhaps the most crucial question will be whether robots can be taught how to recognize wrongdoing by themselves.
Many professionals in the artificial intelligence industry hope that they will never have to find out the answer.