Movement grows against killer robots

The miniature UAV, smaller than a human palm, zips right to its human target — identified through facial recognition technology — and pierces the forehead with a projectile, for an instant kill.

That harrowing scene takes place in a seven-minute viral video issued by autonomousweapons.org, a non-profit sounding warning bells over potential automation of weapons. Its Campaign to Stop Killer Robots (#BANKILLEROBOTS) seeks a preemptive international ban on “fully autonomous weapons which enable strikes to be carried out without human intervention.”

“Allowing machines to choose to kill humans will be devastating our security and freedom,” warns Stuart Russel, professor of computer science at the University of California at Berkeley, on the video.

What feels like science fiction to those of us raised on the Terminator franchise could be closer than we think. Because of this, a new U.S. Army report emphasizes the need to develop countermeasures against swarming drones and other unmanned weapons.

The Army and U.S. Department of Defense have invested significantly in technologies in response to these threats, often focusing on detecting radio frequency transmissions of the UAVs or their operators.

However, as the report points out, today’s consumer and customized UAS increasingly can operate without radio frequency command-and-control links by using automated target recognition and tracking, obstacle avoidance, and other capabilities enabled by software.

The U.S. Army discusses the pros and cons of autonomous weapons in a June 2017 article in Military Review, saying an international ban should be considered on “fully autonomous weapons with missions that cannot be aborted and that cannot be recalled once they are launched. If they malfunction and target civilian centers, there is no way to stop them.”

Sobering thoughts about a future that may not be too distant.