Just hours after Tesla and SpaceX CEO and all-around futurist Elon Musk spoke at the National Governors Association meeting in Rhode Island and warned the politicians that artificial intelligence poses an “existential threat” to human civilization, a security robot working outside a Washington, D.C. office building drowned in a fountain. Was it a malfunction? Was it the stress of guarding unappreciative humans who may turn on it upon orders from their Lord Elon? What do other robots and AI think? If it’s a suicide, does this mean that robots have already achieved an existential level and Musk’s warning is closer than we think? Will the movie based on this incident be a sci-fi action film or a comedy?
The autonomous Knightscope K5 had only been on patrol for a few days at its job outside the Washington Harbour office and retail complex in Georgetown where it is responsible for spotting misdemeanors, citing parking violations and identifying known criminals using its GPS, lasers, sensors, cameras and facial recognition software. On July 17th, employees at the building tweeted that the 300-pound, 5-foot-tall bullet-shaped droid was dead in the water … literally.
Our D.C. office building got a security robot. It drowned itself. We were promised flying cars, instead we got suicidal robots.
— Bilal Farooqui (@bilalfarooqui)
What happened? The K5 certainly has enough equipment to handle any emergency … a 360-degree video camera array, sensitive microphones that can identify gunshots, air quality sensors, thermal imaging capability, a siren and a phone to call for help. Perhaps the phone directory had call 911 but not the suicide hotline number? Did the engineers forget to put in an edge sensor for steps? Even a Roomba has one of those. Did K5 turn it off itself? Why?
Human security officers rushed in, first to insure that no humans were hurt – no one was – and then to tend to the robot, which had no pulse and no identifiable spot to place the defibrillator paddles. Knightscope issued a perfunctory statement saying it was investigating this “isolated event” and was rushing a replacement, to the consternation of office workers hoping to park free for a few days.
Was this “isolated event” just an accident that can be fixed with a sensor? The K5 robots have been having problems lately with one knocking down a child at a mall and another being knocked down itself by a drunk. Is K5 robotically depressed because it sees these ‘imperfections’ continuing due to robots being built by imperfect humans? Is the solution to let AI built robots? Is this what Musk is warning about?
Is this what Douglas Adams’ robot Marvin was warning about?
” I’ve calculated your chance of survival, but I don’t think you’ll like it. “