This is how scientists are working to prevent a robot rebellion

It's something every science-fiction reader knows to fear: the robot rebellion, when artificial intelligence will learn how to override and outsmart human commands. Now scientists at Google are working to develop a "kill switch" to prevent AI from figuring out they got the short end of the deal (or just getting mischievous), BBC News reports.

Working with researchers at Oxford University, Google's AI division, DeepMind, plans to ensure that humans always remain in charge. "Now and then it may be necessary for a human operator to press a big red button to prevent the agent from continuing a harmful sequence of actions," the researchers wrote — a vaguely terrifying reassurance.

Artificial intelligence won't "behave optimally all the time," but the important thing is that people are able to override when a robot "learns" to misbehave, such as when a Tetris-playing robot figured out that if it paused a game forever, it would never lose, or when Microsoft's Tay chatbot became racist and sexist.

The Week

Escape your echo chamber. Get the facts behind the news, plus analysis from multiple perspectives.

SUBSCRIBE & SAVE

Sign up for The Week's Free Newsletters

Join more than 350,000 subscribers and keep yourself informed with a selection of The Week’s most interesting, enlightening and entertaining stories - plus daily puzzles.

Join more than 350,000 subscribers and keep yourself informed with a selection of The Week’s most interesting, enlightening and entertaining stories - plus daily puzzles.

Sign up

"No system is ever going to be foolproof — it is matter of making it as good as possible, and this is one of the first steps," DeepMind's Dr. Laurent Orseau said.