Robots will become smarter and faster than humans, an Oxford University professor said recently.
And not only will robots be better than us at a lot of things, they'll eventually take over and make humans redundant, the professor reckons.
Dr. Stuart Armstrong, of Oxford's Future of Humanity Institute, thinks the future relationship between humans and robots is not going to turn out well for the humans.
Among the prophecies? Armstrong thinks humans could be wiped out because robots' Artificial General Intelligence (AGI) comprehension is going to be too literal. For example, the robots could interpret an instruction such as "prevent human suffering" as "kill all humans."
Likewise, an instruction like "keep humans safe and happy" could be interpreted as "entomb everyone in concrete coffins on heroin drips," Armstrong says.
The problem will be compounded by the fact that the robots will be considerably more powerful than humans, so humans won't be able to counter the machines.
Add speed to the equation and we're in trouble. The robots will do everything at super-speed—much faster than humans can perform tasks.
In other words, conceivably, a robot could misinterpret the nuance of an instruction and complete a task—possibly in an undesirable way—before the human can explain better what's required.
AGI will do things "very, very fast," Armstrong said, being quoted in the Daily Mail newspaper.
The threat from artificial intelligence is "not just a Hollywood fantasy," Patrick Sawer says, writing in the Telegraph newspaper.
Another issue with robots, the Telegraph says, is that as the machines harness huge amounts of computing power they will eventually create global networks with each other.
They'll communicate without human interference, and that could be a problem. They could start making decisions among each other.
It doesn't stop there. Other problems Armstrong envisages, in addition to machines deliberately wiping out the human race because they don't understand the subtleties of our language, is that they won't understand moral codes. Humans find it difficult to separate right and wrong, so they are not good role models when it comes to behavior.
Plus, it's going to be difficult to tell if the artificial intelligence robots are developing a "benign or deadly direction," Armstrong said.
AGI technology might appear to be beneficial, like Apple's voice-control app Siri, but it has the potential to "logically take over," Armstrong thinks.
"Humans steer the future not because we're the strongest or the fastest, but because we're the smartest. When machines become smarter than humans, we'll be handing them the steering wheel," Armstrong said.
But it might not all be doom and gloom.
"Since when did an IT system work as well in practice as it did in the marketing blurb?" Telegraph commenter 123db says after reading Armstrong's remarks, implying that we shouldn't worry because the things aren't going to work anyway.
"Just make sure each machine comes with a giant fuse in an easy-to-remove location on the back of their heads," the commenter suggests.
This article is published as part of the IDG Contributor Network. Want to Join?