Google is one of the organizations at the forefront of robotics and computerized reasoning research, and being in that position implies they have the most to stress over. The possibility of a robot takeover may at present be a theoretical, science anecdotal idea to us, however, Google has really assembled a rundown of practices that would cause them great concern, both for efficiency and safety in the future.
Among the recurring themes is the possibility of deceit: that robots would learn to “accomplish goals” by hiding evidence that more work needs to be done.
The researchers also laid out the possibility of smart robots avoiding humans to evade further assignments
You can read the rest here.
Maybe we and our artificial companions of the future have a lot in common after all. So much for robots proving more efficient than humans, if they’re planning to take two-hour lunch breaks.