Tech giant Google announced it is working on a system known as the “kill switch” which would prevent artificial intelligence (AI) machines from acting against humans’ will.
Google's artificial intelligence division, DeepMind, and Oxford University said machines can be coded to prevent programmers from overriding a robot’s actions.
Google said that they will ensure that humans can always “press the big red button” against robots.
The company published a paper with details of the development of their new “Safely Interruptible Agents,” that will prevent robots from acting on their own.
Artificial intelligence allows robots to learn from their own experiences and develop themselves, according to researchers.
“It’s unlikely for AI agents to behave in an optimal manner all the time, especially in the real world interaction,” says the statement.
According to the paper, machines can learn to override a command, as was the case with the 2013 AI that was taught to play Tetris and learned to pause a game forever to avoid losing.