- Google DeepMind is working on a ‘kill switch’ to prevent AI from going rogue
- Researchers said AI agents are “unlikely to behave optimally all the time”
- To prevent robots from learning to ignore human intervention, the team developed an algorithm called ‘safe interruptability’
Since the dawn of artificial intelligence (AI), mankind has been worried that robots will become the dominant intelligent being on Earth and may become a threat to humanity. To put these concerns to rest, Google DeepMind is developing a ‘kill switch’ to prevent a robot apocalypse.
The researchers from DeepMind explained, in a paper published in collaboration with the Future of Humanity Institute of Oxford University, that AI agents are “unlikely to behave optimally all the time,”, as per an article published by RT.
“It may be necessary for a human operator to press the big red button to prevent the agent from continuing a harmful sequence of actions and lead the agent into a safer situation,” said researchers Laurent Orseau and Stuart Armstrong.
The researchers also claimed the system would prevent robots from learning to ignore human intervention, which they describe as “an undesirable outcome.” For this, the team developed an algorithm called ‘safe interruptability’,
“Safe interruptibility can be useful to take control of a robot that is misbehaving and may lead to irreversible consequences, or to take it out of a delicate situation, or even to temporarily use it to achieve a task it did not learn to perform or would not normally receive rewards for,” the researchers said.
Speaking at a debate on artificial intelligence in London earlier this year, Armstrong warned that humans could be wiped out even if robots are instructed to protect people, as per an article published by Daily Mail.
“There is a risk of this kind of pernicious behavior by an AI,” he said, adding that human language is subtle and can be easily misinterpreted. “You can give AI controls, and it will be under the controls it was given. But these may not be the controls that were meant.”
With the leaps mankind is making in technological advancement, it only make sense to take precautionary measures needed to prevent these robots, as the researchers noted, from going rogue and wiping out human existence with their metal fists.