Artificial intelligence has always been a subject of much apprehension, what with Elon Musk and Stephen Hawking being wary of what the technology could do. Apparently, Google has also been thinking along the company has recently published documents that reveals that Google is currently working on a way to implement a Kill Switch into their DeepMind self-learning platform.
The document, titled Safely Interruptible Agents, is published on the website of the Machine Intelligence Research Institute (MIRI) and was written by Laurent Orseau, a search scientist at Google DeepMind. According to the paper, this kill switch would allow a ‘human operator’ to repeatedly and safely interrupt an AI, while making sure that the AI doesn’t learn how to prevent or induce the interruptions.
“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 this,” states the papers.
Source: Business Insider
