The secretive AI division of Google is trying to develop a big red button that will be able to halt its artificial intelligence software. Google was previously warned by researchers that AI could threaten humanity. An expert has also admitted that Google’s historic win over Go world champion proves AI can be unpredictable and immoral.
Google’s DeepMind team’s answer to this is an off switch. Two of the researchers that is Laurent Orseau of Deep Mind and Dr Stuart Armstrong at the Future of Humanity Institute at Oxford University, explained that it is unlikely that the A1 agents will behave optimally all the time. Now and then 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 – harmful either for the agent or for the environment – and lead the agent into a safer situation.According to them the key to it is safe interruptibility.
They added that 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. At the same time they also admit that it may not be possible to halt all systems. The authors admit that it is unclear if all algorithms can be easily made safely interruptible.
Earlier, Dr Armstrong has warned that it is a race against time to develop safeguards around artificial intelligence research, before robots outwit us or even accidentally cause loss of life. According to the Telegraph report based at the Future of Humanity Institute at Oxford University Dr Armstrong predicted that robots will become very essential to our everyday lives for doing menial tasks but later on they will make humans unneeded and take over.
