There are two unmistakable sides to the debate concerning the future of artificial intelligence service . In the “ gold rush ” corner are ship’s company like Google , Facebook , Amazon , and Microsoft aggressively investing in technology to make AI systems smarter and smarter . And in the “ doomsday ” corner are prominent thinkers like Elon Musk and Stephen Hawking who ’ve say that AI is like “ summoning the demon . ”
Now , one of the the most advanced AI outfit , Google ’s DeepMind , istaking rubber measuresin vitrine homo operator need to “ take control of a robot that is misbehaving [ that ] may conduce to irreversible consequences , ” which I sham includes but is not limited to pop all man . However , this paper does n’t get nigh so revelatory and keeps examples bare , like level-headed automaton work in a factory .
The published document was a joint effort between Google ’s DeepMind and Oxford University ’s Future of Humanity Institute , which as its name suggests , wants there to be a time to come for world . Founding director Nick Bostrom has been outspoken about the potential dangers of developing AI for decades and has publish whole book discussing the implications of A-one - intelligent golem .
This finical newspaper publisher , style “ Safely Interruptible Agents , ” investigates how to turn off AI if it starts doing something its human operator does n’t want it to do . The report is make full with math 99 percent of us will never understand , which essentially describes method for build what the paper brashly calls a “ fully grown blood-red button ” into AI .
The research worker have seen the same movies you have . You know , the one where the robot see to push aside a grow - off instruction . They ’re prepared for that .
This theme explore a way to check that a see agent will not con to preclude ( or try ! ) being interrupted by the environment or a human wheeler dealer .
It may seem like overkill or a uncalled-for limit look at the most telling AI achievement human beings has ever seen is that it is really good atboard game . But Bostrom has theorizedbeforethat it really only takes constructing a human - degree AI to speedily sling golem brains beyond our own :
Once artificial intelligence reaches human point , there will be a positive feedback loop that will give the development a further boost . AI would help constructing better AIs , which in turn would help build good ai , and so forth .
Better safe than sorry .
[ Intelligence.orgviaBusiness Insider ]
Daily Newsletter
Get the practiced technical school , skill , and acculturation intelligence in your inbox day by day .
newsworthiness from the future , delivered to your present .
You May Also Like
