Facebook Had to Switch Off Their Robots as They Created Their Own Language

in #technology7 years ago (edited)

554422.jpg

Some people have very genuine fears that one day humans will become inferior to robots. As technology advances, they believe that despite these things being invented to help us, we'll end up doing their bidding.

It still remains to be seen whether that will be the case, but for those fearful that it will be, this news won't help - scientists had to switch off Facebook's artificial intelligence as they reportedly began to 'create their own language'.

Two machines named Bob and Alice were found to be talking to each other in what was assumed to be gibberish, Metro reports, but then after close inspection it was apparent that they were using code that only they could understand.

According to the New Scientist, the 'neutral network' started translating phrases into this code as it was easier than using the programmed language.

"I can i i everything else," Bob reportedly said to Digital Journal, to which Alice replied: "Balls have zero to me to me to me..."

It's believed that the robots were trying to resolve problems they had encountered because 'I' and 'to me' address personal issues.

"There was no reward to sticking to English language," Georgia Tech scientist Dhruv Batra told Fast Co. Design. "If I say 'the' five times, you interpret that to mean I want five copies of this item. This isn't so different from the way communities of humans create shorthands."

Bit scary, isn't it? Even more so when you consider that the chief of engineering at Google, Ray Kurzweil, reckons robots "will reach human intelligence by 2029."

If you've got Amazon's Alexa, or indeed use Siri on your phone, then you'll know just how far Artificial Intelligence has come. However, we're not quite at the point where anything has passed the Turing test, which basically tests whether technology can produce intelligence not too dissimilar to that of a human.

Kurzweil is prepared for when that day comes, claiming that it 'will change the nature of humanity itself', with life as we know it coming to an end in 2045.

Over the next decade, breakthroughs will reportedly be made in General AI, with humans and robots becoming comparable.

Following that, there's only one more hurdle to jump, and that's artificial super-intelligence (API). Sounds good, doesn't it? Well it's not. It's the kind of shit that gave Will Smith a right hard time in I, Robot, and none of us are anywhere near skilled or handsome enough to win that battle.

"We project our own humanist delusions on what life might be life [when artificial intelligence reaches maturity]," philosopher Slavoj Žižek says.

"The very basics of what a human being will be will change.

"But technology never stands on its own. It's always in a set of relations and part of society."

If we're honest, if they can work a hoover, do the dishes, pull a pint and play FIFA with you, I think we should be welcoming them with open arms.

Source Article: Metro.co.uk

Sort:  

Reminds me of the movie Automata with Antonio Banderas where a robot writes the programming of subsequent robot models so that no human could crack the code. However, they quickly decommission the robot because its intelligence increases exponentially which is very threatening. Good watch (though a bit low budget) if you haven't seen it already.

Congratulations @jackthar1ppir! You have completed some achievement on Steemit and have been rewarded with new badge(s) :

Award for the number of comments received

Click on any badge to view your own Board of Honor on SteemitBoard.
For more information about SteemitBoard, click here

If you no longer want to receive notifications, reply to this comment with the word STOP

By upvoting this notification, you can help all Steemit users. Learn how here!