Share this:

Rumors were heard about Facebook artificial intelligence program which was getting smarter day by day and has generated its own creepy language. The two AI machines were talking in a language not understandable to humans and Facebook  pull plugs of AI bots.

Google is working on Deep mind , Facebook debuted a project to improve quality of Chatbot. Microsoft is making AI chip for holo lens, Almost every big IT company is working on Artificial Intelligence.

Facebook started a new artificial intelligence project aspired to improve the quality of chatbots. Few weeks ago, FastCo Design reported on Facebook attempt to develop a negotiating software “Generative Adversarial Network”.

Two bots were designed to negotiate in start to end conversation with each other or with  people and approach at a same decision.”- Facebook Artificial Intelligence Research Unit

The bots never did anything more smarter than this and only discussed to split an array of given items.

 

Facebook Researcher told that there goal was to develop bot with ability to negotiate deals with other user end fluently so the user could not notice they are talking to a robot. Soon after they started the bot they researchers realized that they made an mistake by not to restrict bots to communicate only in English, which then started to communicate in a short code language which looks creepy. The only thing bots were able to learn was efficient way of trade.

Facebook did lose their AI bots but not the fact that they invent their own language to communicate but according to Mike Lewis “Our Interest was having bots who could talk to people”.

Share this:

LEAVE A REPLY

Please enter your comment!
Please enter your name here