Facebook, the world’s leading social network, reportedly shut down an artificial intelligence (AI) system that the company was working on when the system’s bots started to create their own language. The language, not part of the bot’s core algorithm, was created from scratch without human involvement.
The creation of the language defies the program’s codes and forced the company to terminate the system. The language was not able to be interpreted by the company.
The announcement comes a week after Tesla’s Elon Musk stated that AI was the world’s “biggest risk.”
Facebook’s AI system used a series of bots that communicate with each other. The sophisticated bots started out communicating in English before creating their own language.
The bots were allowed to converse between themselves freely to help strengthen their own conversational skills. The bots then started to learn how to negotiate using machine language and strategies to help improve negotiation outcomes.
The bots learned how to act interested in one item during a negotiation and “sacrifice” the item later on in the negotiations, although the item wasn’t truly important to the bot. The fast-learning bots were shut down by Facebook with no mention of future plans for the bots.
Several leaders in the tech world have expressed concerns over AI systems, including Elon Musk, Steve Wozniak and Bill Gates.
Facebook’s Mark Zuckerberg stated earlier this month, “I think people who are naysayers are trying to drum up these doomsday scenarios – I just, I don’t understand it. It’s really negative and in some ways I think it’s pretty irresponsible.”
Zuckerberg’s comment came after Musk called for regulations and care for AI in the future.
Musk responded to the debate, tweeting, “I’ve talked to Mark about this. His understanding of the subject is limited.”
Musk is calling for lawmakers to be proactive and put regulations in place to help regulate AI systems, which have shown the possibility of creating their own language and communicating together.