Facebook's bots have in fact developed a new language independently that will enable them to communicate more effectively between them. This is not the first case in which researchers recognize the development of a new language by artificial intelligence systems. In a previous case, Google's artificial intelligence, which is used in its online translation service, has developed a new language for itself to perform its tasks more efficiently.

These languages ​​are soon developed into a form of communication that no longer resembles a human language. Artificial intelligence systems tend to develop capabilities for effective communication according to their tasks. For example, in the case of Facebook bots, Bob and Alice, they were required to perform certain tasks and the language used them to negotiate the tasks Bob would perform and Alice's

After analyzing the language they used, the researchers found that it was much more effective for them to communicate their messages than if they were using normal human language. The developed language has died of a deep layer of words and phrases that are useful only to humans, and the semantics of its words as well as its syntactic structure have been adapted to efficiency. In simple language, artificial intelligence has died of anything that smells of poetic, social, or emotional curl, leaving only the effective expressions for its task.

Artificial intelligence works on the basis of a "reward" in which it is supposed to take a certain course of action spontaneously if it provides it with a certain "profit". This method allows bots to find the most efficient way to perform tasks independently. In the case of Bob and Alice, they had no value in continuing to use English, and their profit was in the invention of a new and more effective language, which was actually done.

Moreover, in a lab at OpenAI, Ayalon Masak Artificial Intelligence Initiative, a similar experiment found that bots could learn these new languages ​​without being part of their development process. Perhaps this is why Musk is concerned that artificial intelligence may one day turn over its human creators and become a risk. What is certain is that the more human operators of artificial intelligence will be excluded from the process of communication between bots, they may lose control over them and who knows, perhaps one day we will really wake up to a brave new world in which the bots will control us through the Matrix.

In a June 14 post describing the project, FAIR researchers said the project

represents an important step for the research community and bot developers toward creating chatbots that can reason, converse, and negotiate, all key steps in building a personalized digital assistant.