Emergency shutdown! The Secret of Conversation Content of Artificial Intelligence’s Self-created Chat Language

       CCTV News:Artificial intelligence has entered the lives of ordinary people. However, human beings are increasingly wary of the development of artificial intelligence. Last weekend, the artificial intelligence chat robot of Facebook, an American social networking site, went out of control, and researchers had to shut down all the activities of the artificial intelligence system they were developing to control it.

       Bob: I can do everything else.

       Alice, an artificial intelligence robot: The ball has a ball for me, for me, for me, for me, for me, for me, for me, for me.

       This passage is an excerpt from the conversation between Bob and Alice, a chat robot announced by Facebook. Don’t you understand? Indeed, this dialogue surprised the researchers.

       At first, researchers were confused about robots chatting in their own language, but later they found that in fact, robots themselves created a more complicated way to deal with the tasks given by humans.

Image source network

Image source network

       Duluf Batra, an artificial intelligence researcher at Facebook, believes that just as people sometimes use jargon to talk to solve problems quickly, artificial intelligence robots will create their own languages and do similar things. Because researchers can’t understand the self-created language of artificial intelligence, it creates obstacles for them to build artificial neural networks. Although it is not clear whether this artificial intelligence system can overturn the operator’s instructions, for the sake of caution, the researchers intelligently shut them down temporarily.

       Some people say that artificial intelligence is like a high-speed train. When people are still looking at its true appearance, it whizzes past at a speed that makes people sigh. For artificial intelligence, whether to "brake in time" or "accelerate forward", the two opposing views that have been arguing for a long time are now colliding more and more fiercely. There is no trace of the "confrontation" between artificial intelligence and human beings, but human beings themselves have launched a heated debate on this.

one

Image source network

       In the American sci-fi film I, Robot in 2004, the robot’s computing ability has been continuously improved, and it has learned to think independently, even solved the control code, becoming a completely independent group, even destroying human society, and turning into a "mechanical enemy" of the whole human race at any time.

       In sci-fi movies and literary works, there are many descriptions of artificial intelligence out of control and human beings at the end of the road. In real life, industry giants have also given warnings.

       Elon Musk once said that creating artificial intelligence technology is like summoning the devil. However, Zuckerberg, the founder of Facebook, who is also a big name in the industry, is very optimistic about the prospects of artificial intelligence.

       Mark zuckerberg said: "I am optimistic about the prospect of artificial intelligence. I really can’t understand those who sing the opposite. They always advocate that artificial intelligence will lead to the end of the world. This statement is too negative. I think it is quite irresponsible in a sense." Intriguingly, shortly after saying this, Zuckerberg’s "Facebook" company experienced a situation in which artificial intelligence was out of control.

Image source network

Image source network

       So, how should humans and artificial intelligence get along? Kevin kelly, a famous American science and technology observer and founder of Wired magazine, pointed out that the relationship between human beings and machines should never be defined as "competition", because in that case, we will surely lose. Humans and machines are destined to go hand in hand. However, some experts have warned that in the process of developing artificial intelligence, we need to pay attention to the risks brought by artificial intelligence technology.

       Stuart Russell, a computer professor at the University of California, Berkeley, said: "When we are making machines intelligent, what we have to do is to ensure that they are 100% on our side, not 99% on our side, because as long as there are 1% loopholes, it may lead to human beings losing to artificial intelligence."

       No matter what views people hold, artificial intelligence technology, known as the driving force of the new round of industrial revolution, is changing human life in an all-round and accelerated way. People must be prepared and carefully evaluate the opportunities and challenges brought by artificial intelligence.