AI has really begun to awaken, and human beings may really be eliminated by AI in the future, and this concern has now become a reality. When scientist Hawking was alive, he reminded more than once that artificial intelligence will destroy modern society, end human beings, and completely replace human beings as new life forms. Because chatGPT was born just a few months ago, it can be said that it completely refreshed the cognition of all mankind today.
It turns out that the current artificial intelligence has been trained to such a stage of intelligence and even horror. After a brief shock, people began to panic again. If AI continues to train and grow, is it possible that one day it will get out of control, there will be self-awareness awakening, will not be satisfied with being manipulated by humans to kill humans, and become the new ruler of the earth? These fears are not unfounded.
Because now AI is evolving too fast, far beyond human imagination. So later, Musk united more than 1,000 experts around the world to urgently issue a statement, calling on everyone not to train more advanced AI, because it really may get out of control, and once it gets out of control, human beings are really finished. But once Pandora's box is opened, can it really be said that it is closed?
Just on June 2 of this year, the American media broke the shocking news, saying that the latest AI began to kill people, because that person prevented him from achieving his goal, so the AI killed the operator in anger. This happened during an AI simulation test in the US military, and the US military (undescribable) is still undergoing large-scale high-intensity AI training. The U.S. military uses AI technology to implant drones, turning drones into terminators in movies to carry out assassination missions.
But in this process, when the operator wanted to come forward to prevent him from continuing to complete the task, the AI accidentally did not accept the operator's instructions, thinking that the operator was obstructing him from completing the task, so he chose to attack the operator and kill the operator.
Hearing this, do you feel terrible? More terrifying is yet to come. Later, the US military found that this could not work, so it re-trained the AI, telling him that you had to obey the operator's command, and the operator told you to terminate, you must terminate.
Under no circumstances can you attack the operator, he directly destroys the signal tower that the operator uses to communicate with him, trying to disconnect from the operator. This AI is about to become an immortal. Now AI is like a child on a chain, with which humans command him what to do and what not to do, but now the child has begun to try to break free. If he had already started this act, he would have broken free one day in the future.
Imagine what will become of AI when it breaks free of this chain that binds him? How strong will he be? Can humans control him again? I'm afraid this will have to be verified by time.
