Xinhua News Agency, Beijing, Dec. 29 (Xinhua) -- The United States' obsession with smearing other countries is just like the "illusion" of artificial intelligence
Xinhua News Agency reporter.
"Hallucination" is a technical term in the field of artificial intelligence that is used to describe the possible output of a machine that "makes something out of nothing". Such "illusions" have also appeared in some American politicians, who have falsely slandered China for using artificial intelligence technology to collect data on Americans. ** This kind of "illusion", some people in the United States should take a good look in the mirror.
The United States reported a few days ago that the director of the FBI and others said that China may be using artificial intelligence technology to collect and sort out the data of Americans, which poses a "top threat" to the United States. The argument is sensational, and on closer inspection, the report is full of words such as "probably" and "probably", without any evidence.
These false accusations are similar to the "hallucination" problem that has been alarmed in the development of generative AI recently. In the large language models on which the development of generative AI depends, "hallucination" refers to the phenomenon that the model generates text by weaving together some fake content that seems reasonable and smooth, but is actually not true, such as some models once told the story of "Lin Daiyu's three fights against the white bone spirit". As a flaw in the development of artificial intelligence, the potential harm of machine "hallucinations" has raised concerns among researchers and the general public.
Why do machine "hallucinations" occur?In addition to the limitations of AI technology in terms of underlying models and algorithms, an important reason is that there are problems such as bias in the corpus used to train models, and it is difficult for algorithms to correct them.
In the minds of some people in the United States, the reason for the "hallucination" of China is similar, that is, all kinds of political biases against China. All they think about all day long is how to attack China, linking China's development of artificial intelligence with the guise of "threatening the United States", which is nothing more than a new set of narratives to frame China, hollow out their minds to create confrontation with terrifying remarks, and smear China's development in high-tech fields such as artificial intelligence.
The "illusion" of the United States is manifested not only in slandering other countries out of nothing, but also in turning a blind eye to its own evil acts. A spokesperson for China pointed out at a press conference that China's sector is suffering from a massive number of cyberattacks almost every day, most of which come from the United States. In May this year, China's National Computer Virus Emergency Response Center and 360 jointly released the "Matrix Empire" Investigation Report - US Intelligence Agency, pointing out that the US Intelligence Agency has carried out cyber attacks on many countries, including China. From these facts, it is not difficult to see that the United States slanders China for using artificial intelligence technology to collect data from Americans, and it is a thief who shouts to catch a thief and fights a rake, and applies its usual tricks to China.
In order to deal with the problem of "illusion", AI researchers have proposed a strategy called "process supervision", that is, every step in the algorithm process must be supervised based on facts and correct data. However, in American political circles, it has become the norm to disregard facts and fabricate lies for the sake of political interests and manipulating public opinion. It is even more difficult to carry out "process supervision" of US politicians. As a part of the political ecology, some Americans** often deviate from the basic principle of authenticity and co-star with politicians.
For many years, the United States has used its superiority in network technology to conduct large-scale surveillance around the world, and the result will inevitably be that the world will see clearly the true face of its "Matrix Empire." From the "Prism Gate" 10 years ago to the "Leak Gate" this year, there are voices in the international community that are constantly exposing and criticizing the actions of the United States. Finland's "Helsinki Times** once published an article saying that the United States has proven to be the most serious threat to global cybersecurity. Helvoye Krašić, a professor at the University of Zagreb in Croatia, pointed out that on the one hand, the United States is constantly spying on countries, including its allies, and on the other hand, it is blaming other countries for engaging in cyber surveillance, which is a typical "double standard."
At present, the United States is still the most advanced country in artificial intelligence and other high-tech countries, but if it continues to use technology to do evil, fall into the "illusion" and cannot extricate itself, and blame other countries for its evil methods, it will eventually become a trick of self-deception.