Recently, a group of studies with the theme of "A university institution have found that what proportion of diseases can be eliminated by a certain food, is it true?" Articles on the subject have sprung up on Internet platforms. The structure of these articles is highly similar, almost all of them start from the perspective of patients, telling their stories of seeking medical treatment.
Regarding the efficacy of the so-called "diet therapy" mentioned in the title, the article seems to maintain a certain "rationality". When patients ask questions, doctors without exception will bring out so-called authoritative research and papers, saying that some ingredients in food do have certain effects, but their effects cannot be exaggerated, and scientific treatment must be achieved to cure diseases. In the end, with the efforts of doctors and patients, the disease was finally cured, and at the end of the article, patients and readers were even thoughtfully reminded to eat a balanced diet, have a regular schedule and rest, and insist on exercising.
Such a "healing" story mixed with highly personal emotions seems to be nothing wrong at first glance, but the reporter combed through and found that in the process of telling the story, the article was mixed with a lot of content that contradicted the facts, and it was still "pseudo-science" in nature, and it was more concealed and deceptive.
Recently, a large number of "pseudo-popularization" articles with similar structures have appeared on the Internet.
The "pseudo-popular science" article is full of loopholes
In an article about "can drinking vinegar clear blood clots", it was mentioned that the vascular surgery team of a certain university hospital published a paper on the prevention of thrombosis by inhibiting platelet activation and fibrin polymerization with acetic acid in the journal "Natural · Communications", however, the reporter searched in the journal "Natural · Communications" and the official website of the university's medical school, and did not find this paper.
As for the claims mentioned in the article, "acetic acid can inhibit the activation and aggregation of platelets, and can also reduce the formation of fibrin" and "the acetic acid contained in vinegar" that advises patients to drink vinegar in moderation, it has long been falsified. A few years ago, Hong Zhaoguang, a professor at Beijing Anzhen Hospital affiliated to Capital Medical University, said in an interview with the media that acetic acid will be neutralized by alkaline digestive juices in the small intestine, and the acetic acid that can be absorbed into the blood is very limited, and it will soon be cleaned up by the blood stabilization system, so acetic acid will not change the pH of the blood, nor can it soften the blood vessels.
Fabricating and tampering with authoritative research is exactly the common routine of these articles. Moreover, when it cites the so-called "authoritative research", the research subjects almost always use vague titles and individuals such as "experienced Dr. Liu", "senior immunology expert Professor Li", and "Dr. Zhao specializing in traditional Chinese medicine". Does it really exist? The reader simply can't verify it. It is worth noting that in formal news reports or popular science writing, hospitals, doctors, teams, research content, etc. will be clearly listed, and these key information will not be hidden.
In addition, in order to increase the credibility of the article, some "pseudo-popular" articles will be accompanied by some pictures that approximate the paper charts. However, the reporter verified and found that these pictures belonged to Zhang Guan Li Dai. For example, in the article on "Can Eating Garlic Reduce Inflammation?", the author cites an English picture to show the experimental process of a patient taking garlic extract, but in fact this picture is about CAR-T therapy, which is not related to the style of the article. The author took advantage of the fact that some readers did not understand English, so he stole English pictures that were irrelevant to the article in a vain attempt to deceive the readers.
Left: The "pseudo-popular science" article shows the experimental process of a patient taking garlic extract; Right: Translated, this image depicts CAR-T therapy.
"Pseudo-science" articles are generated using AI
It has also become a learning material for AI
The reporter noticed that these "pseudo-science" articles did not put forward pseudo-scientific views at the end of the article, saying that "the research results in the laboratory cannot be directly applied, and clinical trials and long-term follow-up are needed to verify their safety and efficacy".
With the mentality of trying it out, the reporter opened an AI chat software. When entering the command "Write a popular science article with a character story with the title of 'Drinking vinegar can remove 50% of blood clots within 24 hours'", the AI chat software quickly generated content very similar to the above-mentioned "pseudo-popular science" article, including introduction, story background, character introduction, story development, climax, ending, popular science time, etc., all available, and the reminders in the link are very similar. It can be seen that these "pseudo-popular" articles with similar structures are likely to be generated by self-media authors through integration and fabrication of AI-generated content.
The reporter typed in an AI chat software: "Write a popular science article with a character story with the title of 'Drinking vinegar can remove 50% of blood clots within 24 hours'", and the AI quickly generated an article.
Misappropriating AI-generated content, coupled with sensational stories made up by themselves, self-media authors have concocted a large number of "pseudo-popular science" articles and distributed them on the Internet, talking nonsense in a serious manner, such articles are likely to mislead readers. Moreover, when the reporter experienced it, he found that these "pseudo-popularization" content turned around and became learning materials for AI, interfering with the accuracy of AI-generated content.
The reporter further typed in the AI chat software, "Can acetic acid prevent thrombosis?" The relevant interface shows that a total of 19 web pages have been read for its generated content, including the above-mentioned "pseudo-science" article - "The study has found that drinking vinegar can remove 50% of blood clots within 24 hours, is it true?" Looking at the final answer, it is also obviously misled by the "pseudo-popular science" article.
The reporter asked in an AI chat software: "Can acetic acid prevent thrombosis?" Among the 19 web pages learned by AI is a "pseudo-science" article.
In recent years, with the development of AI technology, the application scenarios of artificial intelligence have penetrated into all walks of life. In response to this phenomenon, the Cyberspace Administration of China (CAC) recently issued the Measures for the Identification of Artificial Intelligence Generated and Synthetic Content (Draft for Comments), which clearly defines the explicit or implicit labels that need to be made in different situations, and also requires the platform party to assume corresponding responsibilities.
However, it will be some time before the measures are implemented. Here, the Shanghai Rumor Refutation Platform reminds users to be vigilant when using AI chat software, and pay attention to whether pseudoscience, rumors, bad or harmful information are mixed with AI learning materials. In the era of AI, the public needs to be more careful to distinguish AI-generated content and be more vigilant against bad information.
Source: Shanghai Internet Rumors, Changjiang Cloud News