The Yan Report is a misleading article masquerading as science, which falsely claims that the novel coronavirus was made in a Chinese lab. An example of cloaked science, it was released during a time of intense uncertainty; as scientists raced for answers about COVID-19, sharing unvetted data as preprints in open science repositories became an essential mode of international collaboration. The increasing openness of the scientific community, though, is a vulnerability that can be leveraged by media manipulators, especially during times of crisis. On April 28, 2020, Dr. Li-Meng Yan, a researcher at the University of Hong Kong (HKU), fled to the United States with support from Steve Bannon and Guo Wengui. They used Yan’s story — that she was a whistleblower — to exploit the contentious wedge issue of the unknown origin of COVID-19.
This media manipulation campaign involved planting misleading evidence into the scientific literature, muddying the waters about COVID-19 and providing the veneer of scientific legitimacy for the political claim that coronavirus was a Chinese bioweapon. Subsequently, the Yan Report was amplified through right-wing media networks, leading to nearly a million views of the report on Zenodo, an open-access research data repository. While social media platforms moderated information about the Yan Report after scientists at several universities debunked it, two follow-up Yan Reports were uploaded to open science repositories that even more bluntly pushed the bioweapon narrative, while also refuting the academic responses to the first report. Seeding the Yan reports in the scientific community as cloaked science allowed those who linked to them on social media to claim legitimacy, while also providing the empirical basis for furthering the political aims of the funders of the reports.
Stage 1: Manipulation Campaign Planning and Origins
Within a few weeks of the novel coronavirus spreading from China to the rest of the world, a pernicious narrative began to take root online: the suggestion that the virus SARS-CoV-2 was a biological weapon created in a lab.1
In mid-January 2020, Dr. Li-Meng Yan, a researcher at the University of Hong Kong (HKU), gave credence to this idea when she told her favorite YouTuber — Wang Dinggang, a vocal critic of the Chinese government, and close associate of exiled Chinese billionaire Guo Wengui2 — about rumors she had heard about the virus’ origins. Wang repeated the conversations on his channel without naming her “because officials could make the person disappear. ”3
On January 25, 2020, a hyperpartisan news outlet called G News published an article further pushing the bioweapon conspiracy theory. It was titled, “Breaking news: China will admit coronavirus coming from its P4 lab.”4 G News was not the only media outlet pushing the narrative, but its involvement was significant because G News is a media outlet associated with Guo and Steve Bannon, former Breitbart executive and ally of President Trump. The two have formed a partisan alliance to push their shared anti-CCP agenda through the Rule of Law Foundation and the Rule of Law Society, which they founded in October 2017.5 Funded by Guo and managed by Bannon, Rule of Law aims to “protect and assist individuals victimized in China, particularly those penalized for speaking out against injustice.”6 Guo and Bannon were drawn together because both “naturally despise the Chinese Communist Party (CCP),”7 according to Guo. They also both have media backgrounds. Bannon was previously Trump's chief strategist, and, before that he ran the right-wing news site Breitbart. Guo founded and funds G Media,8 which often posts anti-CCP stories across social media platforms, and prominently on the social media platform, Parler.
Prior to the first Yan report being released, Bannon said Wang’s YouTube episode featuring Yan was shown to and translated for him.9 As doubts about the origins of COVID-19 continued to proliferate across right-wing media networks, Guo and Bannon connected with Yan. This is when the pieces fell into place for what would become the Yan Report media manipulation campaign.
In interviews, Yan had been arguing that both China and the World Health Organization knew about the novel coronavirus earlier than they admitted. This is not a unique claim. Similar narratives, for example, were also circulating using the hashtag #chinaliedpeopledied, which went viral in March and April 2020.10 Yan put forward that she had evidence that the notion that the virus transferred from animals to humans was a “smokescreen” to hide its true origin, which she claims is a lab in Wuhan with close connections to the Chinese Communist Party.11 These claims are unfounded and have been debunked.12
While similar claims had been made by others, Yan stands out due to her background in science. Her CV says she has a medical degree from Xiangya Medical College of Central South University and a PhD from Southern Medical University.13 Prior to joining Bannon and Guo, Yan held a postdoctoral fellowship at Hong Kong University (HKU), where she was the first author on a COVID research paper that was published by Nature — one of the most prestigious biology journals in the world.14
Guo flew Yan to the US on April 28, 2020.15 In her most recent articles, she lists the Rule of Law Foundation and the Rule of Law Society as her formal affiliations.
In late April and early May, President Trump and Secretary of State Mike Pompeo fanned the flames of the January rumor that COVID-19 was made in a lab.16 Trump contradicted his own intelligence team by saying he had seen evidence and that he had a “high degree of confidence” that it was created and released in Wuhan.17
On July 10, 2020, Yan traded her story up the chain from online and hyperpartisan news outlets to an interview on Fox News.18 During the interview she said she was in hiding from the Chinese government and referred to herself as a whistleblower, claiming to have worked on human-to-human COVID transmission as early as December 2019.19 Fox, in turn, called her a whistleblower in its article about the interview.20 It is the earliest article labeling her as such.
The interview garnered a lot of attention online with over 2.7 million views on YouTube and 18K reactions on Facebook.21 The YouTube video has since been removed by Facebook according to data from CrowdTangle, a “public insights tool” owned by Facebook.22 The day after this interview, HKU put out a press release refuting what Yan told Fox News about her research.23
Yan continued to ramp up media appearances. On September 9, she repeated her claims to Raheem Kassam, co-host of Bannon’s popular podcast and YouTube show War Room: Pandemic, editor-in-chief of the conservative news site National Pulse and former editor-in-chief of Breitbart London.24 On September 11, Yan told Loose Women, a British talk show, that she would publish the evidence.25 The interview is available on YouTube, and has over 1.4M views.26 Then the New York Post quoted Yan’s interview with Loose Women,27 as did Tech Times (8.8K followers on Twitter),28 the Daily Mail (2.6M followers),29 and Mint (1.9M followers).30
With Yan, Guo and Bannon were able to effectively exploit the active crisis of the coronavirus by spreading her claims about its origins and the wedge issueof mistrust of the CCP. However, reporters were still skeptical of her claims and comments across social media repeatedly clamored for evidence.
A New York Post tweet publicizing Yan’s claims days before she uploaded her preprint. Screenshot by TaSC.
STAGE 2: Seeding Campaign Across Social Platforms and Web
All of this publicity put Yan’s name into circulation in the US and primed the public for the next phase of the campaign, which was the release of a preprint scientific paper.31 The paper exploited the vulnerability of open science to further muddy the waters about the origin of COVID-19 and push an anti-CCP narrative. The paper is an example of cloaked science — a discrete piece of scientific-looking misinformation — which gave credibility to conspiracies suggesting that COVID-19 was a bioweapon and that it had been engineered by China.
On September 14, 2020, Yan, along with three other names,32 released a preprint paper on an open-access research data repository called Zenodo. The title of the paper used scientific keywords central to the bioweapon misinformation narrative: “Unusual Features of the SARS-CoV-2 Genome Suggesting Sophisticated Laboratory Modification Rather Than Natural Evolution and Delineation of Its Probable Synthetic Route.”33 Its entire claim is in its title: COVID-19 was created in a lab. She joined Twitter and tweeted the Zenodo link to her preprint.34
Once considered controversial, the publication of preprints during the pandemic has increased in order to share data quickly during the crisis.35 But preprint publications can also be a vulnerability in the information ecosystem. Because preprints are not scrutinized in the same way as peer-reviewed journal articles and can be uploaded by anyone, they can become vectors for misinformation. “Preprints are the Facebook of science,” said Dr. Pamela Silver, who sits on the advisory board of bioRxiv, the premier preprint server for biology.36
Yan’s choices to release the paper as a preprint, and to use Zenodo as the terrain to disseminate it, appear strategic. Zenodo is hosted at the European Organization for Nuclear Research (CERN) and funded by OpenAIRE and the European Commission. Uploads to Zenodo are automatically assigned a DOI (Document Object Indicator), which gives it a baseline-level of legitimacy and a way for other scientists to cite it. Additionally, in contrast to bioRxiv, Zenodo is designed so that anyone with an email address can upload their ideas.37
BioRxiv also has a four-to-five day delay between author upload and public viewing, whereas Zenodo makes papers available immediately. If Yan had uploaded to bioRxiv, she would have had to wait while the team reviewed the uploaded preprint to determine whether the information was original and scientifically sound.38 Furthermore, on bioRxiv, authors need to be affiliated with an institution with an ORCID (an academic identifier), but Zenodo does not have the same requirement. Yan listed her affiliation on her Zenodo paper as Bannon and Guo’s Rule of Law Foundation and Rule of Law Society.
Once Yan had uploaded the preprint to Zenodo, it wasshared by others across multiple platforms. Raheem Kassam uploaded it to Scribd as “The Yan Report,” emphasizing her authorship. Using the same naming convention, it has been uploaded to YouTube as audio and added to publicly viewable Google Drives. Tweets and articles also describe the preprint as the “Yan Report.”
The speed of open science meant that Yan’s preprint could travel far — and quickly.39 On the day of its publication, Twitter mentions of Yan’s report were made by the likes of India’s WION News (292.5K followers)40 and News18 (4.5M followers);41 right-wing accounts such as ZeroHedge (842K followers);42 and retired NBCUniversal Senior Executive Mike Sington (24.2K followers).43
Peter Navarro (92.5K followers), Assistant to the President and Director of the Office of Trade and Manufacturing Policy, tweeted, “It would be unbelievable if it weren't so believable. #CCPLiedPeopleDied.” The tweet included a link to a ZeroHedge article about Yan joining Twitter and tweeting a link to her paper. It received 12.7K retweets and 19.1K likes.44 The Yan Report also sparked conversation in online forums, such as Reddit.45
According to a capture by the Internet Archive, Yan’s preprint had 156,769 views and 104,708 downloads on the day it was uploaded to Zenodo, making it instantly one of the most popular papers about COVID-19.
In an interview with National Geographic, Dr. Angela Rasumussen, a virologist at Columbia University, explained that Yan’s paper “looks legitimate because they use a lot of technical jargon. But in reality, a lot of what they're saying doesn't really make any sense.”46 Yan’s paper includes graphs, datasets, and cellular models, making it look like she has the evidence to back up her claims.
STAGE 3: Responses by Industry, Activists, Politicians, and Journalists
On September 15, the day after publication, Yan appeared on Tucker Carlson’s Fox News show. On the segment, she discussed her report. It was the fourth-most-watched program on TV that night, with an estimated 4.85 million viewers tuned in.47 G News released a statement about Yan’s paper that included a direct link to the paper.48
Carlson posted the clip of Yan’s appearance to Facebook and tweeted it. On Facebook, it has 73K reactions, 16K comments, and 78K shares.49 On Twitter, the clip has 56K likes and 27K retweets.50 The interview, also uploaded to YouTube, has around 2.8M views.51 The link was promoted by Carlson’s colleagues, such as Tammy Bruce, whose preview of the episode attracted 21.6K Twitter likes.52
For comparison, other Fox stories that day received much less attention. The top two results of a Facebook search show 1K and 2.3K engagements, respectively.53
On the day of the Carlson interview, Yan uploaded her publication to ResearchGate.54 Having the preprint on other repositories increases the likelihood that someone searching online would find the paper, and redundancy acts as a backup in case it is removed from one site. By September 16, the views on her Zenodo paper had reached 404,163, per the Internet Archive.
Bannon went on Tucker Carlson’s show two days after Yan, where he discussed Yan and her paper.55 He interviewed Yan on War Room: Pandemic 11 times between July 28 and October 2, 2020.56 Bannon and Kassam continue to host her and Navarro, a major booster of her report, periodically.57 She appeared on the show a total of 22 times in 2020, according to a count of the show notes on Pandemic: War Room’s website.58
A Tucker Carlson tweet that links to a clip from his interview with Yan. Screenshot by TaSC.
Tweet by Tammy Bruce publicizing Yan’s interview with Tucker Carlson. Screenshot by TaSC.
A tweet by Peter Navarro that links to the Yan Report. Screenshot by TaSC.
Conservative news sites covered the Yan Report uncritically, such as National Review,59 as well as the same media outlets that reported on Yan before her Zenodo upload: Fox,60 New York Post,61 and Tech Times.62 It also received international media exposure via reporting by Australia’s New.Com.Au,63 Spain’s AS English,64 and India’s WION.65
Influential conservatives voices continued to amplify the Yan Report on social media. Interview clips, commentary, and links were shared as Facebook posts, tweets, and retweets by Senator Marsha Blackburn, Rev. Franklin Graham, and President Trump,66 signifying a degree of political adoption, at least by some members of the president’s party.67 By arguing that COVID-19 was designed in a Chinese lab, the Yan Report affirms Trump’s repeated claim that “it’s China’s fault.”68
An analysis by TaSC found that links to the preprint have been tweeted and retweeted 15,731 times. Just three tweets (two from Yan and one from Navarro) received two-thirds of the retweets.69
The graph (shown below) charts the virality of the Yan preprint. Yan’s preprint received about 850,000 views between its upload and October 8, the date she uploaded her next report. This graph is based on data captured by the Internet Archive. To put the Yan Report’s virality in context, it is important to note that other papers uploaded around this time have only a few hundred views.
As of February 10, 2021, the Yan Report had 1,043,337 views and 735,879 downloads.70
Total daily views (blue line) and downloads (red line) of the first Yan Report from the day of its upload (September 14, 2020) through to the publication of this case study (February 10, 2021). Data was taken from the Internet Archive captures of the Zenodo page hosting the Yan Report, which refreshes its page views and downloads once daily. The numbers are the same on September 19, 20, and 21 as Zenodo had a small outage on those days. Note that no data is available for the following dates: September 28, 29, October 26, November 14, 22, 24, 28, 30, December 7, 10, 14, 22, 27, and January 3, 13, 14, 15, 18, 21, 25. Chart by TaSC.
STAGE 4: Mitigation efforts
The media attention that the Yan Report received led academic scientists to issue refutations of the preprint, and that analysis was then covered by critical press. There is scientific and journalistic consensus that the content of Yan’s preprint is not convincing, and furthermore, is medical misinformation.71
A week after its upload to Zenodo, the preprint was fact-checked by Johns Hopkins’ Center for Health Security, which said that the paper “offers contradictory and inaccurate information that does not support their argument.”72 Another problem with the paper is that it cites “multiple papers in their reference section that have weaknesses or flaws to make their case.”73
In an interview with CNN, Nancy D. Connell of Johns Hopkins, explained why her team decided to engage with the paper: “It was clear on social media that the paper was getting more and more attention. We talked carefully and thought for a long time whether to do it.”74
On September 24, MIT scientists concluded,“This manuscript does not demonstrate sufficient scientific evidence to support its claims. Claims are at times baseless and are not supported by the data and methods used. Decision-makers should consider the author's claims in this study misleading.”75 The Yan Report was also fact-checked by scientists at Columbia University,76 Carnegie Mellon University, and the University of Washington.77
In addition to critical press and debunking, Twitter suspended Yan’s Twitter account two days after she created it to share the preprint, though Twitter made no public statement about the cause of the suspension.78 Twitter also de-indexedthe longform version of the link to her preprint, so that no results appear when it is searched on Twitter. Shorter versions of the link, however, still lead to tweets about the preprint.
Facebook and Instagram added warnings to Tucker Carlson’s posts about his interview with Yan. YouTube videos — from Tucker Carlson, Fox News, and Loose Women — all have a small banner across the bottom with an option to view CDC information.79
Screenshot of Tucker Carlson’s Facebook post sharing his interview with Yan. Credit: TaSC.
Social media moderation, however, did not quell the interest in the paper. After attempts by the platforms to reduce the report’s spread, Infowars posted an article saying that Facebook had “censored” the Tucker Carlson interview.80 Sara Carter, a Fox commentator, also brought attention to content moderation and urged her followers to watch the interview clip.81 The New York Times also reported on social media responses.82
Yan has claimed that censorship kept her work at the preprint stage,83 though we have found no evidence that Yan and her named co-authors sought publication in any scientific journal, as customary with a preprint like this. Instead, these claims of censorship echo media manipulation strategies that make unfounded claims about bias in order to create an aura of illicit information.84
In an interview with Zenodo’s curators and representatives from OpenAire, the team told TaSC they weighed removing the preprint from the repository. They decided against it because it would be antithetical to the spirit of their platform — and their terms of service. They state that an upload isn’t an endorsement. Rather, they maintain Zenodo is a platform where scientific conversation can begin. Keeping the Yan Report online, they said, means that it can be discussed and debunked. When scientists fact-check it, they have a paper to reference and a DOI to cite. However, the site currently does not provide a discussion forum or labels that identify when there is significant refutation of content on their site.85
STAGE 5: Adjustments by campaign operators
Yan published a second paper on October 8, however, with sometactical adjustment,specifically in the language of the title and the content of the paper. The title of this second paper is more straightforward, and less scientific-sounding: “SARS-CoV-2 Is an Unrestricted Bioweapon: A Truth Revealed through Uncovering a Large-Scale, Organized Scientific Fraud.” This title covers top keyword search terms for disinformation about bioweapons, which topped the most shared stories this year. Per The New York Times, “media intelligence platform” Zignal Labs found that content about bioweapons accounted for 24.2 percent of the misinformation spread in 2020.86
This second paper was uploaded to Zenodo, but unlike the first one, was not accepted to the site’s Coronavirus Disease Research Community - COVID-19. The second Yan Report was also uploaded to ResearchGate, and it is part of the site’s Coronavirus Community.87
As part of her tactical adjustment, Yan created a second Twitter account in order to tweet her second preprint. During the observation period, it received 16.3K likes and 14.1K retweets and quote tweets.88 It was immediately shared by right-wing Twitter accounts, such as ZeroHedge, which posted the link along with a mention of how quickly Yan’s account was removed after her last paper.
Anticipating more platform moderation, Yan also tweeted a link to her Gab account (should her second Twitter account be removed), which represents further adjustment. Gab is a minor platform frequented by white supremacists, conspiracists, and others who seek unmoderated online discussion forums. As of February 10, 2021, Yan’s latest Twitter account was still active and has nearly 76K followers.89 Her second paper has 229,918 views on Zenodo and has been downloaded 117,651 times.90
Notably, for the second Yan Report, CNN accused Yan of copying passages from an anonymous blogger into her preprint.91 This resulted in a third working paper, which is a direct response to CNN. Also uploaded to Zenodo, it is titled “CNN Used Lies and Misinformation to Muddle the Water on the Origin of SARS-CoV-2.”92 The paper attempts to refute CNN’s plagiarism claims and contends that CNN used its media power to discredit Yan’s work and reputation. Uploaded on November 21, 2020, this paper has received only 30,080 views and 9,403 downloads. Unlike the previous reports that included misleading science, this upload appears to be a press release, wherein Zenodo is acting as part of the distribution infrastructure for Yan’s public replies.
Conclusion
By exploiting open science93 during a health crisis to further their political aims, Bannon and Guo utilized Yan’s status as a research scientist fleeing Hong Kong to sustain public attention to the “COVID-19 as a bioweapon” narrative. While data repositories and preprint servers have provided crucial infrastructure for international collaboration on COVID-19, like other online platforms, they can also be used in disinformation campaigns, due to the legitimacy they confer by association. When lay publics and some journalists view these sites, they may inadvertently assume the content has been vetted or assessed in some official capacity, and is therefore sound science. Cloaked science can be particularly deceiving when it is surrounded by research from top tier scientists, universities, and institutes.
锦心，让我喋喋不休地唠叨，好剧吗，还是想和朋友们讨论的，谢谢大家理解。
编剧吗，估计是商业模式催化的，需要又长又长的集数，水水啦！
除了老妇人和冬青，其他都演得好！
我追到39集，现在因为迷乒乓暂时停一下。等完结了再一口气看完。
里算中等长度，毕竟几个家族角色众多，这剧算是古装家族群像剧。《爱的迫降》我也很喜欢。其实言情剧，都有点不真实，有点小套路。那为什么好的言情剧很少见呢？一个是编剧一个是演员，这才是互相成就。李政赫同志有幸遇见了好编剧，徐令宜运气稍微差一点。
爱的迫降我也喜欢，题材新颖，演员也给力。
链接没有了，试试这个
Overview
The Yan Report is a misleading article masquerading as science, which falsely claims that the novel coronavirus was made in a Chinese lab. An example of cloaked science, it was released during a time of intense uncertainty; as scientists raced for answers about COVID-19, sharing unvetted data as preprints in open science repositories became an essential mode of international collaboration. The increasing openness of the scientific community, though, is a vulnerability that can be leveraged by media manipulators, especially during times of crisis. On April 28, 2020, Dr. Li-Meng Yan, a researcher at the University of Hong Kong (HKU), fled to the United States with support from Steve Bannon and Guo Wengui. They used Yan’s story — that she was a whistleblower — to exploit the contentious wedge issue of the unknown origin of COVID-19.
This media manipulation campaign involved planting misleading evidence into the scientific literature, muddying the waters about COVID-19 and providing the veneer of scientific legitimacy for the political claim that coronavirus was a Chinese bioweapon. Subsequently, the Yan Report was amplified through right-wing media networks, leading to nearly a million views of the report on Zenodo, an open-access research data repository. While social media platforms moderated information about the Yan Report after scientists at several universities debunked it, two follow-up Yan Reports were uploaded to open science repositories that even more bluntly pushed the bioweapon narrative, while also refuting the academic responses to the first report. Seeding the Yan reports in the scientific community as cloaked science allowed those who linked to them on social media to claim legitimacy, while also providing the empirical basis for furthering the political aims of the funders of the reports.
Stage 1: Manipulation Campaign Planning and OriginsWithin a few weeks of the novel coronavirus spreading from China to the rest of the world, a pernicious narrative began to take root online: the suggestion that the virus SARS-CoV-2 was a biological weapon created in a lab.1
In mid-January 2020, Dr. Li-Meng Yan, a researcher at the University of Hong Kong (HKU), gave credence to this idea when she told her favorite YouTuber — Wang Dinggang, a vocal critic of the Chinese government, and close associate of exiled Chinese billionaire Guo Wengui2 — about rumors she had heard about the virus’ origins. Wang repeated the conversations on his channel without naming her “because officials could make the person disappear. ”3
On January 25, 2020, a hyperpartisan news outlet called G News published an article further pushing the bioweapon conspiracy theory. It was titled, “Breaking news: China will admit coronavirus coming from its P4 lab.”4 G News was not the only media outlet pushing the narrative, but its involvement was significant because G News is a media outlet associated with Guo and Steve Bannon, former Breitbart executive and ally of President Trump. The two have formed a partisan alliance to push their shared anti-CCP agenda through the Rule of Law Foundation and the Rule of Law Society, which they founded in October 2017.5 Funded by Guo and managed by Bannon, Rule of Law aims to “protect and assist individuals victimized in China, particularly those penalized for speaking out against injustice.”6 Guo and Bannon were drawn together because both “naturally despise the Chinese Communist Party (CCP),”7 according to Guo. They also both have media backgrounds. Bannon was previously Trump's chief strategist, and, before that he ran the right-wing news site Breitbart. Guo founded and funds G Media,8 which often posts anti-CCP stories across social media platforms, and prominently on the social media platform, Parler.
Prior to the first Yan report being released, Bannon said Wang’s YouTube episode featuring Yan was shown to and translated for him.9 As doubts about the origins of COVID-19 continued to proliferate across right-wing media networks, Guo and Bannon connected with Yan. This is when the pieces fell into place for what would become the Yan Report media manipulation campaign.
In interviews, Yan had been arguing that both China and the World Health Organization knew about the novel coronavirus earlier than they admitted. This is not a unique claim. Similar narratives, for example, were also circulating using the hashtag #chinaliedpeopledied, which went viral in March and April 2020.10 Yan put forward that she had evidence that the notion that the virus transferred from animals to humans was a “smokescreen” to hide its true origin, which she claims is a lab in Wuhan with close connections to the Chinese Communist Party.11 These claims are unfounded and have been debunked.12
While similar claims had been made by others, Yan stands out due to her background in science. Her CV says she has a medical degree from Xiangya Medical College of Central South University and a PhD from Southern Medical University.13 Prior to joining Bannon and Guo, Yan held a postdoctoral fellowship at Hong Kong University (HKU), where she was the first author on a COVID research paper that was published by Nature — one of the most prestigious biology journals in the world.14
Guo flew Yan to the US on April 28, 2020.15 In her most recent articles, she lists the Rule of Law Foundation and the Rule of Law Society as her formal affiliations.
In late April and early May, President Trump and Secretary of State Mike Pompeo fanned the flames of the January rumor that COVID-19 was made in a lab.16 Trump contradicted his own intelligence team by saying he had seen evidence and that he had a “high degree of confidence” that it was created and released in Wuhan.17
On July 10, 2020, Yan traded her story up the chain from online and hyperpartisan news outlets to an interview on Fox News.18 During the interview she said she was in hiding from the Chinese government and referred to herself as a whistleblower, claiming to have worked on human-to-human COVID transmission as early as December 2019.19 Fox, in turn, called her a whistleblower in its article about the interview.20 It is the earliest article labeling her as such.
The interview garnered a lot of attention online with over 2.7 million views on YouTube and 18K reactions on Facebook.21 The YouTube video has since been removed by Facebook according to data from CrowdTangle, a “public insights tool” owned by Facebook.22 The day after this interview, HKU put out a press release refuting what Yan told Fox News about her research.23
Yan continued to ramp up media appearances. On September 9, she repeated her claims to Raheem Kassam, co-host of Bannon’s popular podcast and YouTube show War Room: Pandemic, editor-in-chief of the conservative news site National Pulse and former editor-in-chief of Breitbart London.24 On September 11, Yan told Loose Women, a British talk show, that she would publish the evidence.25 The interview is available on YouTube, and has over 1.4M views.26 Then the New York Post quoted Yan’s interview with Loose Women,27 as did Tech Times (8.8K followers on Twitter),28 the Daily Mail (2.6M followers),29 and Mint (1.9M followers).30
With Yan, Guo and Bannon were able to effectively exploit the active crisis of the coronavirus by spreading her claims about its origins and the wedge issue of mistrust of the CCP. However, reporters were still skeptical of her claims and comments across social media repeatedly clamored for evidence.
A New York Post tweet publicizing Yan’s claims days before she uploaded her preprint. Screenshot by TaSC.
STAGE 2: Seeding Campaign Across Social Platforms and WebAll of this publicity put Yan’s name into circulation in the US and primed the public for the next phase of the campaign, which was the release of a preprint scientific paper.31 The paper exploited the vulnerability of open science to further muddy the waters about the origin of COVID-19 and push an anti-CCP narrative. The paper is an example of cloaked science — a discrete piece of scientific-looking misinformation — which gave credibility to conspiracies suggesting that COVID-19 was a bioweapon and that it had been engineered by China.
On September 14, 2020, Yan, along with three other names,32 released a preprint paper on an open-access research data repository called Zenodo. The title of the paper used scientific keywords central to the bioweapon misinformation narrative: “Unusual Features of the SARS-CoV-2 Genome Suggesting Sophisticated Laboratory Modification Rather Than Natural Evolution and Delineation of Its Probable Synthetic Route.”33 Its entire claim is in its title: COVID-19 was created in a lab. She joined Twitter and tweeted the Zenodo link to her preprint.34
Once considered controversial, the publication of preprints during the pandemic has increased in order to share data quickly during the crisis.35 But preprint publications can also be a vulnerability in the information ecosystem. Because preprints are not scrutinized in the same way as peer-reviewed journal articles and can be uploaded by anyone, they can become vectors for misinformation. “Preprints are the Facebook of science,” said Dr. Pamela Silver, who sits on the advisory board of bioRxiv, the premier preprint server for biology.36
Yan’s choices to release the paper as a preprint, and to use Zenodo as the terrain to disseminate it, appear strategic. Zenodo is hosted at the European Organization for Nuclear Research (CERN) and funded by OpenAIRE and the European Commission. Uploads to Zenodo are automatically assigned a DOI (Document Object Indicator), which gives it a baseline-level of legitimacy and a way for other scientists to cite it. Additionally, in contrast to bioRxiv, Zenodo is designed so that anyone with an email address can upload their ideas.37
BioRxiv also has a four-to-five day delay between author upload and public viewing, whereas Zenodo makes papers available immediately. If Yan had uploaded to bioRxiv, she would have had to wait while the team reviewed the uploaded preprint to determine whether the information was original and scientifically sound.38 Furthermore, on bioRxiv, authors need to be affiliated with an institution with an ORCID (an academic identifier), but Zenodo does not have the same requirement. Yan listed her affiliation on her Zenodo paper as Bannon and Guo’s Rule of Law Foundation and Rule of Law Society.
Once Yan had uploaded the preprint to Zenodo, it was shared by others across multiple platforms. Raheem Kassam uploaded it to Scribd as “The Yan Report,” emphasizing her authorship. Using the same naming convention, it has been uploaded to YouTube as audio and added to publicly viewable Google Drives. Tweets and articles also describe the preprint as the “Yan Report.”
The speed of open science meant that Yan’s preprint could travel far — and quickly.39 On the day of its publication, Twitter mentions of Yan’s report were made by the likes of India’s WION News (292.5K followers)40 and News18 (4.5M followers);41 right-wing accounts such as ZeroHedge (842K followers);42 and retired NBCUniversal Senior Executive Mike Sington (24.2K followers).43
Peter Navarro (92.5K followers), Assistant to the President and Director of the Office of Trade and Manufacturing Policy, tweeted, “It would be unbelievable if it weren't so believable. #CCPLiedPeopleDied.” The tweet included a link to a ZeroHedge article about Yan joining Twitter and tweeting a link to her paper. It received 12.7K retweets and 19.1K likes.44 The Yan Report also sparked conversation in online forums, such as Reddit.45
According to a capture by the Internet Archive, Yan’s preprint had 156,769 views and 104,708 downloads on the day it was uploaded to Zenodo, making it instantly one of the most popular papers about COVID-19.
In an interview with National Geographic, Dr. Angela Rasumussen, a virologist at Columbia University, explained that Yan’s paper “looks legitimate because they use a lot of technical jargon. But in reality, a lot of what they're saying doesn't really make any sense.”46 Yan’s paper includes graphs, datasets, and cellular models, making it look like she has the evidence to back up her claims.
STAGE 3: Responses by Industry, Activists, Politicians, and JournalistsOn September 15, the day after publication, Yan appeared on Tucker Carlson’s Fox News show. On the segment, she discussed her report. It was the fourth-most-watched program on TV that night, with an estimated 4.85 million viewers tuned in.47 G News released a statement about Yan’s paper that included a direct link to the paper.48
Carlson posted the clip of Yan’s appearance to Facebook and tweeted it. On Facebook, it has 73K reactions, 16K comments, and 78K shares.49 On Twitter, the clip has 56K likes and 27K retweets.50 The interview, also uploaded to YouTube, has around 2.8M views.51 The link was promoted by Carlson’s colleagues, such as Tammy Bruce, whose preview of the episode attracted 21.6K Twitter likes.52
For comparison, other Fox stories that day received much less attention. The top two results of a Facebook search show 1K and 2.3K engagements, respectively.53
On the day of the Carlson interview, Yan uploaded her publication to ResearchGate.54 Having the preprint on other repositories increases the likelihood that someone searching online would find the paper, and redundancy acts as a backup in case it is removed from one site. By September 16, the views on her Zenodo paper had reached 404,163, per the Internet Archive.
Bannon went on Tucker Carlson’s show two days after Yan, where he discussed Yan and her paper.55 He interviewed Yan on War Room: Pandemic 11 times between July 28 and October 2, 2020.56 Bannon and Kassam continue to host her and Navarro, a major booster of her report, periodically.57 She appeared on the show a total of 22 times in 2020, according to a count of the show notes on Pandemic: War Room’s website.58
A Tucker Carlson tweet that links to a clip from his interview with Yan. Screenshot by TaSC.
Tweet by Tammy Bruce publicizing Yan’s interview with Tucker Carlson. Screenshot by TaSC.
A tweet by Peter Navarro that links to the Yan Report. Screenshot by TaSC.
Conservative news sites covered the Yan Report uncritically, such as National Review,59 as well as the same media outlets that reported on Yan before her Zenodo upload: Fox,60 New York Post,61 and Tech Times.62 It also received international media exposure via reporting by Australia’s New.Com.Au,63 Spain’s AS English,64 and India’s WION.65
Influential conservatives voices continued to amplify the Yan Report on social media. Interview clips, commentary, and links were shared as Facebook posts, tweets, and retweets by Senator Marsha Blackburn, Rev. Franklin Graham, and President Trump,66 signifying a degree of political adoption, at least by some members of the president’s party.67 By arguing that COVID-19 was designed in a Chinese lab, the Yan Report affirms Trump’s repeated claim that “it’s China’s fault.”68
An analysis by TaSC found that links to the preprint have been tweeted and retweeted 15,731 times. Just three tweets (two from Yan and one from Navarro) received two-thirds of the retweets.69
The graph (shown below) charts the virality of the Yan preprint. Yan’s preprint received about 850,000 views between its upload and October 8, the date she uploaded her next report. This graph is based on data captured by the Internet Archive. To put the Yan Report’s virality in context, it is important to note that other papers uploaded around this time have only a few hundred views.
As of February 10, 2021, the Yan Report had 1,043,337 views and 735,879 downloads.70
Total daily views (blue line) and downloads (red line) of the first Yan Report from the day of its upload (September 14, 2020) through to the publication of this case study (February 10, 2021). Data was taken from the Internet Archive captures of the Zenodo page hosting the Yan Report, which refreshes its page views and downloads once daily. The numbers are the same on September 19, 20, and 21 as Zenodo had a small outage on those days. Note that no data is available for the following dates: September 28, 29, October 26, November 14, 22, 24, 28, 30, December 7, 10, 14, 22, 27, and January 3, 13, 14, 15, 18, 21, 25. Chart by TaSC.
STAGE 4: Mitigation effortsThe media attention that the Yan Report received led academic scientists to issue refutations of the preprint, and that analysis was then covered by critical press. There is scientific and journalistic consensus that the content of Yan’s preprint is not convincing, and furthermore, is medical misinformation.71
A week after its upload to Zenodo, the preprint was fact-checked by Johns Hopkins’ Center for Health Security, which said that the paper “offers contradictory and inaccurate information that does not support their argument.”72 Another problem with the paper is that it cites “multiple papers in their reference section that have weaknesses or flaws to make their case.”73
In an interview with CNN, Nancy D. Connell of Johns Hopkins, explained why her team decided to engage with the paper: “It was clear on social media that the paper was getting more and more attention. We talked carefully and thought for a long time whether to do it.”74
On September 24, MIT scientists concluded,“This manuscript does not demonstrate sufficient scientific evidence to support its claims. Claims are at times baseless and are not supported by the data and methods used. Decision-makers should consider the author's claims in this study misleading.”75 The Yan Report was also fact-checked by scientists at Columbia University,76 Carnegie Mellon University, and the University of Washington.77
In addition to critical press and debunking, Twitter suspended Yan’s Twitter account two days after she created it to share the preprint, though Twitter made no public statement about the cause of the suspension.78 Twitter also de-indexed the longform version of the link to her preprint, so that no results appear when it is searched on Twitter. Shorter versions of the link, however, still lead to tweets about the preprint.
Facebook and Instagram added warnings to Tucker Carlson’s posts about his interview with Yan. YouTube videos — from Tucker Carlson, Fox News, and Loose Women — all have a small banner across the bottom with an option to view CDC information.79
Screenshot of Tucker Carlson’s Facebook post sharing his interview with Yan. Credit: TaSC.
Social media moderation, however, did not quell the interest in the paper. After attempts by the platforms to reduce the report’s spread, Infowars posted an article saying that Facebook had “censored” the Tucker Carlson interview.80 Sara Carter, a Fox commentator, also brought attention to content moderation and urged her followers to watch the interview clip.81 The New York Times also reported on social media responses.82
Yan has claimed that censorship kept her work at the preprint stage,83 though we have found no evidence that Yan and her named co-authors sought publication in any scientific journal, as customary with a preprint like this. Instead, these claims of censorship echo media manipulation strategies that make unfounded claims about bias in order to create an aura of illicit information.84
In an interview with Zenodo’s curators and representatives from OpenAire, the team told TaSC they weighed removing the preprint from the repository. They decided against it because it would be antithetical to the spirit of their platform — and their terms of service. They state that an upload isn’t an endorsement. Rather, they maintain Zenodo is a platform where scientific conversation can begin. Keeping the Yan Report online, they said, means that it can be discussed and debunked. When scientists fact-check it, they have a paper to reference and a DOI to cite. However, the site currently does not provide a discussion forum or labels that identify when there is significant refutation of content on their site.85
STAGE 5: Adjustments by campaign operatorsYan published a second paper on October 8, however, with some tactical adjustment, specifically in the language of the title and the content of the paper. The title of this second paper is more straightforward, and less scientific-sounding: “SARS-CoV-2 Is an Unrestricted Bioweapon: A Truth Revealed through Uncovering a Large-Scale, Organized Scientific Fraud.” This title covers top keyword search terms for disinformation about bioweapons, which topped the most shared stories this year. Per The New York Times, “media intelligence platform” Zignal Labs found that content about bioweapons accounted for 24.2 percent of the misinformation spread in 2020.86
This second paper was uploaded to Zenodo, but unlike the first one, was not accepted to the site’s Coronavirus Disease Research Community - COVID-19. The second Yan Report was also uploaded to ResearchGate, and it is part of the site’s Coronavirus Community.87
As part of her tactical adjustment, Yan created a second Twitter account in order to tweet her second preprint. During the observation period, it received 16.3K likes and 14.1K retweets and quote tweets.88 It was immediately shared by right-wing Twitter accounts, such as ZeroHedge, which posted the link along with a mention of how quickly Yan’s account was removed after her last paper.
Anticipating more platform moderation, Yan also tweeted a link to her Gab account (should her second Twitter account be removed), which represents further adjustment. Gab is a minor platform frequented by white supremacists, conspiracists, and others who seek unmoderated online discussion forums. As of February 10, 2021, Yan’s latest Twitter account was still active and has nearly 76K followers.89 Her second paper has 229,918 views on Zenodo and has been downloaded 117,651 times.90
Notably, for the second Yan Report, CNN accused Yan of copying passages from an anonymous blogger into her preprint.91 This resulted in a third working paper, which is a direct response to CNN. Also uploaded to Zenodo, it is titled “CNN Used Lies and Misinformation to Muddle the Water on the Origin of SARS-CoV-2.”92 The paper attempts to refute CNN’s plagiarism claims and contends that CNN used its media power to discredit Yan’s work and reputation. Uploaded on November 21, 2020, this paper has received only 30,080 views and 9,403 downloads. Unlike the previous reports that included misleading science, this upload appears to be a press release, wherein Zenodo is acting as part of the distribution infrastructure for Yan’s public replies.
ConclusionBy exploiting open science93 during a health crisis to further their political aims, Bannon and Guo utilized Yan’s status as a research scientist fleeing Hong Kong to sustain public attention to the “COVID-19 as a bioweapon” narrative. While data repositories and preprint servers have provided crucial infrastructure for international collaboration on COVID-19, like other online platforms, they can also be used in disinformation campaigns, due to the legitimacy they confer by association. When lay publics and some journalists view these sites, they may inadvertently assume the content has been vetted or assessed in some official capacity, and is therefore sound science. Cloaked science can be particularly deceiving when it is surrounded by research from top tier scientists, universities, and institutes.