This phenomenon is not merely a return to old habits or rural mysticism. It has become a nationwide consumer frenzy, driven by the very demographic the Communist Party hoped would be its most rational constituency: the young and educated. In chasing these modern symbols of hope, they are losing more than just money.

[…] In today’s China, the most popular “spiritual” items aren’t books or teachings but small objects—especially crystals. These are sold not only as fashion items but as tools for cosmic power. Supposedly, they bring wealth, block bad energy, and balance inner forces. Livestreams offer quick lessons in “crystal basics,” and influencers promote them with the excitement once shown for new tech.

[…] Along with the crystal craze, astrology, tarot, and fortune-telling have become small but growing businesses. Highly educated youth—graduates, civil servants, tech workers—are quitting their jobs to become full-time “mystics.” On platforms like Taobao and WeChat, paid readings are everywhere. In many cities, you’ll find stylish little shops doing tarot readings, often run by baristas turned fortune-tellers.

This is happening despite—or maybe because of—government crackdowns. In 2021, China banned religious content on e-commerce sites and tightened rules on spiritual services. But the demand only adapted. Tarot readers now call themselves “emotional consultants.” Horoscope sellers move to foreign platforms like Discord. The state fights superstition with censorship, and loses every time.

[…] It would be wrong to see this wave of superstition as a uniquely Chinese flaw. But since 2024, China’s superstition boom has become a pressure cooker where many deep problems have gathered: economic slowdown, job stress, burnout, pushy online systems, and a desperate need for meaning.

Young Chinese are not naturally more superstitious. But they are trapped in an unstable system, and with no clear future, they are buying ready-made ones. These crystals and tarot cards aren’t ancient traditions—they’re quick-fix stories built from what’s left in the marketplace. Meanwhile, sellers and platforms continue testing how much people are willing to pay to ease their fears. [Source]

The Economist in January similarly described trends such as app-based horoscopes and fortune-telling and offline “metaphysical bars,” fueled by frustration at “a sluggish economy, a tight job market and intense competition in many aspects of life.” (Another Economist report the week before noted similar phenomena in the U.S. and India.)

The Communist Party has long tried to rid itself of what it calls “feudal superstition”. Last year the Central Party School, a training academy for officials, expressed concern about the number of members and cadres “believing in ghosts and gods”. It tried to clarify the party’s restrictions by publishing a Q&A on the matter. Occasionally participating in local folk customs or consulting a fortune-teller on a name for your baby? That’s fine. Spending a lot of time and money, especially public funds, on superstitious activities? Unacceptable.

The masses are also discouraged from embracing such practices. A notice issued by the city of Sanming in 2023 stated: “The public should improve their scientific literacy, enhance their psychological immunity to superstitious activities and not seek spiritual comfort through ‘fortune-telling’ when encountering real setbacks.” Other cities have followed suit. Last year some local governments cracked down on the burning of fake money and other paper offerings to the dead during the annual grave-sweeping festival.

State censors, with the help of internet firms, have tried to curb the spread of superstitious beliefs and divination services online. Search terms such as “astrology” and “fortune-telling” have been blocked on Taobao, an e-commerce market. But on Weibo, a social-media site, popular astrologers have accumulated tens of millions of followers. Some speak of playing a cat-and-mouse game with the authorities. A 24-year-old tarot-card reader in Shanghai jokes that she tries to divine her own fate—to see if jail time is in the offing. [Source]

There is also online hay to be made from confronting superstition. In April, South China Morning Post’s Zoey Zhang reported on Shandong-based influencer Zhang Shulin, who has built a following with video stunts debunking beliefs such as hauntings, shamanism, and ghost marriages. This, too, can be a hazardous approach if targets include traditional practices favored with official endorsement, however. Mixed martial artist Xu Xiaodong was hit with censorship, travel restrictions, financial penalties, and forced apologies following his efforts to puncture the inflated claims of purported kung fu masters, some of whom he flattened in bouts lasting only seconds. In 2022, a number of prominent online voices were silenced in apparent retaliation for their criticism of Lianhua Qingwen, a traditional Chinese medicine-based herbal product promoted by Chinese authorities for treatment of COVID.

A pair of translations at CDT last month described how other frustrated young Chinese are turning to another old ritual: the annual civil service exams.

The controversy entered the public eye in late April, courtesy of a leaked letter from the estranged wife of Dr. Xiao Fei, a thoracic surgeon at Beijing’s prestigious China-Japan Friendship Hospital, to the hospital disciplinary committee. The letter contained details of the doctor’s alleged affairs with several colleagues (including one Dong Xiying, a young resident at the hospital), and an allegation that Dr. Xiao left a patient anesthetized on the operating table for 40 minutes while he left the operating theatre to comfort Ms. Dong. After a brief investigation, Dr. Xiao was sacked by the hospital and expelled from the Chinese Communist Party.

But that was simply the first act in what would become a much larger controversy. Internet sleuths who dug into Ms. Dong’s background discovered that she was a “returnee” who had earned an economics degree at Barnard College in the U.S., was from a fairly influential family background, and had enjoyed an academic and career trajectory that may have been helped along by nepotism and string-pulling. Perhaps most controversial was Dong’s rapid rise via the “4+4” accelerated-degree pilot program at Peking Union Medical College (PUMC), which allows a small number of “elite” university graduates—even those with undergraduate degrees unrelated to medicine or biology—to attain a medical degree in only four years, a much shorter timeline than is typical for medical students in China. Netizens also raised questions about Ms. Dong’s publication history, including a graduate thesis that was suspiciously short, and author credits on research papers for projects she didn’t seem to have played much of a role in. (For more background on the multifaceted scandal, we recommend What’s On Weibo’s excellent account of the key protagonists and events.)

Discussion of the scandal proved so popular that at one point, it accounted for more than half of the top 50 “hot search” topics on Weibo, according to WeChat blogger “History Rhymes.” But just a few days later, as the blogger noted on May 5, they had disappeared from the list:

I checked Weibo’s “hot search” list today, and there are no longer any topics about Miss Dong, Peking Union Medical College, etc.

Keep in mind that just a couple of days ago, more than half of the top 50 “hot search” topics were about or related to her.

But just because it’s not trending, doesn’t mean that people aren’t discussing it. Netizens are still digging into the matter. [Chinese]

In addition to Weibo apparently muting the topic, there was also self-censorship on the part of PUMC, which removed content related to Ms. Dong from its website and edited her name out of a 2023 commencement speech given by the college president. Ms. Dong’s graduate thesis and other publications mysteriously disappeared from the academic database portal CNKI (China National Knowledge Infrastructure). After their removal was noticed, it briefly became the top trending search topic on Weibo.

CDT Chinese editors have archived 21 articles and essays related to the “4+4” scandal and its various corollaries; at least three of these have since been censored. The first of these deleted articles appeared on May 1, under the headline, “Could Miss Dong’s Family Be Considered Beijing Brahmins?” Written by journalist Wang Mingyuan, who runs the WeChat public account Fuchengmen Courtyard No. 6, the article argues that Ms. Dong’s family are simply upper middle class, not highly privileged cadres, suggesting that the kind of string-pulling behind her career could be even more pervasive and concerning. Wang’s article also includes a widely circulated (and now censored) meme poking fun at hospital corruption. In the now iconic cartoon, every doctor, nurse, patient, lamp, and piece of medical equipment in the operating theater claims to have gotten into the hospital by leveraging connections:

Patient on the operating table: I got in here through connections.
Doctor: Me, too.
Row of nurses: I did, too. Me too! Same here.
Various medicine cabinets and items of medical equipment: Me, too.
Overhead surgical lamp: Same here!
Another item of medical equipment, with a thought bubble: Hey, didn’t we all?
(source: Wang Mingyuan/WeChat)

Another now-deleted article, published on May 4 by WeChat account Elephant News, provides details about PUMC’s accelerated “4+4” doctoral degree program, and compares it to the usual "5+3+4" route for Chinese medical students: five years of undergraduate-level medical education, followed by three years of master’s-level medical coursework and four years of doctoral-level medical coursework. The author notes how unusual it was that Dong Xiying, whose undergraduate degree was in economics, was allowed to help perform surgery when she was only in her second year of medical school. The article also includes a screenshot showing angry comments left under a PUMC social media account, with netizens complaining that going to see a doctor now feels like “making a holy pilgrimage,” expressing concerns that their doctors might not be qualified if they graduated from the accelerated “4+4” program, and demanding that PUMC make the list of “4+4” program participants public.

On May 5, CDT Chinese editors archived a deleted WeChat article by Sina Finance, which reproduced the answer given by Tencent’s Yuanbao AI chatbot in response to a Sina editor’s query about what other individuals might be implicated in violating the regulations of PUMC’s “4+4” program. The chatbot gave a detailed answer organized into four headings, the first of which listed known participants in the “4+4” program. The second listed individuals who may have benefited from personal connections or affiliations. The third section, enumerating some characteristic examples of systemic privilege, claimed that “35% of the ‘4+4’ program participants have parents who are departmental-level or higher-level cadres, which far exceeds the proportion found among students in typical medical school programs (2.1%).” The chatbot also claimed to have found admission loopholes (“Some of the ‘4+4’ students did not meet the pre-med course requirements”) and possible academic fraud (“Some of the students’ thesis papers did not meet the required page length, with some only 12 pages long”). The fourth and final section in the AI’s answer discussed the broader public opinion controversy over the “4+4” scandal.

One WeChat essay, published on May 6 and still available online, provides an interesting personal and historical perspective on how socioeconomic privilege has evolved since the beginning of China’s Reform and Opening. In "Deteriorating Circumstances Have Given Rise to ‘Second-Generation Privilege,’” essayist and commentator Xipo (“Western Slope”) explores how slowing economic growth, fiercer competition, and fewer opportunities for social mobility in recent years have spurred those with privilege to resort to ever more extreme measures to pass on that privilege to their children:

After publishing my last article [“All that Remains of the ‘4+4’ Scandal Is the Meme About It”], a friend and I discussed the phenomenon of “second-generation privilege.” That discussion made me realize that the unchecked proliferation of second-generation privilege is actually the result of deteriorating [socioeconomic] circumstances. It took me a while to realize this, but now that I do, it makes a lot of sense.

This friend of mine works at a scientific research institute in southern China. He was at university around the year 2000, a critical juncture in time [for the purpose of our discussion]. I won’t mention his field of study, but let us call him “Professor A.”

Professor A recalls that when he was at university, few of his classmates were what we might call “second-generation scions.” While there were some who excelled at their studies and followed conventional paths mapped out for them by parents, most students pursued majors in different fields from those of their parents.

Back then, of course, the overall population was much less educated than it is today. Many university students had parents who were farmers or factory workers, which is something we should keep in mind.

Professor A observes that when he was a student, even the children of professors and department heads rarely followed in their parents’ footsteps. “In those years, there was an abundance of choice when it came to academic majors and career paths. The children of faculty members chose various majors, regardless of what their parents happened to be teaching.”

But there has been a palpable shift over the past six or seven years, he notes. Now an academic advisor to university students, Professor A has found that most of his colleagues’ children are pursuing the same fields of study as their parents.

Thinking back over the news and public discourse of the past few years, I found that many things instantly clicked into place. That oft-repeated term “involution” [内卷 nèijuǎn, a profound sense of burnout caused by cutthroat academic and socioeconomic competition], suddenly took on a concrete form.

As the old saying goes, “Pavilions situated closest to the water are the first to bask in the moonlight” [近水楼台先得月, jìnshuǐlóutái xiān dé yuè; in other words, proximity has its benefits]. But in order to benefit from structural proximity, there must first be a structure in place. If we examine the history of China’s gaokao [university entrance exam], the most illuminating example can be found in the large cohort of post-Cultural Revolution exam-takers. [This cohort encompassed individuals across an unusually diverse age range, from teens to thirty-somethings whose education had been interrupted by the chaos of the Cultural Revolution.] Back then, teachers and students alike were starting from scratch, and everyone was positioned at the same starting line.

As the educational system gradually returned to normal and became more standardized, a certain group of people (or more accurately, a certain group of families) came to occupy central positions in the hierarchy of academia, scientific research, and resource allocation. This is not to dismiss them entirely, of course, for their contributions were essential as China was starting again from scratch.

During the phase of socioeconomic expansion, this wasn’t too big a problem. Right around the year 2000, for example, socioeconomic conflict was still largely centered in rural areas, and the “three rural issues” (agricultural production, rural development, and rural income) commanded nationwide attention. Although this was only a couple of decades ago, it now feels like a distant memory.

Naturally, by then, some far-sighted types had already begun grooming their second-generation successors. But it was also a time in which emerging industries were booming, culture was vibrant, and society was suffused with ambition and optimism. Even privileged members of the “second generation” didn’t just want to ride their parents’ coattails: they wanted to outdo them, to leave them in the dust.

But as China began to transition from one phase to the next, from expansion to contraction, both the first and second generations came to realize that the most reliable path to success was for children to follow in their parents’ footsteps.

By the latter half of the 2010s, China’s period of breakneck urbanization and industrialization was drawing to a close, and [socioeconomic mobility] had begun to congeal. There were also harbingers that China’s integration into the international economic system had run its course.

Now that we’re as materially well-off as other countries, and are more or less able to compete at the same level, our once-blue oceans of opportunity have become churning red seas of competition.

This is the point at which the first generation advises their children to follow in their footsteps, the better to avail themselves of a wealth of parental first-hand experience and ready-made resources. If the children demur, the parents might say, “Fine, go out and try to make your own way in the world. See how you like competing with a mob of people, all fighting over the same lousy job.” And after taking a quick look around and sizing up the competition, the second generation might think to themselves, “Sure, I’ll take your advice. Work is work. What more could I want?”

Uncertainty about the future is spurring those who already occupy lofty positions to marshal all available resources to pass their competitive advantages on to the next generation. This type of survival strategy does not differ fundamentally from that seen in the animal and plant kingdoms.

Naturally, these sorts of collective choices can have extremely negative consequences. Amid deteriorating circumstances, second-generation scions may happily “settle” for enjoying their second-generation privilege, but today’s bona-fide “first generation,” those with no parental legacy to lean on, suffer a dual blow. There are fewer opportunities available to them, and increasingly unfair competition for the few opportunities that do remain.

With this in mind, I have even more empathy for young people today. The ones wailing in frustration are those who bear the brunt of this “dual blow.”

Yet I would still advise them not to conflate their own career development with critical analysis of the socioeconomic environment. As I’ve said before, we can’t wait for society to improve before we start living our own lives. Even in conditions of unfair competition, we must take the initiative and find our own ladder to success. But I now have a deeper understanding of the dejection that so many are feeling right now.

And to those privileged first and second generations, I would like to say: “Other people still exist, even if you don’t see them. Other voices still exist, even if you don’t hear them. They are not simply your competitors; they are emblematic of shared opportunity and a path forward for everyone.”

Although humans are part of the animal kingdom, too, we should be able to do better than simply adhere to the doctrine of “survival of the fittest.” Even beavers know how to shape the environment to their advantage by building dams. Human beings, especially those who consider themselves “elites,” must learn to take responsibility for the environment they shape and inhabit.

After all, someone needs to think about the long-term prospects and overall health of our society. [Chinese]

It is not just drones that are promising – or threatening – to upend the tempo of urban life in China. Humanoid robots are particularly buzzy. The highlight of this year’s Spring festival gala, which was viewed nearly 17bn times, was a dance performed by a troupe of humanoid robots made by a company called Unitree. On Saturday, the world’s first humanoid v human race – a half marathon – took place on the outskirts of Beijing.

“Applying artificial intelligence to robots basically really kicked into high gear last year,” says Rui Ma, a China technology analyst and investor based in San Francisco. The shift could allow the industry to grow at a much faster rate in 2025 than in previous years. Reinforcement learning, which means training robots to learn from experience rather than relying on rigid models, allows humanoid robots to be trained in months rather than years, hastening the pace of innovation. Toy robot dogs are already part of daily life in China. At a wholesale market in Yiwu, a trading hub in east China’s Zhejiang province, a child plays with a robot dog while his mother haggles with exporters over the price of false eyelashes. On the streets of Shanghai, a woman walks her robot dog, which is helpfully carrying a basket of shopping on its back.

The development of China’s robotics industry is intimately linked to advances in AI. For years, China has been trying to catch up with the United States. Xi wants to drive economic growth through “new quality productive forces”, a concept that includes advanced technologies. [Source]

Some scientists and observers have tried to manage the public’s expectations of China’s AI rollout. Zeyi Yang at WIRED described the setbacks of the “stumbling and overheating” humanoid robots competing in the half marathon, and also noted that “by the end of the race, many people who tuned into the livestream started to comment on how exhausted the robots’ human operators looked.” At a meeting last Wednesday, the Chinese Ministry of Industry and Information Technology ordered representatives from dozens of automakers to stop hyping up the capabilities of smart driving technology. (This followed a highly publicized incident last month, in which a Xiaomi electric car crashed and killed three passengers while the vehicle’s autopilot feature was engaged.) Similarly, leading AI scientist Zhu Songchun recently warned that AI hype and reality have become detached in China, partially because the media have fed the public “exaggerated” stories about AI. Alex Colville at China Media Project described how Zhu’s critical stance appears to have been ignored, if not overridden:

Zhu’s critique of the propaganda-driven approach appears to have fallen victim to precisely the dynamic of hype he described. While his remarks found outlets in more market-oriented publications like Tencent Technology, Caixin and The Paper, flagship state media organizations like Xinhua and the People’s Daily conspicuously omitted his warnings from their coverage. Instead, these Party organs continued to showcase a parade of applications and robots — the very surface-level achievements that Zhu suggested are distracting China from the deeper scientific work needed to truly lead in artificial intelligence. In a system where positive messaging trumps critical analysis, even warnings from one of the nation’s top AI scientists can be edited out of the narrative.

[...] This disconnect was illustrated once again over the weekend, as Beijing hosted a half marathon where Chinese-built robots raced alongside human competitors. The CCP’s official People’s Daily described the event as a “fierce competition” that had pushed the robots to their limits. Xinhua sang about “infinite possibilities,” and proclaimed in its headline that the racing event had “closed the distance between us and the future.” The less stellar reality, alluded to in a report by Guangzhou’s Southern Metropolis Daily that noted the “many problems” holding the race down, was that the robots had suffered constant failures and necessitated nearly constant repairs by the exhausted human crews running alongside them. In the end, only six of the 21 robot entries completed the race, and one quite literally lost its head.

But in another sense, the race pointed the way toward the possibility of a healthier, more open and more self-critical attitude toward technology and progress — an alternative to the propaganda of constant rise. The Global Times, though in English-language coverage only, remarked somewhat disingenuously that “[behind] this ‘imperfect’ robot half-marathon is the mature atmosphere of tolerance, understanding and acceptance of failure that has developed in Chinese society from top to bottom toward the high-tech industry.” If that were true, of course, no public moderation of Zhu Songchun’s remarks behind closed doors would have been necessary. It would be perfectly acceptable to say: We are getting this wrong. But the Global Times was on to something. [Source]

China’s rapid deployment of embodied AI is at least partially fueled by its competition with the U.S. for technological superiority. Liu Gang, a professor at Nankai University in Tianjin, stated, “We are picking a path where we lower the costs for innovation and industrialization,” adding, “When many can do things with a comparable quality, whoever makes it more cheaply will have a bigger chance to win.” This is evident in the significantly lower costs of Chinese robot dogs and humanoid robots compared to American ones, and in the free rollout of DeepSeek’s AI chatbot. A recent Foreign Affairs article titled “What America Gets Wrong About the AI Race” underlined this dynamic: “The real lesson of DeepSeek’s success is that AI competition is not simply about which country develops the most advanced models but also about which can adopt them faster across its economy and government.”

Indeed, China’s robotics industry is surging ahead. According to a recent industry report, China will likely produce over 10,000 humanoid robots this year, amounting to over half of global production. Late last month, Du Zhihang, Bao Hongyun, Liu Peilin, and Han Wei from Caixin Global provided a deep dive into China’s production of humanoid robots, underscoring the government’s major investments in this booming sector:

Embodied intelligence has become a buzzword in AI and investment circles in the past two years. Unlike traditional industrial robots, such as mechanical arms, embodied intelligence focuses on humanoid robots that can adaptively perceive and interact with their environments using human-like physical forms. These robots emphasize advanced motor coordination — akin to a cerebellum — and cognitive abilities in vision, language and movement, enabling more natural and versatile interactions. This shift represents a move beyond factory automation toward robots capable of more complex, human tasks.

[...] Since 2024, several Chinese regions have rolled out incentive policies to support AI and robotics industries as the country aims to lead the global tech race. Hangzhou in the eastern Zhejiang province offers up to 5 million yuan in rewards and 25% project funding subsidies, while Beijing has established a 100-billion-yuan government investment fund to support startups. Southern Guangdong province provides up to 50 million yuan for robotics companies and 10 million yuan for AI firms, and Shenzhen has set bold targets for embodied intelligence, aiming to cultivate more than 10 companies valued at more than 10 billion yuan and achieve an industry scale exceeding 100 billion yuan by 2027. Shanghai and the southwestern Sichuan province have also introduced supportive measures.

[...] While China and the United States are seen as equals in AI and robotics technology, China’s strengths in large-scale manufacturing and software optimization give it an edge in industrializing humanoid robots.

[...] As China’s governments increase support for AI and embodied intelligence, leading companies stand to benefit. In March 2025, embodied intelligence was named one of four future industries — alongside biomanufacturing, quantum technology and 6G — in the government work report. [Source]

"Return to Wuhan: The Unfinished Story From Five Years Ago,” a now-deleted longform article from WeChat public account “Aquarius Era” (水瓶纪元, shuǐpíng jìyuán) includes interviews with journalists, doctors, artists, activists, and others about the early days of the COVID pandemic and the lockdown of Wuhan. (Although the article has been censored on WeChat, it remains available through the Substack account @aquariuseras.) The article also chronicles more recent attempts to combat official “amnesia” with online and offline commemorations of the whistleblowers, citizen journalists, and victims of the pandemic. The translated excerpt below describes some of the restrictions on remembrances of Dr. Li Wenliang, and the daunting challenges of keeping his memory alive:

Li Wenliang was buried in the Wuhan’s Jiufengshan Cemetery of Revolutionary Martyrs, along with Peng Yinhua, Liu Fan, and others who were recognized as “martyrs” because they sacrificed their lives in the fight against COVID-19. Many local residents who went there to mourn discovered that to enter area two, the section of the cemetery where Dr. Li is buried, they were required to hand in their mobile phones; sign a register with their names, household registration, health codes, and other information; and be escorted by cemetery staff to visit Dr. Li’s grave. Sometimes staff members would say kindly, “If everyone were as cooperative as you, our jobs wouldn’t be so hard.” Other times, staffers would scold coldly, “Make sure to write your ID number clearly, because we have to check it!” Five years on, we no longer have the right to freely visit someone’s grave.

[...As far back as] June 2020, not long after the pandemic in Wuhan had abated, the artist known as "Brother Nut" found that his personal Shimo account [Shimo, which translates as “Graphite,” is a Chinese cloud-storage and document-sharing service similar to Google Docs] was blocked because he included the name "Li Wenliang" in a document title, and he was unable to export any of content from the stored document. When he attempted to defend his rights by filing a complaint, the response from platform customer service was: "You published a document advocating large-scale collective rights protection," and "Your content is politically sensitive."

“The impact of the pandemic period seems much like the virus itself: invisible, intangible, and traceless," said [journalist] Wang Shengnan. [Chinese]

Another longform article noting the fifth anniversary of the Wuhan lockdown was published in early March by RFA-affiliated media outlet Wainao. The now-defunct Chinese-language organization, also known as WHYNOT, was forced to close in mid-March due to steep U.S. funding cuts to the U.S. Agency for Global Media (USAGM). Titled "Five Years After the Wuhan Lockdown: The Traces of Pain That Remain, and the Ordinary People Resisting ‘Amnesia’ in Their Daily Lives," the article explores the phenomenon of collective amnesia while noting some exceptions—such as the many commemorations still being posted to “Li Wenliang’s Wailing Wall,” the popular comment section under Dr. Li’s final Weibo post.

In Wuhan, to this day, "76" remains a number of special significance. From the beginning of the lockdown on January 23, 2020, until its lifting on April 8, Wuhan residents experienced exactly 76 days of being confined at home and almost completely deprived of their freedom.

Despite all the difficulties that Wuhan residents endured, nowhere in the city is there any acknowledgement of what the people of Wuhan went through during that period—no memorial, no exhibition hall, no genuine commemoration of their suffering.

[...] Wuhan Central Hospital, where Dr. Li Wenliang once worked, was one of the hospitals hardest hit by COVID infections among doctors, nurses, and other medical personnel.

The hospital is located on Nanjing Road, opposite the historic cultural district Xian’an Fang, famed for its narrow alleyways and red-brick exteriors. Nowadays, apart from some chain restaurants, the area is mainly home to an array of distinctive boutiques. Memories of the pandemic occasionally resurface here.

[...] In 2021, as reported in the media, a nearby café menu once featured a coffee item named "Whistleblower Coffee—100% Controversial.” Today, this café no longer exists. In a stylishly decorated bar [near Wuhan Central Hospital], there is a feminist-themed book display where customers can leave books and post book recommendations. The bookshelf display contains many comments about death and even a book about ophthalmology, but there is no trace of Dr. Li Wenliang. It seems that there is a tacit understanding not to publicly mention Dr. Li.

Dr. Li Wenliang, who worked as an ophthalmologist at Wuhan Central Hospital, later became known as “the whistleblower of the COVID-19 pandemic.” As one of the first to sound the alarm about the emerging coronavirus—in an online alumni group chat—he was admonished by authorities and labeled a "rumormonger." In the early hours of the morning on February 7, 2020, Wuhan Central Hospital announced that 34-year-old Li Wenliang had died of COVID-19, setting off shock waves on the internet.

At Exit F of Wuhan’s Xunlimen subway station, only an eight-minute bike ride from the hospital where Li Wenliang once worked, there is a large shopping mall featuring shops similar to those found in many other Chinese cities: bubble tea shops, beauty salons, and a food court. Across from the mall, there is a small kiosk that specializes in replacing phone screen protectors. "Who is Li Wenliang?” asks the kiosk’s owner [in response to our question], as he wipes a mobile phone screen with an alcohol swab. “I don’t know him!"

Upon further questioning, it turns out that the kiosk owner is not a local, but had come from Hunan to work in Wuhan three years ago. Most Wuhan locals have heard of Li Wenliang. Nowadays, when his name is mentioned, they are likely to respond: "Oh, that doctor who died."

At the hospital where he once worked, the name of the late Dr. Li Wenliang is absent from a wall displaying the names and photos of hospital specialists. Staffers at the hospital information desk answered our questions cautiously: "He used to work here, but we didn’t know him, and don’t know much about it." The owner of a newsstand at the hospital entrance said, "He died, and the pandemic is over. I don’t know much about it. It’s not something we should be blabbing about. Go look it up online if you want to know more.”

In February 2025, a patient walks past the “specialist wall” inside Wuhan Central Hospital. (source: Wainao/photographer Zu Weina)

On the internet, DeepSeek AI—which Chinese people proudly herald as a rival to ChatGPT—is incapable of answering the question "Who is Li Wenliang?" Instead, DeepSeek offers this line of text: "Hello, I am unable to answer this question at the moment. How about we change the topic and chat about something else?"

But there are other places in which Li Wenliang has not been forgotten. On Sina Weibo, Dr. Li’s last public post, dated February 1, 2020, remains frozen in time: “Today, [my] nucleic acid test results came back positive. The dust has settled, there is finally a diagnosis.”

Ever since, there have been daily updates to the comments under that post, with over one million comments posted. [That number is likely higher, but the counter under the post is capped at “one million plus.”] On Valentine’s Day 2025, one commenter confided chattily to Dr. Li: "We split up right around the Lantern Festival and Valentine’s Day. I know it’s the right decision, but who’s ever happy about breaking up, right? And I’m going to take my driving test soon. Hope I pass on the first try." Another expressed their longing thus: "Doc Li, the flowers in Beijing are about to bloom."

While in Wuhan, there is nary a tribute to Dr. Li, countless Chinese people continue to remember him fondly. [Chinese]

The comments section under Li Wenliang’s final Weibo post, mentioned in the article above, has become known as China’s “Wailing Wall,” a place netizens come to mourn and to celebrate, to mark personal milestones or comment on current events, and to wish Dr. Li well and assure him that his sacrifice will not be forgotten. (CDT editors continue to regularly archive and publish updates on recent Wailing Wall content.) The most recent update includes comments left during the April 4 Qingming festival, also known as “Tomb-Sweeping Day,” when many Chinese remember or visit the gravesites of deceased family members. The following Wailing Wall comments were compiled between April 1-April 5:

四灵妖王: Doctor Li, we haven’t forgotten you.

溜溜溜只洋芋: The cherry blossoms are in bloom, Dr. Li.

不想熬夜的夜猫子666: It’s another Qingming Festival, and I miss you. 🕯️ I hope you are doing well in the other world. 🕯️

momomokoo: Dr. Li, I’m afraid of blind dates, and even more afraid of being rejected.

GEVEYteam: "At Qingming, in the drizzling rain / the bereft wander through the lanes." It rained today. [The quoted lines are from the well-known poem “Qingming” by Tang Dynasty poet Du Mu.]

想翻身的孩子: Dr. Li, nothing has changed here, or if anything, it’s worse. I hope you are safe and well, and that things are better now.

不被团结的nevermore: Good evening, Dr. Li. You’ve had a hard day, too, so turn in early and get some rest.

积极创想曲: When I’m feeling lost, I like to visit this Weibo’s comment section and read about people from all walks of life.

我想我喜欢你1997: Dr. Li, I’m confused about my future, what should I do?

晓阳205011: Hi Dr. Li, I’m back. The unbearable month of March is finally over. I hope you’ll bless me with a bit of better luck in April. Because life’s been rough, too rough.

潮汐夕阳杨桃-: I’ve been daydreaming about a lot of things, but it’s these daydreams that keep me going. [Chinese]

CDT’s Wailing Wall archive is compiled by Tony Hu.

According to officials in Beijing’s Changping district, the City Management Command Center has used Deepseek to break down “information barriers in grid management,” accurately meeting complex needs like “cross-domain collaboration.” In addition, the local government has integrated DeepSeek’s deep analysis technology with HD video from the Sharp Eyes project (a rural grid management-based “mass public security prevention and control project” using networked video surveillance), establishing an “All-Weather Urban Awareness Network,” which has “eliminated the need for time-consuming, labor-intensive patrols.”

The Public Security Bureau in Inner Mongolia’s Uxin Banner said that DeepSeek has improved the precision of safety and security work for major events and proven effective at preventing and controlling potential security risks. They explained that DeepSeek’s real-time analysis of factors such as participant information and event-site conditions can promptly detect unusual activity and provide early warning, “allowing for flawless safety and security work.”

The Public Security Bureau in Chongqing’s Rongchang district said that since DeepSeek went online, analytical law-enforcement tasks that would previously have taken three officers three days can now be accomplished by a single officer in 15 minutes, with “outstanding results in the field.”

A community propaganda worker in Shenzhen’s Bao’an district said that DeepSeek allows them to handle “public opinion” more “skillfully,” and has greatly boosted their productivity by helping them to quickly grasp “key public-opinion topics” and perform “quantitative analysis of public-opinion trends.” The head of the district’s neighborhood law-enforcement team also indicated that DeepSeek enabled them to more accurately pinpoint the sources influencing public opinion, anticipate risks, and provide strong support for their work responding to public opinion.

Cyberspace Administration authorities for Inner Mongolia’s Hinggan and Xilingol Leagues noted DeepSeek’s marked advantages in recognizing contextual complexities and identifying potential hazards, and said they would continue to push ahead with the use of AI in areas such as content management, “public opinion analysis,” and cybersecurity.

Shandong Province Internet Media Group, a province-level Party media organization, offered more details on DeepSeek’s application in "public opinion monitoring." The group indicated that since it adopted DeepSeek, it has seen great advances in both its efficiency at parsing "public opinion data" from across the internet and its ability to filter out noise, and that it can now more quickly identify potential hazards when monitoring hot topics. In addition, DeepSeek can automatically generate "strategic recommendations for public opinion response" based on its analysis of vast quantities of data, and provide smarter "suggestions for handling public opinion."

Topsec, an internet security monitoring company that works closely with the authorities, said that it is already using DeepSeek’s deep content recognition technology—together with other methods such as keyword scanning and optical character recognition—to comprehensively monitor sensitive information and promptly block activity that violates regulations.

Many local propaganda departments and Party media outlets have said that DeepSeek can automatically generate press releases based on real-time information, helping state-media journalists to quickly compose news reports.

The municipal government of Xinxiang, Henan, released a guide to a Smart Official Document Composition Assistant powered by DeepSeek. This reportedly includes a corpus of "official document templates from Party and state administrative units," and can automatically cross-reference with the latest official lists of words to be avoided or treated with caution in public materials. It can also automatically screen for core political terminology like the "Two Safeguards" and "Two Establishes" and compare it against central [government] documents.

The Party committee for Altay in Xinjiang said that DeepSeek enables thoroughly "smart" Party-building work. If you want to study central policy documents and really grasp their spirit, you can simply upload them and DeepSeek will generate a summary of the key points, highlighting and explaining specialist jargon and implementation challenges (e.g. “grassroots Party organization election procedures”), thus helping avoid misinterpretation by grassroots cadres. [Chinese]

The original post at Diyin includes links to source materials, some of which have since gone offline.

One now deleted report at Nanfang Daily, archived at CDT Chinese, focused on DeepSeek’s use in online public opinion monitoring in a Shenzhen subdistrict, noting public suspicions—denied by subdistrict officials—that AI systems may be used to delete or respond to public posts by local residents:

On February 20, the Xixiang subdistrict of Shenzhen’s Bao’an district held a training session on “DeepSeek + Public Opinion Response” for 50 leaders from departments and communities within its jurisdiction. Its goal was to draw on AI technology to enhance the subdistrict’s overall public opinion monitoring, analysis, and response capabilities, and enable it to more rapidly assess and respond to the concerns and requests of city residents.

While the news received “Likes” from many netizens, others had doubts: were government departments using AI in negative ways, such as responding to or deleting online posts? Xiao Bin, propaganda chief for Xixiang’s Office of Party, Government, and Legislative Affairs, responded that AI is not being used for content deletion and other such tasks, only for rapidly sifting through and organizing vast quantities of data to ensure that every resident’s concerns and requests can be handled and answered promptly.

Grassroots Cadres: More Professional Confidence, Less Stress

At the start of the training, the instructor emphasized the necessity of learning and using AI: “You won’t be replaced by AI, but by people who know how to use it.”

They explained the main points of DeepSeek’s technical features, applicability, and operation using a threefold framework of feature analysis, application scenarios, and practical demonstration. In the public opinion response segment of the training, the instructor gave a comprehensive introduction to workflow and methodology for online public opinion response, from monitoring and analysis to response and resolution.

The instructor stressed that in the digital age, public opinion can spread faster and further, so each subdistrict department and community needs to heighten its vigilance and, putting the people first, promptly respond to the masses’ concerns. At the same time, malicious rumors and incitement must be firmly dealt with in accordance with the law. The session also included practical discussion and simulated drills.

A propaganda committee member from Xixiang’s Liutang community commented, “In the past, I’ve often felt flustered when dealing with public opinion. We never had all the information, and responses were haphazard. Now that I understand DeepSeek’s formidable capabilities, like quickly picking the crucial points of public opinion from the flood of online data and performing quantitative analysis of public opinion trends, I can be more confident in follow-up response work, and less stressed.”

An official from Xixiang described the training as an important effort to innovate the subdistrict’s grassroots governance methods, and said that by adopting advanced AI technologies, they would be able to more effectively capture public opinion data and create a positive public opinion environment amenable to the subdistrict’s harmonious and stable development.

Subdistrict Clarifies: No AI Post Deletions

Xixiang’s “DeepSeek + Public Opinion Response” training drew widespread public attention. While much of this was positive, there was also suspicion among some netizens.

Therefore, I decided to speak with Xiao Bin, the propaganda chief for Xixiang’s Office of Party, Government, and Legislative Affairs. Xiao Bin explained that the subdistrict has always attached great importance to public opinion work. Using platforms such as “Quick Response to Public Opinion,” “Mediation and Dispute Resolution,” “Bao’an District’s ‘Scan for Instant Resolution of Unpaid Wages,’” and “Bao! You’re Hired!” to quickly gather information on the public’s various complaints and comments so that “when the people call, we answer.”

As a major Shenzhen subdistrict with over a million inhabitants, Xixiang processed, verified, handled, and responded to 1,050 public opinion cases of one kind or another in 2024. Issues like market supervision, urban management, transportation, and the environment featured heavily. Handling these cases brought swift resolution and response to a great many of the problems reported by city residents.

Xiao Bin refuted claims that government departments use AI for negative activities like post deletions. She noted that a real-name account is required for replying to posts, and registering pseudonymous accounts for automatic AI-generated responses would be not only technically difficult, but also probably illegal. She said the subdistrict’s public-opinion workers strive to be objective, fair, and responsible, taking every public-opinion case seriously and engaging in active communication with the public for the sole purpose of conscientiously resolving their problems.

“The significance of AI in public opinion work,” Xiao Bin said, “is in helping us rapidly filter and compile masses of data so that relevant departments in the subdistrict can accurately assign resources to quickly verify and deal with various issues, and ensure that every resident’s concerns and requests receive prompt resolution and response.”

Asked about plans for future work, Xiao Bin said that they would continue to optimize public opinion work mechanisms, treating the public’s online views and suggestions as “opportunities to deliver government services right to people’s doors,” and taking full advantage of AI to further improve the efficiency of public-opinion handling and provide the public with even better and more effective service, promoting harmonious and stable societal development. [Chinese]

At the Journal of Democracy last month, Valentin Weber placed DeepSeek in the context of the PRC’s steadily evolving surveillance ecosystem, writing that it “has massive potential to enhance China’s already pervasive surveillance state, and it will bring the Chinese Communist Party (CCP) closer than ever to its goal of possessing an automated, autonomous, and scientific tool for repressing its people.” The shortcomings of AI outputs, such as “hallucination” of false positives, only add to the potential risks of such a system.

There is some degree of performativity in the surge of very vocal DeepSeek adoption. The New York Times’ Meaghan Tobin and Claire Fu noted that “The enthusiastic embrace of the technology by China’s bureaucracy reflects, in part, what often happens when Mr. Xi, China’s most dominant leader in decades, puts his stamp of approval on something. (Mr. Xi has set off frenzies over soccer, winter sports and high-end manufacturing, for instance.) […] But it can be hard to parse the substance from the hype. While scores of officials have pledged to use DeepSeek in their work, few have described specific examples in which the technology has made that work more effective or efficient.”

Similarly, Wired’s Zeyi Yang wrote that while some companies “have found genuine uses for the domestic, affordable AI model with cutting-edge capabilities, […] others are merely doing it for the publicity boost or to virtue-signal their national pride.” (Yang noted that this AI gold rush does not have uniquely Chinese characteristics: “The whole frenzy resembles what happened in late 2022 when ChatGPT launched and a wave of American and European companies scrambled to find ways to signal to customers and investors they were engaging with what was then the most cutting-edge innovation in AI.”)

But there’s also another factor that has helped make DeepSeek particularly trendy in China: the fact that the West freaked out about it. “Its strong reception overseas has further boosted its popularity in China, serving as the firm’s best marketing campaign,” says Angela Huyue Zhang, a law professor who studies Chinese technology policy at the University of Southern California.

The narrative that DeepSeek is challenging US dominance in AI has contributed to a growing sense of national pride within China. A central part of the company’s heroic origin story is its development of resource-efficient models, which was seen as a direct response to US policies designed to cut off China’s access to cutting-edge semiconductors. As a result, DeepSeek’s success has fueled a growing belief in China that those measures may eventually fail.

“Where there is blockade, there is breakthrough; where there is suppression, there is innovation,” Wang Yi, China’s minister of foreign affairs, said in a speech on March 7 in which he also compared DeepSeek to China’s previous technological breakthroughs in areas like nuclear weapons development and space exploration. [Source]

The performative aspects of the DeepSeek wave, and the arguably overhyped technical leap that DeepSeek represents, do not necessarily diminish its significance. The Economist argued last week that “The true winner of the AI race […] may not be the country that invents the best models. It is more likely to be the country where governments, businesses and ordinary people use AI at scale every day. For everything from economic growth to military power, technological diffusion ultimately matters more than technological innovation. On that front, the race is closer than many in America believe.”

First uncovered by Marc Hofer of the NetAskari newsletter, the data is essentially a reservoir of articles that require labeling, each article in the dataset containing a repeated instruction to prompt the LLM in its work: “As a meticulous and serious data annotator for public sentiment management, you must fully analyse article content and determine the category in which it belongs,” the prompt reads. “The ultimate goal is to filter the information for use in public opinion monitoring services.”

[…] First, it reveals a sophisticated classification system with 38 distinct categories, running from more mundane topics like “culture” and “sports” to more politically sensitive ones. Tellingly, the three categories marked as “highest priority” in the dataset align distinctly with state interests as opposed to commercial ones. Topping the list is “information related to the military field,” followed by “social developments” (社会动态) and “current affairs developments” (时政动态). This prioritization underscores how private tech companies like Baidu — though it could not be confirmed as the source of this dataset — are being enlisted in the Party-state’s comprehensive effort to monitor and shape online discourse.

[…] The exact purpose of this dataset remains unclear. Were these classifications developed internally by Baidu — or were they mandated by state regulators? Nevertheless, the unsecured data offers a glimpse into the inner workings of China’s AI content dragnet. What was once a labor-intensive system requiring thousands of human censors is rapidly evolving, thanks to the possibilities of AI, into an automated surveillance machine capable of processing and categorizing massive volumes of online content. [Source]

On Wednesday, Charles Rollet at TechCrunch shared experts’ views on how such AI-powered content categorization for the purpose of “public opinion work” would ultimately enhance the state’s censorship abilities:

[CDT’s] Xiao Qiang, a researcher at UC Berkeley who studies Chinese censorship and who also examined the dataset, told TechCrunch that it was “clear evidence” that the Chinese government or its affiliates want to use LLMs to improve repression.

“Unlike traditional censorship mechanisms, which rely on human labor for keyword-based filtering and manual review, an LLM trained on such instructions would significantly improve the efficiency and granularity of state-led information control,” Qiang told TechCrunch.

[…] Michael Caster, the Asia program manager of rights organization Article 19, explained that “public opinion work” is overseen by a powerful Chinese government regulator, the Cyberspace Administration of China (CAC), and typically refers to censorship and propaganda efforts.

The end goal is ensuring Chinese government narratives are protected online, while any alternative views are purged. Chinese president Xi Jinping has himself described the internet as the “frontline” of the CCP’s “public opinion work.” [Source]

Recent developments reflect ongoing attempts to instrumentalize AI for online surveillance and censorship. In February, OpenAI reported that a Chinese-origin network used ChatGPT to build a tool for collecting social media activity on sensitive political and social topics and marketing it to Chinese authorities. Another data leak discovered last month by Sentinel Labs revealed additional types of collaboration between public and private actors in China for the purpose of monitoring and censoring content on the Chinese internet. (On a related note, a whistleblower alleged that in its failed attempt to access the Chinese market, Facebook “developed a censorship system for China in 2015 and planned to install a ‘chief editor’ who would decide what content to remove and could shut down the entire site during times of ‘social unrest,’” as The Washington Post reported earlier this month.)

Beyond surveillance and censorship, AI has been employed across a wide swath of Chinese society. Its ubiquity has led the Cyberspace Administration of China to mandate that online platforms accurately label any content that is generated by A.I. China Media Project’s China Chatbot column has also documented the intersection of Chinese media and AI. Over the past few weeks, dozens of Chinese automakers, medical and pharmaceutical companies, banks, insurance companies, brokerage firms, and even local governments have announced that they are using DeepSeek AI in their products, research, and training to enhance a variety of tasks. Amber Wang at the South China Morning Post reported on Sunday that even China’s PLA is using DeepSeek AI for non-combat support, and potentially other military tasks:

DeepSeek’s open-source large language models (LLMs), which have drawn global attention and praise, are being used in PLA hospitals, People’s Armed Police (PAP), and national defence mobilisation organs, according to publicly available information.

[…] Some units of PAP – a paramilitary police force under the command of the Central Military Commission, which also directs the PLA – are using the app for daily physical training and psychological counselling.

[…] The PLA has called for the incorporation of high-end technology, particularly AI, to strengthen its combat capabilities. This would include boosting the effectiveness of drone swarm tactics, improving the efficiency and realism of pilot training, and battlefield decision-making support.

[…] Fu Qianshao, a Chinese military analyst, said DeepSeek’s applications in routine physical training and logistical support “demonstrate the PLA’s commitment to ‘staying up-to-date and fully utilising AI technology to enhance comprehensive combat capabilities’.”

“It cannot be ruled out that DeepSeek has been used for other combat functions,” Fu said, adding that “the integration of AI into command systems has been under way for a considerable time”. [Source]

The investigation was launched earlier this month, and Belgian authorities last week arrested four people on charges of corruption and being a member of a criminal organisation. A fifth person was charged with money laundering but released after questioning.

Among the arrested suspects is a lobbyist for Huawei who used to work as an assistant in the European parliament. He is suspected of orchestrating the payment of bribes to parliament workers — notably to secure support for a letter several MEPs signed defending the Chinese company’s interests.

[…] The letter, signed by eight EU lawmakers and sent to the European Commission in January 2021, warns about the “politicisation of the deployment of 5G technology” and criticises the ban of foreign 5G devices based out of an “unsubstantiated fear of national security risks”.

[…] Two people familiar with the investigation said the alleged bribes also included offers of Huawei smartphones and tickets to see a match of the local Anderlecht football team. [Source]

According to the investigative judge in charge of the Belgian probe, “A sum of €15,000 [over $16,000 U.S. dollars] was offered to the writer of the 5G letter, while each co-signatory was offered €1,500,” and “suspicious payments” worth tens of thousands of euros were allegedly arranged in exchange for the related services. Huawei said it takes the allegations “seriously” and has a “zero tolerance policy towards corruption or other wrongdoing.” After the corruption case was announced, E.U. institutions expelled Huawei lobbyists from their premises, but some lawmakers expressed concern about loopholes that the lobbyists could use to circumvent the ban, including using a visitor pass or lobbying through umbrella trade associations that continue to represent Huawei. Last week, Politico described Huawei’s extensive record of lobbying within the E.U.:

The company threw lavish parties in glamorous venues featuring fancy buffets and dance performances — like its reception celebrating the Chinese new year at the Concert Noble in Brussels — and was known for thanking contacts with generous gift bags, some including a Huawei phone.

[…] To navigate the geopolitical storm [over security and espionage concerns that spiked after 2019], the firm offered six-figure salaries to former Western journalists and politicians with direct lines to places of power like the Elysée and Westminster.

[…] According to EU transparency register data, Huawei Technologies spent between €2 million and €2.25 million on EU lobbying in 2021, 2022 and 2023 — a lot, but still below its lobbying costs in preceding years, which were estimated at around €3 million in 2018, 2019 and 2020.

Huawei in October declared it had 11 full-time EU lobbyists, nine of whom were accredited to access the European Parliament. At its peak it declared 21. [Source]

On Monday, Alexander Fanta and Simon van Dorpe at Dutch investigative news outlet Follow the Money published an article about the scandal, underlining the alleged, long-running role of top Huawei executives:

The bribery allegations, including the amounts allegedly paid to the signatories, correspond to a complaint that the NGO Transparency International EU received in 2022 and forwarded to the EU anti-fraud office OLAF. However, OLAF decided not to open an investigation, stating there was “no sufficient suspicion”.

[…] According to the arrest warrant, the money transfers intended to disguise the bribes “would have been endorsed by the Chinese executives of Huawei, in particular Abraham Liu”, who at the time was the company’s vice-president for the European region and its chief EU representative. The warrant does not state why investigators believe Chinese executives were involved.

[Valerio] Ottati, Huawei’s Belgian-Italian head of public affairs [who is accused of orchestrating the bribery operation], allegedly told a Polish Huawei employee that Brussels lobbyists “often cross the line and even pay for amendments”, according to a wiretap in his Volkswagen Tiguan, referred to in the arrest warrant.

Former Huawei staff members in Brussels told Follow the Money, Knack and Le Soir that Ottati’s problematic behaviour was widely known within the company, but tolerated by higher-ups. Several employees said they had raised concerns about Ottati, including warnings about possible corruption, but that these were never seriously followed up on. [Source]

The current Huawei corruption scandal evokes the 2022 Qatargate corruption scandal, in which European authorities seized 1.5 million euros in cash and arrested four MEPs on corruption charges related to illicit influence by Qatar, Morocco, and Mauritania. (The full investigation is still ongoing.) Politico noted that one of the offices sealed in the Huawei probe belonged to Adam Mouchtar, co-founder of a group that had as its president Eva Kaili, who was arrested in the Qatargate probe. In reaction to the Huawei probe, Nicholas Aiossa, Director at Transparency International EU, said “These new allegations are as sweeping and serious as Qatargate and make a mockery of democracy at the European Parliament. For too long, MEPs have taken a carefree approach to ethics and continue to exist in a culture of impunity. If MEPs want to protect the integrity of the Parliament, they need to bring about swift, wide-ranging, and substantial ethics reform.”

Against the backdrop of the Huawei scandal is a shifting and at times contradictory E.U. policy landscape vis-a-vis China. Last week, The Financial Times reported that the European Commission had begun an investigation into whether China provided unfair subsidies for a BYD electric car plant in Hungary, underscoring the E.U.’s concern with Chinese economic and security threats. But recent data also shows that 17 E.U. member states have not fully implemented the Commission’s 5G cybersecurity toolbox from 2020 that mandates a ban on Huawei and ZTE from their networks. The nationalist tabloid Global Times pounced on these divergences in three recent editorials, arguing that “Europe should dismantle the barriers of suspicion” when it comes to Chinese telecom suppliers and make “a rational choice to turn further toward China.” However, the Global Times has so far made no mention of the Huawei corruption probe. Acknowledging the growing fractures in the transatlantic relationship, a recent article in The Economist suggested that “Europe will have to zip its lips over China’s abuses” in order to safeguard its broader trade goals.

• Between 2015 and 2022, companies in multiple business sectors in China posted more than 1.7 million censorship-related job ads, varying by different levels of censorship duties. The popularity of social media, especially video-based platforms, has led to dramatic increases in the demand for censorship labor.

• In the Chinese censorship market, there are four major players: traditional content-based companies, such as news and social media companies; nontraditional censorship-requiring companies, who incorporate censorship into the responsibilities of other conventional roles (for example: a marketing professional required to censor promotional content, in addition to their primary tasks); human resource companies that offer outsourced censorship labor; and state-owned media agencies that offer their own suite of censorship services.

• Censorship tasks were considered part-time duties at the start of the researched time period, but over the seven years quickly evolved into full-time professional work. Though essential to companies, censorship work is considered low-skill labor in China, and is characteristically labor-intensive and underpaid.

• To lower the cost of operating a large censorship team, some companies choose to outsource their censorship work to human resource companies or their own subsidiary companies. The research uncovered over 3,000 human resource companies engaged in the outsourced censorship labor market.

• The increasing demand for outsourced censorship labor has resulted in a prominent trend of geographical redistribution of censorship workforces, from more developed coastal areas, to developing inland regions. [Source]

In addition to OTF’s macro-level analysis, CDT and others have published more micro-level insights into China’s human censors. Last October, CDT translated excerpts from an interview by independent Chinese-language magazine Mangmang with a Gen Z censor, who described forbidden words, salary and working conditions, and those he viewed as the real culprits of censorship in China. In 2023, Teacher Li interviewed a former censor on his YouTube channel, and Global Voices translated excerpts of the interview. In 2022, CDT translated part of a report from Chinese digital media outlet Late Post that provided an in-depth analysis of the grueling work culture of Bilibli’s content moderation department, after a Bilibili content moderator died from a brain hemorrhage due to what many suspected was overwork. RFA has also profiled Eric Liu, a former analyst at CDT Chinese who was previously a content moderator for Weibo.

More recently, in January, RedNote (Xiaohongshu) posted multiple recruitment ads for content moderators in order to help adjust to the massive influx of American “TikTok refugees.” CDT reported on the regulatory conundrum this posed for the platform and highlighted instances of censorship. In 2022, CDT reported on a leaked internal document from Xiaohongshu revealing detailed censorship protocols instructing the company’s content moderators how to deal with “sudden incidents.”

Probing the implementation of censorship and other aspects of digital authoritarianism in China is an increasingly difficult task. Ariane Ollier-Malaterre, Emilie Szwajnoch, Alexander Trauth-Goik, Ausma Bernot, Fan Liang, and Ashley Poon explored the various challenges to this research in an article published on Friday in the Journal of Contemporary China: “Navigating Through The Fog: Reflexive Accounts on Researching China’s Digital Surveillance, Censorship, and Other Sensitive Topics.” In one section, Fan Liang lists censorship-related impediments to obtaining high-quality digital data: government-imposed restrictions on data access, industry (or platform-based) content moderation, and self-censorship on the part of survey respondents. Below, Fan provides more detail on platform-based restrictions on data access:

The second type of restriction comes from industry moderation, which affects areas ranging from social media platforms to survey firms. Many scholars collect social media data to analyze public opinion and policy processes. However, censorship and content moderation have been viewed as key tools used by platforms to monitor and regulate online content. More recently, major platforms have limited data access through their Application Programming Interfaces (APIs), making it increasingly difficult for scholars to download and collect large amounts of data from Chinese social media. While web scraping offers an alternative, this method may violate social media platforms’ policies and even China’s Data Security Law. Some scholars have developed solutions such as Weiboscope and WeChat scope, open-source platforms for collecting and visualizing data from Weibo and WeChat, respectively. Yet, these databases face challenges from both the platforms and China’s political environment. Moreover, industry moderation also involves the screening conducted by survey firms. For example, one of my research projects collected survey data to understand people’s attitudes toward official policies. My co-authors and I worked with a Chinese university and a large survey firm to conduct the online survey. In addition to the ethical board review at that university, our questionnaires were also screened by the survey firm and the Department of Industry and Information Technology. As a result, three questions were recommended for changes, as they focused on perceptions of trust in the central government. These questions were considered sensitive by the firm and the government, so my co-authors decided to remove them to launch the survey as soon as possible. This type of political screening has become a routine challenge for scholars conducting surveys in China, especially on politically sensitive topics. [Source]

Despite the growing censorship industry, Chinese and overseas internet users are constantly pursuing ways to circumvent online restrictions. Last week, on the occasion of World Day Against Cyber Censorship, Reporters Without Borders highlighted its ongoing work in developing Operation Collateral Freedom, a tool to help Chinese internet users access media outlets blocked by China’s Great Firewall:

Collateral Freedom is an initiative launched by Reporters Without Borders (RSF) in 2015, which uses the method of mirroring websites of media outlets to guarantee their access in countries in which they are blocked by the censors. Under this initiative, RSF has already unblocked more than 150 media outlets’ websites worldwide, listed in its [GitHub] page except when security concerns require media outlets to remain discreet.

RSF currently enables 33 media outlets censored in China to overcome the “Great Firewall of China,” one of the most sophisticated systems of digital censorship and control in the world, which was strengthened when Chinese leader Xi Jinping came to power. Through the created mirror sites, Chinese citizens can now access independent sources of information as an alternative to the official propaganda, often touching on increasingly taboo topics in China, such as political news, social conflicts, and the human rights violations carried out by the Beijing regime. [Source]

Departments which republished the article “Wuhan Novel Coronavirus Pneumonia Patient Dies After Leaving Hospital,” please immediately withdraw it! (March 5, 2020) [Chinese]

Please give low-key handling to reports on our plans to send charter planes to return some Chinese citizens from Iran. Brief factual reports are permissible, but do not extrapolate, decipher, comment, or republish foreign media reports. (March 5, 2020) [Chinese]

If reporting on confirmed cases among Chinese citizens repatriated from Iran, take information published by authoritative departments as standard. Do not stress patients’ ethnic minority status, do not comment, do not hype, do not exaggerate, and avoid triggering social panic. (March 5, 2020) [Chinese]

These continue a stream of epidemic-related orders over the previous several weeks. The first of the Iran-focused directives echoes a longer set of instructions issued on March 3, whose other subjects included repatriation of Hong Kong and Macao citizens from Hubei. Earlier directives had similarly restricted coverage of Hong Kong and U.S. citizens’ evacuation. Another order issued on January 15 contained extensive guidance on coverage of Iran’s downing of a Ukrainian passenger plane, whose 176 passengers and crew were all killed.

For now, do not report on the commendations from three ministries this morning. (March 5, 2020) [Chinese]

On March 5, three government agencies—the National Health Commission, Ministry of Human Resources and Social Security, and National Administration of Traditional Chinese Medicine—issued commendations to 113 healthcare teams and 473 individual healthcare workers, as well as several healthcare workers who died, "in the field of epidemic prevention and control." Those awarded posthumously include Dr. Li Wenliang, who was initially reprimanded for sharing information with healthcare colleagues about the then-unknown coronavirus, and who later died from the virus. His death became a rallying point for internet users demanding free speech and greater government transparency and accountability.

When reporting on the fight against the epidemic and the return to work and production, do not cover Xinjiang’s organizing of work positions for Uyghurs and other ethnic minority members in the interior. (March 5, 2020) [Chinese]

The Australian Strategic Policy Institute had published a widely reported paper on March 1 describing "the mass transfer of Uyghur and other ethnic minority citizens from the far west region of Xinjiang to factories across the country. Under conditions that strongly suggest forced labour, Uyghurs are working in factories that are in the supply chains of at least 82 well-known global brands in the technology, clothing and automotive sectors, including Apple, BMW, Gap, Huawei, Nike, Samsung, Sony and Volkswagen. This report estimates that more than 80,000 Uyghurs were transferred out of Xinjiang to work in factories across China between 2017 and 2019, and some of them were sent directly from detention camps."

•••

Be sure to use standard Party terms for grassroots organizations and related positions.
1. Do not describe a “Party committee secretary” as “community secretary”: “secretary” is a Party position, and the expression “community secretary” is non-standard.
2. Do not confuse “Party branch secretary” with “general Party branch secretary”: Party branches and general Party branches are both grassroots organizations of the Party, but there are differences in their scope and functions, and they cannot be mixed up. Likewise, general Party branch secretary and Party branch secretary are two different posts. (March 5, 2020) [Chinese]

Several other directives had similarly urged recipients to ensure terminological orthodoxy on matters from COVID-19 to territorial claims in the South China Sea. Others were proscriptive rather than prescriptive, barring the word "evacuate" to describe repatriations from Iran, "potentially controversial" terms such as "tracking" and "tracing" from reports on coronavirus-related big data analysis, and the derisive "copying homework" as a jibe at foreign countries’ handling of the epidemic.

Since directives are sometimes communicated orally to journalists and editors, who then leak them online, the wording published here may not be exact. Some instructions are issued by local authorities or to specific sectors, and may not apply universally across China. The date given may indicate when the directive was leaked, rather than when it was issued. CDT does its utmost to verify dates and wording, but also takes precautions to protect the source. See CDT’s collection of Directives from the Ministry of Truth since 2011.

The Guardian’s Amy Hawkins reported on the broader “stability maintenance” regime surrounding the meetings:

Security is heightened. Extra uniformed personnel have been deployed to stand guard on Beijing’s bridges – lest anyone attempt a stunt inspired by Peng Lifa’s protest at Sitong Bridge ahead of the 20th party congress in 2022. Guards at busy subway stations subject commuters to random scans of their identification cards.

Virtual private networks – apps used to tunnel through the firewall of internet censorship – slow down, as the authorities try to tighten their grip on the exchange of information with the outside world. It is imperative to the Communist party that the parallel sessions of the “Chinese People’s Political Consultative Conference”, an advisory body, and the National People’s Congress (NPC), China’s rubber-stamp parliament, run smoothly. Put together, the meetings are known as the Two Sessions, and represent the most important annual event in China’s political calendar. [Source]

The profile change blocks were highlighted on X by Teacher Li on Monday. Users reportedly received notifications such as the following, attributing the temporary change to “technical upgrades”:

Dear Blued users, greetings! We will be carrying out technical upgrades between 00:00 on March 3 and 23:59 on March 8. The following functions will be affected: suspend the modification of nicknames and signatures. We apologize for any inconvenience this may cause you, and appreciate your understanding and support! [Chinese]

A set of leaked internal corporate directions published by CDT in 2022 relayed instructions for similar temporary restrictions. These similarly applied to profile changes as well as new users’ profiles, emoji, gifs, and other functions, and ordered the use of “strategic phrasing” such as “upgrades/maintenance/improvements” rather than “supervisory requirements” or “sensitive period”:

Urgent! Network control notice—network control arrangements for 20:00 on June 2 to 08:00 on June 6

@Everyone, hello, we have received a notice from supervisors in Shenzhen and Xiamen, the specific requirements are as follows:

1. Customization function restrictions and content self-examination for 20:00 on June 3 to 20:00 on June 5 (network control period)

• All games must shut down customization features for all users including [changes to] display names and avatars, chat, and so on. Newly registered accounts can use default settings for display names, avatars, in-game vocalizations (random display names, random avatars, default vocalizations, etc)
• Forums and communities must bar users from changing avatars, display names, user profiles, etc. Other user-posted content must be individually inspected before publication; if this is not possible, these functions must be shut down.
• Self-inspection and usage blocks for sensitive content: all game forums and communities must conduct self-examination for emoji packs and other material incorporating candles, tanks, “89,” “64,” or potentially component elements (numbers, gifts, pictographs, audiovisual material). If these are found, please block their use.

2. Content inspection filtering and emergency duty for 20:00 on June 2 to 08:00 on June 6 (strict examination duty period)

• Content security platforms will coordinate to adjust strategy for strictly intercepting and attacking sensitive content
• Those that have not already implemented content security measures must, outside the aforementioned network control period, strengthen checks on newly registered users’ display names, avatars, and user profiles; step up content monitoring; and increase the frequency of inspections of user-posted content such as chat messages, forum posts, etc.
• During the strict examination duty period, all project teams are requested to make arrangements and contingency plans for staffing schedules. Throughout the Dragon Boat Festival holiday period, each shift must have at least two staff members constantly reachable by phone and able to respond to supervisory instructions or sudden incidents within 5-10 minutes.

3. Related announcements and compensation

• External announcements: notices such as product announcements, emails, etc. [about the above changes and restrictions] may not include terms such as “supervisory requirements,” “sensitive period,” etc.: use strategic phrasing like “game function upgrades/maintenance/improvements.”
• Reward/compensation distribution: Please refer to previous compensation [guidelines] for the January 30 network control [period]. (June 2, 2022) [Chinese]

Similar controls had also been introduced on Weibo, WeChat, and Alipay during the Party’s 19th National Congress in October, 2017, “in order to provide users with a better experience on our platform.”

At Rest of World in October 2023, Viola Zhou and Andrew Deck described how Blued had navigated the sensitive terrain of LGBTQ+ politics in China over the previous decade:

Blued, China’s most popular gay dating app, has set its sights on becoming the world’s largest social network for the LGBTQIA community, starting with an expansion in Southeast Asia and the U.S. But back home, Blued is facing growing uncertainty due to China’s tightening control over LGBTQIA content, and growing competition for younger users from other platforms.

Launched in 2012 by entrepreneur Ma Baoli, better known by his alias Geng Le, Blued has since deftly navigated China’s precarious political environment for LGBTQIA communities to become the country’s most used gay dating app. A central part of Blued’s strategy has been to offer HIV prevention and sexual health services that align with state public health initiatives, raising awareness of LGBTQIA issues, all the while steering clear of rights-based advocacy. Blued has survived, even as other queer dating apps in China have shut down: Competitors such as Zank were forced to close, and the lesbian-focused app Rela was pulled twice from Chinese app stores.

[…] The strategy shift [toward international markets] comes as the Chinese government intensifies crackdowns on LGBTQIA-focused organizations, events, and social media accounts, making it difficult for Blued to grow its business domestically. The company shut down its surrogacy service BluedBaby, which connected gay men in China with California-based surrogacy agencies, in the wake of a headline-making surrogacy scandal in 2021.[Source]

1. Do not reprint or cite foreign media commentary on sensitive issues involving Xinjiang.
2. Do not republish or hype inaccurate reports on epidemic control in Russia.
3. Strengthen checks on maps and place name labels and markers involving the South China Sea. (March 3, 2020) [Chinese]

Reminders related to the novel coronavirus pneumonia epidemic:

1. Give low-key handling to our government’s coordination of commercial charter flights to repatriate Chinese citizens from Iran. Brief factual reports are permissible, but do not extrapolate, decipher, comment, republish foreign media reports, draw connections to Sino-Iranian relations and cooperation, or use the term "evacuate." If reporting on accommodations for Chinese citizens following their repatriation, do so in accordance with information published by the relevant provinces and authoritative departments, and do not quote unverified online information. Do not conduct interviews with repatriated people or their friends and relatives in the absence of unified arrangements.
2. In general, do not publish reports on online information such as "vegetables given to support Hubei by other regions left to go to waste."
3. If reporting on the number of fatalities from novel coronavirus pneumonia in welfare institutions, senior citizens’ homes, mental hospitals etc., proceed in accordance with information published by authoritative departments. Do not cite data published by social organizations or foreign organizations, or information circulating online.
4. In the next few days, the Hong Kong and Macao governments will repatriate groups of their citizens from Hubei. Do not create reports for domestic audiences; reports for foreign audiences may proceed on the basis of unified deployment. (March 3, 2020) [Chinese]

These directives’ instructions on the COVID-19 epidemic echo many themes from previous orders, including the tone of coverage of the epidemic in other countries, avoidance of potentially inflammatory terminology and topics such as evacuations of Hong Kong, U.S. and other citizens, and standardization of sourcing with heavy emphasis on official releases. Other recent orders had also similarly called for rectification of names involving the South China Sea.

The order about "foreign media commentary on sensitive issues involving Xinjiang" came two weeks after The New York Times’ release of reporting on a major trove of leaked documents on China’s mass detention campaign in the region. It also coincided with a report from the Australian Strategic Policy Institute on various Western firms’ suspected use of forced labor in the region, and with criticism of the detention campaign by NGOs at the United Nations. March 3 saw the opening of an exhibition at the the Palais Des Nations in Geneva intended to rebut such condemnation during the 43rd regular session of the U.N. Human Rights Council. The display included "more than 100 pictures and videos presenting a beautiful, open, and richly-endowed Xinjiang" in which "people of different ethnic groups, thanks to social stability, are able to share the fruits of development and enjoy their life and work."

Since directives are sometimes communicated orally to journalists and editors, who then leak them online, the wording published here may not be exact. Some instructions are issued by local authorities or to specific sectors, and may not apply universally across China. The date given may indicate when the directive was leaked, rather than when it was issued. CDT does its utmost to verify dates and wording, but also takes precautions to protect the source. See CDT’s collection of Directives from the Ministry of Truth since 2011.

1. Without prior arrangement, do not report information that is likely to trigger public panic involving big data analysis of the numbers and destinations of people who left Wuhan and Hubei, the number of people with whom they have come into close contact, emerging epidemic situation hotspots, and so on. Do not use potentially controversial terms such as "tracking," "categorizing," "locating," "tracing," "routing," and do not report personal information such as names and phone numbers.
2. Reports on the epidemic and control situations in foreign countries should contain accurate, comprehensive, and objective information, and should not overly criticize or ridicule the "loopholes" or "mistakes" of the countries concerned. Do not make simple comparisons to China’s prevention and control measures, and do not use terms like "copying homework." (March 1, 2020) [Chinese]

Both parts of this order reflect the general push of directives on coverage of the COVID-19 epidemic to "control the temperature" of public opinion on topics including the initial outbreak; the new virus’s prominent early victim, whistleblowing doctor Li Wenliang; the prospects for a cure; and the epidemic’s economic impact. The second instruction echoes part of an extensive earlier directive issued on February 26, which also cautioned against potentially inflammatory coverage of other countries’ border restrictions against Chinese citizens. Other earlier directives aimed to avoid stoking international tensions by calling for "low-key" coverage of China’s efforts to obtain protective equipment from abroad "to avoid a public opinion backlash in the countries concerned and consequent obstructions to our overseas procurement work."

Since directives are sometimes communicated orally to journalists and editors, who then leak them online, the wording published here may not be exact. Some instructions are issued by local authorities or to specific sectors, and may not apply universally across China. The date given may indicate when the directive was leaked, rather than when it was issued. CDT does its utmost to verify dates and wording, but also takes precautions to protect the source. See CDT’s collection of Directives from the Ministry of Truth since 2011.

Please strictly enforce use of labels and place names in line with the Chinese government’s in maps of the South China Sea; avoid the use of maps or place names that violate our South China Sea policy or stance in news reports or on web pages to avoid the transmission of erroneous information to foreign countries. Materials for foreign publication, especially in countries with involvement in the South China Sea and those including maps or place names related to the South China Sea, should be stringently examined and strictly reviewed, and do not mention foreign publications that violate our South China Sea policies, claims, and positions. (February 29, 2020) [Source]

The immediate trigger for this directive is unclear, but its instructions reflect Chinese authorities’ longstanding insistence on rigid adherence to its territorial claims in written or graphical materials.

•••

In accordance with the National Development and Reform Commission and National Health Commission’s guidance on the selection and use of face masks, it is permissible not to wear them in well-ventilated outdoor areas while maintaining a suitable distance from others. In order to guide standardized protection measures amid the return to work and resumption of production in all departments and across the country, people wearing face masks should not be shown in video or photographs in reports on spring plowing, outdoor construction work, etc. in areas outside Hubei. (February 29, 2020) [Source]

These instructions show efforts to prevent unnecessary mask use in the face of stubborn supply constraints even for those on the front lines. A comparison of national guidance on mask use published in The Lancet on May 1 noted that although "some provinces and municipalities in China [including Shanghai] have enforced compulsory face mask policies in public areas […] China’s national guideline has adopted a risk-based approach in offering recommendations for using face masks among health-care workers and the general public." Various reports in the days and weeks surrounding this directive described mask shortages among medical workers, manufacturers’ struggles to meet demand, Chinese authorities’ seizure of millions of fake or substandard masks, and China’s efforts first to procure masks from abroad—a recurring subject in earlier directives—and then to reserve enough of its own output for domestic use as global demand surged. Governments elsewhere faced similar challenges in managing supply and demand of vital protective equipment. See more directives on the coronavirus pandemic from CDT.

Since directives are sometimes communicated orally to journalists and editors, who then leak them online, the wording published here may not be exact. Some instructions are issued by local authorities or to specific sectors, and may not apply universally across China. The date given may indicate when the directive was leaked, rather than when it was issued. CDT does its utmost to verify dates and wording, but also takes precautions to protect the source. See CDT’s collection of Directives from the Ministry of Truth since 2011.

Regarding the arbitration result in the case of swimmer Sun Yang and the International Swimming Federation (FINA), take the Chinese Swimming Association’s official statements as standard without exception, and do not infer, decipher, or comment. Do not translate foreign media reports without authorization. Strictly control the temperature, do not put stories on the news or main front pages, and do not send news app push alerts. Strictly manage all kinds of attacking or defamatory commentary. (February 28, 2020) [Chinese]

On February 28, three-time Chinese Olympic swimming champion Sun Yang was suspended from competing for eight years for a drug testing violation. A ruling by the Switzerland-based Court of Arbitration for Sport in a complaint by the World Anti-Doping Agency banned Sun from the upcoming games in Tokyo, and would "most likely end his career," reported The New York Times.

•••

1. The National Development and Reform Commission will soon publicly release minimum purchase price levels for the 2020 rice crop, and will also set limits on purchase amounts. In the absence of unified arrangements, do not report, comment, or republish.

2. Regarding the Central Propaganda Bureau and Central Culture Office’s launch in Wuhan of the “Voluntary Service and Care Campaign” for epidemic prevention and control, do not report for now until arrangements have been made. (February 28, 2020) [Chinese]

CDT editors were unable to find news in English or Chinese on the NDRC release mentioned in this directive.

The second point listed in this directive is the latest in a near daily succession of propaganda orders on the coronavirus pandemic. Several previous directives had also limited coverage of civil society aid in the midst of the crisis, especially donations from “disadvantaged social groups.”

•••

When some websites reported on the National Tax Office’s February 27 press conference, they abbreviated “digitized special VAT receipts” as “digital receipts,” creating a serious inaccuracy. Please change “digital receipts” to “digitized special VAT receipts” as soon as possible. If it cannot be changed, please immediately delete the relevant reports. (February 28, 2020) [Chinese]

On February 27 Chinese authorities pledged to finish drafting new financial legislation, including a law on value-added tax.

•••

For reports about research findings on the efficacy of pharmaceuticals, Chinese medicines, or vaccines against the novel coronavirus pneumonia, standardize the distribution workflow for news dispatches, and proceed in strict accordance with information published by the State Council Joint Prevention mechanism and authoritative departments such as the National Health Commission, Ministry of Science and Technology, etc. Do not rush to publish unverified information, and do not overstate or exaggerate curative effects. (February 28, 2020) [Chinese]

Among the many recent COVID-19-related propaganda directives since early January, several were aimed at medical research, vaccine development, and the promotion of Traditional Chinese Medicine for coronavirus prevention and treatment.

Since directives are sometimes communicated orally to journalists and editors, who then leak them online, the wording published here may not be exact. Some instructions are issued by local authorities or to specific sectors, and may not apply universally across China. The date given may indicate when the directive was leaked, rather than when it was issued. CDT does its utmost to verify dates and wording, but also takes precautions to protect the source. See CDT’s collection of Directives from the Ministry of Truth since 2011.

Reminders related to the novel coronavirus epidemic:

1. Keep a firm grasp when reporting on the coronavirus situation and control measures in foreign countries. Do not overly criticize or ridicule their "loopholes" or "mistakes." Do not make simple comparisons to China’s prevention and control measures. In order to avoid backlash from foreign countries, do not use terms like “copying homework.”

2. Concerning the adoption of necessary entry prevention, control, and quarantine measures taken towards South Korea, Japan, and other countries, report in strict accordance with statements from the Foreign Ministry and other authoritative departments. Do not report control measures taken by relevant locations without unified arrangement to avoid inflaming public opinion. Do not use extreme wording or triggering descriptions such as "uniformly prohibit," "strictly prevent," or "completely close." "Clickbait" behavior must be eliminated.

3. Regarding work resuming in tourist areas, restaurants, malls, etc., this must be handled in accordance with the principles of moderation. Do not highlight or aggregate reports. Guide the public to avoid large gatherings and to exercise sound judgment about going outside.

4. Concerning research and development into a vaccine for the novel coronavirus, further standardization of the publication and information control process is needed. Use only authoritative information issued by the State Council Joint Prevention mechanism, the Ministry of Science and Technology, and the National Health Commission. Don’t hastily publish unconfirmed information, and do not exaggerate the efficacy of vaccines developed by social organizations. Do not speculate about the vaccine prospects, and prevent inaccurate reports and confusing information. (February 26, 2020) [Chinese]

On the matter of the Hong Kong Special Autonomous Regional government’s announcement of arrangements for retrieving Hong Kong residents stranded in the mainland, do not push news segments in pop-ups, do not post on front pages, do not start hot topics in interactive sections, do not show in trends, and promptly manage extreme comments. (February 26, 2020) [Chinese]

Earlier entries in the stream of coronavirus-related directives throughout February had included similar guidance on the issues of foreign opinion, border closures and other restrictions, and medical research.

Hong Kong’s government had announced on February 24 that it would begin to evacuate some 2,700 citizens from Hubei after a month of lockdown. Such moves had reportedly been blocked by central authorities aggrieved at similar moves by foreign countries; one earlier directive contained detailed restrictions on coverage of the American withdrawal of consular staff from Wuhan soon after the lockdown began.

•••

All departments and terminals publishing video pieces must persist in using higher authorities’ source information. Do not arbitrarily use source content from other media outlets. (February 26, 2020) [Chinese]

On the basis of this sample, please clean up and delete the articles, photos, etc. [listed]. Report relevant data on the clean-up to the editor on duty. Take note, this sample should not be circulated. (February 26, 2020) [Chinese]

Please search for and delete videos and articles related to this video sample. If any information is cleaned up please report to the editor on duty. (February 26, 2020) [Chinese]

CDT has not confirmed what specific content was targeted in the last three directives above.

Since directives are sometimes communicated orally to journalists and editors, who then leak them online, the wording published here may not be exact. Some instructions are issued by local authorities or to specific sectors, and may not apply universally across China. The date given may indicate when the directive was leaked, rather than when it was issued. CDT does its utmost to verify dates and wording, but also takes precautions to protect the source. See CDT’s collection of Directives from the Ministry of Truth since 2011.