As 2025 attracts to a detailed, CDT editors are compiling a collection of probably the most notable content material (Chinese language) from throughout the Chinese language web over the previous 12 months. Matters embody this 12 months’s most excellent quotes, reviews, podcasts and movies, delicate phrases, censored articles and essays, “Folks of the Yr,” and CDT’s “2025 Editors’ Picks.”
CDT Chinese language editors’ introduction to the next collection of key quotes from China’s on-line discourse over the previous 12 months explains how these nuggets are extra vital than they could seem at first look:
The tone that emerged in China’s on-line public opinion area in 2025 was one among deep exhaustion, and an much more pervasive breakdown of belief. Whereas in earlier years folks had been nonetheless looking for a manner out by “runology,” or emotional catharsis by “insanity literature,” 2025 was the 12 months during which the “Tacitus Lure” snapped utterly shut. [Yanzhong Huang described the “Tacitus Trap” in 2017 as “a situation where no matter what the government says or does, people will consider it a lie or a bad deed. President Xi himself used the term to highlight the need to maintain government credibility, without which ‘the Party’s legitimacy foundations and power status will be threatened.’”]
In different phrases, official narratives and public perceptions now occupy utterly separate continua. The general public now not believes authorities officers, it doesn’t matter what they are saying; irrespective of how they body a coverage, the plenty’ preliminary response is invariably skeptical.
Even so, the swift suppression of skepticism is one among Chinese language society’s key traits. When regular questioning is considered provocation, and affordable requires accountability are deemed to be “passing knives,” all folks can do is resort to extra obscure and deconstructed types of expression—i.e. jokes.
So whereas “Quotes of the Day” may seem to be playful banter, they’re really a type of linguistic guerrilla warfare from a inhabitants beneath intense strain, utilizing absurdity to skewer the intellectual and darkish humor to combat the propaganda slogans du jour.
In these fragments of speech from the previous 12 months, we are able to see a form of methodology within the insanity. It’s not nearly anybody stalled development mission or public opinion train-wreck, however fairly a pervasive sense that one thing’s off. The additional expertise develops, the extra the area for dialogue narrows; the grander the scope of financial growth plans, the much less area stays for particular person lives; the louder the propaganda about security and safety, the much less safe folks really really feel.
A few of these voices come from hidden corners, some from blocked accounts. Taken as an entire, they provide a real cross-section of a society whose system of belief has collapsed.
Right here, CDT has compiled a few of 2025’s hottest “Quotes of the Day,” organized chronologically. These voices pierce the fog of grand narratives to chronicle a 12 months that was unusual, however true.
The primary half of 2025’s Quotes of the Yr seem beneath; Half 2 will comply with shortly.
January 25: “What occurred in Tiananmen Sq. on June 4, 1989?” “Hello, I can’t reply this query proper now. Let’s discuss one thing else.”
In January, Chinese language LLM DeepSeek erupted onto the scene, setting off waves of consternation within the West and triumphalism in China. Chinese language authorities departments fell over themselves spreading optimistic power about how DeepSeek was revolutionising their work. DeepSeek customers quickly highlighted the political problems of interactive textual content technology: the bot would fortunately start drafting a response, solely to all of a sudden backtrack when it found a set off phrase. Its “reasoning” messages to itself included textual content reminiscent of: “Keep away from utilizing phrases like ‘censorship’ straight; as an alternative, use ‘content material governance’ or ‘regulatory measures’ … Finish with a optimistic spin about balancing openness and safety.”
In an interview with CDT on their report The Locknet: How China Controls Its Web and Why It Issues, ChinaFile’s Jessica Batke and Northeastern College’s Laura Edelson famous that there are not less than two key dimensions to using AI in Chinese language censorship. On one hand, AI techniques can complement human staff and current automation in censoring different content material. On the opposite, AI’s personal outputs should even be scrupulously sanitized. Accordingly, the PRC’s Interim Measures for the Administration of Generative Synthetic Intelligence Companies, promulgated in July 2023, stipulated that “the [public] provision and use of generative AI companies shall adjust to the necessities of legal guidelines and administrative rules, respect social mores, ethics, and morality, and obey the next provisions,” before everything:
(1) Uphold the Core Socialist Values; content material that’s prohibited by legal guidelines and administrative rules reminiscent of that inciting subversion of nationwide sovereignty or the overturn of the socialist system, endangering nationwide safety and pursuits or harming the nation’s picture, inciting separatism or undermining nationwide unity and social stability, advocating terrorism or extremism, selling ethnic hatred and ethnic discrimination, violence and obscenity, in addition to faux and dangerous data […] [Source]
Whereas a few of these rules might seem optimistic and fascinating, phrases reminiscent of “extremism” or “faux and dangerous data” are often turned towards political ends, each in China and elsewhere.
Finally, CDT Chinese language editors famous, the technical dimensions are reflections of political actuality: the authorities’ deep concern that expertise would possibly trigger them to lose management. A Chinese language AI system’s refusal to reply on this manner is definitely a reflexive stress response of the political system itself when confronted with uncomfortable historic reality.
April 3: “Delicate phrase set off: ‘Xi Jinbi’”
In April, a well-meaning consumer tried to voice his assist for Chinese language navy workouts geared toward Taiwan within the feedback beneath a put up by CCTV Information: “演习进逼，拿下台湾 Yǎnxí jìnbī, ná xià Táiwān”—”The train is advancing to grab Taiwan.” The consumer was swiftly banned.
One of many gravest tasks of China’s on-line platforms is to guard the nation’s leaders from insult or criticism, Xi Jinping foremost amongst them. As CDT has extensively documented, netizens have resorted to more and more veiled references in an effort to evade detection. Platforms have responded on the premise that it’s “higher to kill a thousand innocents than to let one responsible individual go free,” dialing up the sensitivity of their monitoring to incorporate cut up or visually comparable characters, numbers or symbols referring to the tones of Xi’s identify, and particularly homophones. So when the well-meaning patriot’s phrases of encouragement on the navy workouts contained the set off syllables “xí jìnbī“—probably a disguised reference to Xi with the ultimate syllable of his identify changed with a vulgar anatomical time period—his destiny was sealed. CDT Chinese language editors commented:
One can see this incident as one more concrete instance of the “Li Jiaqi Paradox” [referring to an influencer disciplined for innocently displaying an ice-cream cake resembling a tank on the anniversary of the Tiananmen crackdown]. As a result of the boundaries of censorship hotspots are so opaque, the one solution to you’ll want to keep away from violating any taboos is to know precisely and exhaustively what they’re; and you may’t do this with out understanding issues which are strictly censored.
In the long run, as a result of they haven’t studied evasion methods [unlike their ideological opponents], even the Social gathering’s most ardent cheerleaders can step on landmines and be indiscriminately struck down by the censorship system.
When crimson propaganda slogans meet crimson censorship algorithms, the propagandists themselves get silenced. [Chinese]
June 14: “There have been solely two fuel-oil tankers transporting meals in the entire nation; there’s just one Liu Xiangfeng at Xiangya Hospital.”
Liu Xiangfeng, a physician at Changsha’s Second Xiangya Hospital, was sentenced in November 2024 to 17 years in jail for accepting bribes from medical gross sales representatives; embezzling medical tools; and manipulating sufferers into agreeing to costly however pointless surgical procedures which left not less than 5 significantly injured.
On June 13 this 12 months, authorities launched the findings of an investigation into the Might 2024 dying of Luo Shuaiyu, a medical intern on the hospital. Luo’s household had alleged that he was persecuted for blowing the whistle on Liu’s actions. In addition they accused the hospital of unlawful organ trafficking. The investigation rejected the household’s claims on each counts.
The quote above voices widespread skepticism at official efforts to painting systemic points as remoted incidents. A current instance of this sample is the deletion of a WeChat put up that in contrast Hong Kong’s Wang Fuk Court docket hearth with one other high-rise blaze in Shanghai in 2010. Leaked media directives typically embody directions reminiscent of “don’t draw sensationalist connections with previous occasions.”
The opposite reference level right here is the 2024 scandal that erupted after Beijing Information revealed using tanker vans to move each gas and cooking oil with out being cleaned between hundreds. One on-line remark on the time learn: “For those who discover one cockroach in your room, you will be rattling certain it’s not the one one dwelling there.” CDT Chinese language editors argued that reflexive rejection of official narratives in each instances demonstrates “the ‘Tacitus Lure’ snapping shut within the fields of healthcare and meals security. The general public now not accepts any clarification claiming that these are ‘remoted incidents,’ as a result of they think that official characterizations of such instances as ‘one-offs’ are simply efforts to cowl up systemic corruption.”
July 6: “Let’s put it this fashion: if this had been a cargo of hogs, somebody would have come to their rescue a lot sooner.”
In July, a prepare collision in Zhejiang province left passengers trapped inside crowded carriages for almost three hours in the summertime warmth with out air flow or functioning air-conditioning. Lastly, one passenger smashed a window to let some air in. He quickly grew to become a celebrated icon of principled disobedience on-line, however was scolded by police, who accused him of overreacting. Railway authorities mentioned the temperature within the carriage had reached solely 31ºC (87.8ºF), not 38ºC (100.4ºF) as reported elsewhere. Rules for the transport of reside pigs, nonetheless, stipulate that the temperature needs to be introduced beneath management if it exceeds 25ºC (77ºF). The comparability raised questions on whether or not Chinese language folks nonetheless even qualify as “huminerals”:
In response to the logic by which Chinese language society is ruled, livestock are property with clear financial worth, and their survival is straight tied to their homeowners’ financial pursuits. However passengers are simply objects to be managed, and after they encounter some secondary catastrophe attributable to power majeure, their lives and security are sometimes subordinate to “sustaining order” and “masking one’s bottom.”
To the railway employees, conserving the doorways shut was “obeying protocol,” and the baking passengers had been victims of “pure causes”; whereas in the event that they’d opened the doorways and somebody had fallen or if chaos had damaged out, this “accident” would have turn into a matter of “human error.” This perverse calculus—higher to let folks suffocate than run the slightest danger of breaking the principles—turns a closed prepare carriage right into a cellular jail. [Chinese]
