Interesting Substack article about Chinese state media’s limited success in using overseas social media to influence foreign audiences, with special reference to Twitter…
As yet, not many state media influencers have migrated over [to BlueSky].
Will they? Almost certainly.
These are media professionals who are paid to post, work full-time, and have the pressure of an entire autocratic state behind them. They are fulfilling Xi’s call that: “China must amplify its voice so that the nation’s message is heard overseas.”
The great prize of Twitter was never its number of users: it was who they were. Journalists, academics, and politicians… the Chinese state saw it as the fount of western hegemony, and an ideological battle place. If BlueSky becomes Twitter’s spiritual successor, it too will be a key mount to conquer, drown out, or destroy.
If state media does arrive in earnest, it may take time. It’s noticeable how loose BlueSky’s algorithm currently is, how difficult it is to game. No promoted tweets, blueticks or convoluted algorithm to manipulate here. BlueSky’s simplicity may yet be its best protection for its purity.
With its currently predominantly western liberal leaning, the platform also will be an uphill challenge for those looking to push overtly nationalistic viewpoints.
Even after years of honing their communication on Twitter, many state media accounts and their offshoots remain too abrasive, too anthema to mainstream ideals, and (potentially most damaging of all).. too dull to entice new broad audiences.
Many only became large on Twitter via brute-forcing the system — latterly via Musk’s paid-for-blueticks. And yet, even with this algorithmic help, it’s interesting how KPI’s for many in recent months have plateaued, or gone backwards.
I recently semi-moved to Bluesky, though a lot of worthwhile accounts and material remain on Twitter. But one glaring difference is that it’s relatively easy to block and generally avoid annoying garbage on the newer platform – which clearly angers and frustrates conspiracy theorists, tankies and other bores/bots who want to shove various forms of idiocy into your face. And state media influencers.
The SCMP reports a push to use AI in China’s overseas propaganda efforts…
Communist Party mouthpiece People’s Daily called on the country to enlist AI in an information campaign to debunk US-led narratives about China, highlighting the perceived urgency in a series of front-page articles on Friday.
In the articles, researchers suggested strategies that included harnessing technology to counter Western media, which they said painted an unfair image of China, and hindered the nation’s ability to form a united view of itself.
…In another article, Cheng Manli, a professor at Peking University’s school of journalism and communication, pointed to the “urgency and necessity of advancing the restructuring of the international communication landscape”.
As geopolitical tensions have worsened, the US-led West has continued to escalate its suppression and containment of China, and the Western media’s incorporation of China-related issues into global narratives often distorts facts, according to Cheng.
She said such narratives “not only affect the international community’s accurate understanding of China, but also significantly disrupts the formation of consensus among the Chinese domestic audience”.
“Breaking the West’s monopoly on discourse and dismantling international stereotypes about China have become increasingly important to present a truthful, multidimensional, and comprehensive image of China to the world,” she wrote.
So Western media ‘hinders the nation’s ability to form a united view of itself’ and ‘significantly disrupts the formation of consensus among the Chinese domestic audience’. Does it really do that? Is it even that accessible to most people in China? Or are Beijing’s leaders simply suffering a haunting fear that someone, somewhere might be exposed to differing views? (To paraphrase HL Mencken.)
Desmond Shum describes AI’s ability to write (and explain how to write) official Chinese documents…
Apparently, over 50% of ChatGPT’s mainland Chinese users are CCP bureaucrats seeking help with their official documents. Yes, bureaucrats! ChatGPT has been so thoroughly fed with endless streams of Party prose that it now churns out CCP-style articles with the finesse of a seasoned propaganda officer.
He asked ChatGPT to summarize the characteristics of CCP official writing, and got the response….
“CCP official writing and “party Eight-Legged Essays” are often criticized for being formulaic, verbose, and lacking substance. Key characteristics include:
1.Formality and Routine: Heavy use of fixed slogans and a template structure, prioritizing political correctness over originality or practical solutions.
2.Lack of Substance: Abstract language avoids real issues, with vague terms like “further promote” replacing concrete measures.
3.Verbose and Repetitive: Redundant phrases and repeated points make texts long-winded and unfocused.
4.Authoritative Tone: Commanding language emphasizes infallibility but lacks approachability.
5.Ambiguity: Avoids specifics, leaving statements open to interpretation.
6.Rigid Style: Uniformity and repetition dominate, making content monotonous and unengaging.
While these features ensure ideological consistency, they often neglect logic, flexibility, and real-world relevance, reducing practical impact.”
I asked a free ChatGPT-style app to ‘Write a press release complaining about a critical newspaper article in the style of the Hong Kong government’. The result was a massive – stunningly bland and moderate – fail, with such phrases as…
The government respectfully urges [Insert Newspaper Name] to revisit its editorial practices and strive for a more responsible approach to reporting that prioritizes public interest and community solidarity.
There’s no substitute for authentic human mouth-frothing and ranting.
Why would CCP propagandists bother with Bluesky (20m users) instead of X (600m users)?
I asked ChatGPT the same question as above and it delivered a similarly bland answer. I then asked it to add in a few of the HK government’s favourite angry phrases and it got a bit better. However, not a single use of “so-called” which was very disappointing.
I was going to ask it to reproduce the meaningless word salad of a discourse on ‘Xi Jinping thought’ but didn’t want to be the first person to drive an AI to suicide….
@Marius: because it is growing exponentially at an amazing rate. Already exceeding activity of Threads. Bluesky also allows you to interact freely without having your reply throttled, and replies full of pornbots, Musk-fellators and right-wing trolls only. One can even opt to have no algoritm interfering whatsoever. It’s freeing, try it!
I got kicked out of Twitter 6 years ago after I asked Tung Chee-wah to simply die. Today I signed up with Bluesky and it is refreshing. Yes, try it.
2 and 5 are pretty much the same point…as are 3 and 6… the so-called ‘generational artificial intelligence’ has a way to go before it can undermine the resolute propagandists of our Fatherland