Experts say Spamouflage Dragon, China's yearslong disinformation campaign targeting the US, has largely failed due to China missing cultural context and more
but no one is looking at it. https://www.wired.com/... @WIRED Tori Elliott / @telliotter : There's a lot of talk about China's disinfo/propaganda efforts, but turns out... they're pretty bad at it? This story from @daithaigilbert https://www.wired.com/... Michael Clarke / @meclarke114 : Interesting piece but neglects China's relative success in deploying sustained disinfo and misinformation on Xinjiang over the past decade https://www.wired.com/... David Gilbert / @daithaigilbert : Despite running for over seven years with millions of accounts on dozens of platforms, China's state-run disinformation campaign has been an utter failure. I spoke to @elisethoma5 and @selectedwisdom to try and figure out why. https://www.wired.com/... @mrseanhaines : You all know all of this. And I've been hammering these points for 3 years. But this article is excellent, 100% on point, with this line particularly.. 🔥 https://www.wired.com/... [image] Daniel Funke / @dpfunke : Insightful story from @daithaigilbert on how Chinese disinformation campaigns have so far failed to have much real-world impact https://www.wired.com/... @wired : Earlier this month, a new report from Microsoft's Threat Analysis Center outlined how a Chinese disinformation campaign was now utilizing artificial technology in a campaign designed to inflame divisions in US politics ahead of the 2024 election. https://www.wired.com/... [image] David Gilbert / @daithaigilbert : Here are some of the AI-generated images China's state disinformation campaign has been spreading to trying and stoke division ahead of the up-coming US election. Like all CCP influence operations over the last 7 years no real people have seen them... https://www.wired.com/... [image] Alena Popova / @alenapopova : “I would describe China's disinformation campaigns as Russia 2014. As in, they're 10 years behind,” said the general manager of Microsoft's Threat Analysis Center. “They're trying lots of different things but their sophistication is still very weak.” https://www.wired.com/... @wired : Spamouflage has been running at a massive scale, but like pretty much all of Beijing's disinformation campaigns that have been waged in the last seven years, nobody saw it. Here's why: https://www.wired.com/... Doug Saunders / @dougsaunders : “Why China is So Bad at Disinformation” — a valuable article finally looking at the qualitative side of Beijing's influence campaigns. Russia managed to win itself a US election 8 years ago. China's managed to make everyone notice their ham-fisted efforts https://www.wired.com/... David Gilbert / @daithaigilbert : You may have recently seen headlines warning about the threat posed by China's disinformation campaign — which is now using AI. But the reality is that AI or not, China has been historically terrible at influence operations like this Here's why: https://www.wired.com/... @wired : The Chinese government-linked disinformation campaign, known as Spamouflage Dragon or Dragonbridge, used AI-generated audio and memes designed to grab user attention and boost engagement. https://www.wired.com/... [image]
Context & Ripple Effects
Spamouflage Dragon has been tracked as a large, cross-platform operation that promoted Chinese policies and targeted dissidents, as documented in a prior account of its platform infiltration. This assessment shifts the focus from the campaign's visible scale to whether it generated meaningful real-world influence.
It also reinforces the caution in earlier analysis of overseas Chinese influence operations: sustained activity and large account networks do not by themselves demonstrate persuasion. The reported use of AI-generated media makes that distinction more important as content production becomes easier to scale.
First-order effects
- Spamouflage Dragon's operators face evidence that years of account-building, memes, and synthetic audio have not translated into substantial US influence, with poor cultural fit and execution undermining reach.
- Platforms, researchers, and policymakers gain a clearer basis for separating operational presence from demonstrated audience impact when assessing this campaign.
Second-order effects
- Threat-intelligence teams may put greater weight on engagement quality, audience targeting, and offline resonance rather than using account volume as a proxy for influence.
- AI-generated content can still increase the campaign's output, as in the later overview of AI-enabled Dragonbridge activity, but scaling supply does not solve the cultural-context problem identified here.
Third-order effects
- If this pattern persists, state-backed influence operations will be judged less by the size of their networks and more by evidence of credible audience persuasion, raising the bar for public attribution and policy responses.
- The broader contest may shift toward operations with local knowledge and authentic distribution channels; synthetic media is an amplifier, not a substitute for either.
The trend: This is one data point in the shift from measuring disinformation by content volume and bot networks to measuring whether it reaches and persuades real audiences.