Chinese-linked online network is stoking U.S. political divisions: report
Add Axios as your preferred source to
see more of our stories on Google.
In this photo illustration the logo of US online social media and social networking site 'X' (formerly known as Twitter) is displayed centrally on a smartphone screen alongside that of Threads (L) and Instagram (R). Photo: Matt Cardy/Getty Images
A Chinese influence operation of inauthentic social media accounts to spread anti-Western sentiment is using fake accounts posing as American citizens ahead of the 2024 election, a report released Tuesday found.
The big picture: Americans have been inundated by media manipulated by China, Russia and Iran in recent years attempting to deepen stateside divisions as the election approaches.
- The Chinese state-linked disinformation campaign known as "Spamouflage" has used increasingly aggressive tactics in its attempts to influence American online discourse, intelligence company Graphika reported.
- Some of the most recent political content was "almost certainly AI-generated," the report found.
State of play: Graphika has monitored the operation's use of fake and hijacked accounts since 2019, describing the network to be "ultimately low impact" five years ago and stating in its most recent report that the campaign rarely directly mentioned U.S. elections in 2020.
- But prior to the 2022 midterms, those accounts started to engage with election-related topics, the report found. Since mid-2023, those efforts have only deepened, with Spamouflage pages amplifying content criticizing candidates, casting doubt on the electoral progress and delving into sensitive issues such as gun control and the Israel-Hamas war.
Zoom in: Graphika identified 15 accounts on X and one on TikTok claiming to be Americans or U.S.-focused activists, as well as one cross-platform persona posing as an inauthentic media outlet.
- At least five accounts Graphika observed explicitly claimed to be American voters, soldiers or someone who "love(s) America" but had grown disappointed with the U.S. government and the Biden administration.
- Harlan Report, a cross-platform account purporting as a U.S. conservative media outlet and influencer that amplified pro-Trump content, garnered significant audiences online compared to other Spamouflage pages, with one post mocking President Biden collecting 1.5 million views, per the report.
- The account shifted personas on its X account several times: posing as a New York-based veteran critical of Biden, a 29-year-old Trump supporter with a likely AI-generated profile picture and, most recently, a 31-year-old Republican influencer based in Florida. TikTok has banned the Harlan Report, the Wall Street Journal reported, as well as the account has been suspended on X.
Case in point: One persona that typically presents as a pro-China outlet called "Deep Red" rebranded one of its X accounts to "Common fireman" and has recently focused that page's content on U.S. political issues.
Zoom out: Since at least 2017, Spamouflage has used thousands of accounts to share low quality content across more than 50 platforms and websites, an Institute for Strategic Dialogue report published in April found.
Between the lines: While the ISD's 2024 findings detailed four accounts mimicking Trump supporters — part of a tactic dubbed "MAGAflage" — the latest Graphika report built on that research and presented a broader lineup: Several posts criticized the former president, a method tied to a broader goal to exacerbate social divisions.
- The monitored accounts worked together and with the wider campaign network to amplify posts and content, Graphika found.
- "In conjunction with ISD's analysis, our findings suggest that Spamouflage's attempts to pose as U.S. users are more expansive than previously reported," Graphika's key findings read.
Thought bubble: Chinese influence operations have tried many of these tactics before, Axios' Sam Sabin writes.
- The real test is whether any of these social media narratives will actually stick and spur the division that Beijing wants to see.
Go deeper: OpenAI says its tools were used in foreign influence campaigns
