Graphika: Huawei officials retweeted messages from fake accounts on Twitter to spread a pro-5G influence campaign in Belgium; Huawei is investigating the claim
A covert online push to sway telecommunications policy in favor of the Chinese company may presage a new twist in social manipulation. Source: Graphika .
New York TimesAdam Satariano
Context & Ripple Effects
Graphika's finding lands mid-arc in Europe's 5G vendor fight: Huawei had already grown from an upstart into an infrastructure juggernaut during the 4G cycle (riding licensed and sometimes stolen tech to undercut rivals), and Belgium matters because Brussels is where EU telecom policy gets made. The twist here is that the retweets came from Huawei's own officials' accounts, not just anonymous bots.
The report also foreshadows what came after: within months, investigators documented an army of fake accounts covertly amplifying Chinese diplomats and state media on Twitter (retweeting their posts tens of thousands of times), and by 2023 Belgian intelligence had put Huawei's Brussels lobbying office on its watchlist over possible direct ties to China's government.
First-order effects
Huawei's 'we are investigating' response puts named company officials — not faceless troll farms — at the center of a covert policy-push allegation, raising the stakes for its 5G sales pitch to Belgian and EU decision-makers.
Belgian policymakers weighing 5G vendors now have network-analysis evidence that pro-Huawei messaging reached them through inauthentic accounts, giving security agencies fresh material for vendor assessments.
Second-order effects
The finding feeds directly into the scrutiny that followed: Belgian intelligence's later decision to watchlist Huawei's Brussels lobbying office treats the office itself, not just its messaging, as a potential instrument of state influence.
Twitter faces pressure to act on official accounts that amplify coordinated fakes, while firms like Graphika gain a commercial role as the de facto auditors of influence operations for governments and platforms.
Third-order effects
If the pattern holds, telecom procurement in Europe merges with counterintelligence: vendor selection becomes less about equipment specs than about whether a supplier's public-affairs machinery is arm's-length from its home government.
Influence operations appear to be industrializing around private contractors and layered authenticity — real officials' accounts lending credibility to fake ones — making attribution-dependent defenses like Graphika's a standing part of the policy environment.
The trend: State-linked influence campaigns are evolving from pure bot amplification toward hybrid networks that borrow the credibility of real officials' accounts, with commercial forensics firms like Graphika setting the pace of exposure.
🆕 Our latest report uncovers a covert pro-Huawei influence campaign using fake accounts and AI-generated faces (again!) to criticize the Belgian government's #5G policy. 🗞 In @nytimes here: https://www.nytimes.com/...
“FT also found a handful of what appeared to be fake accounts dedicated to reposting Huawei content. One such account, for example, only had four followers including Huawei's verified German account. Several bot accounts were suspended by Twitter..” https://www.ft.com/...
A cluster of inauthentic Twitter accounts amplified and at times created articles attacking the Belgian government's plans to limit access of “high-risk” suppliers to its #5G network, particularly Chinese firms #Huawei and #ZTE. Read the full report: https://graphika.com/...
Great, important story by Adam. Now think who else might be using the same tactics. It's the direction of travel and the sports world is not immune https://twitter.com/...
„Clumsily executed & easy to identify". How faux journalism sites like EU Reporter & lawyers paid to write pro Huawei opinions become part of Huawei's online campaigns pushed by fake accounts & Huawei trolls. https://www.nytimes.com/...
LR @RepMcCaul: “More proof of #Huawei's nefarious actions & risk to our national security, @satariano exposes another manipulation effort. Huawei steals US IP, threatens 5G security, & supports #CCP genocide & should remain on the #EntityList. (1/2) https://www.nytimes.com/...
I am always impressed with the depth and breadth of @Graphika_NYC's research and this is no different. They uncovered a pro-Huawei influence campaign in Belgium. As these campaigns become more common this kind of research in the hands of leaders is more critical than ever. https:…
hmm...didn't I read something in the EU Digital Services Act about curbing inauthentic behaviour? Do hurry up please. These campaigns are both tiresome and make a laughing stock of any possibility of ‘civic discourse’ (another EU term). https://twitter.com/...
The latest entry in a long line of examples where pro-#Huawei messaging isn't always what it seems ... “The effort suggests a new twist in social media manipulation”: https://twitter.com/...
We found this network when it was boosted by Spamouflage, a pro-China operation. Independently, @mvanhulten of @TI_EU and @ArbiterOfTweets of @Knack found it with different methods. It's not a friendly environment for fake campaigns, folks. https://graphika.com/...
More from @Graphika_NYC & @benimmo about the operation, including how Huawei execs amplified fake Twitter accounts created with A.I.-generated profile pics. Others including @ArbiterOfTweets & @mvanhulten also identified the suspicious activity. https://graphika.com/...
“Kevin Liu, Huawei's president for public affairs and communications in Western Europe, who has a verified Twitter account with 1.1 million followers, shared 60 posts from the fake accounts over three weeks in December” https://twitter.com/...
Not a PR pro or anything, but responding to concerns about your company's good faith and transparency by ordering up a Twitter bot campaign complete with fake GAN-generated profile pics does not seem like the right way to work the problem. https://twitter.com/...
Little-known Huawei execs in Europe all built up hundreds of thousands of followers on Twitter in a matter of weeks, many of which appear inauthentic They've now lost thousands of followers after the FT asked Twitter about the phenomenon Story w/ @SVR13 https://www.ft.com/...
Great report by @satariano on a fake network that @Graphika_NYC (and others) found in December. Twitter accounts with GAN faces, boosting Huawei, boosted by Huawei execs, and attacking Belgium's 5G policies. Not enough evidence to prove who ran them. https://www.nytimes.com/...