facebook phone app
© AP / Jenny Kane
Data provided by the social media giants showed a web of linked accounts pushing pro-US and anti-Russia narratives.

Twitter and Facebook have deleted a network of suspicious accounts said to be involved in "covert influence operations" online, including efforts to promote "pro-Western" sentiment and demonize US adversaries, according to company data analyzed by researchers at Stanford University.

Partnering with social media analytics firm Graphika, the Stanford Internet Observatory issued a report on Wednesday outlining the alleged influence ops, citing datasets provided by both Twitter and Meta, Facebook's parent company.

Between July and August, the platforms removed "two overlapping sets of accounts" for "manipulation," "spam" and "coordinated inauthentic behavior," the researchers said, noting that they had found a similar "interconnected web of accounts" on six other social media sites. The accounts used "deceptive tactics" to "promote pro-Western narratives," and many continued to operate for nearly five years.

"These campaigns consistently advanced narratives promoting the interests of the United States and its allies while opposing countries including Russia, China, and Iran," the report said, adding that the accounts have "heavily criticized" Moscow in particular. "To promote this and other narratives, the accounts sometimes shared news articles from US government-funded media outlets, such as Voice of America and Radio Free Europe, and links to websites sponsored by the US military."

Though neither Twitter or Meta could say who was operating the accounts, Twitter named the US and the UK as the "presumptive countries of origin," while Meta said the activity was traced back to the United States.

The report also highlighted a potential link to the US military, as some of the most-followed accounts in the dataset publicly declared some connection to the Pentagon. While those accounts were not identified in the report, a military spokesman told the Wall Street Journal that officials would look into the issue.

The inauthentic social media handles often created fake personas using AI-generated profile pictures, and "leveraged" memes, short videos, hashtag campaigns and online petitions - in what the report deemed "the most extensive case of covert pro-Western [influence operations] on social media" ever seen by independent researchers.

However, the data also hinted at the limitations of such campaigns, as the "vast majority" of the hundreds of thousands of posts reviewed for the report "received no more than a handful of likes or retweets." Moreover, fewer than one-fifth of the accounts in question, or 19%, had more than 1,000 followers before they were removed, suggesting they had little influence.