Researchers detail a pro-US, five-year influence operation on Twitter, Facebook, Instagram, and other platforms in the Middle East and Central Asia

Stanford Internet Observatory 2022-08-25

Context & Ripple Effects

The Stanford finding adds a pro-US operation to a record that had largely centered on Iran-linked networks: Facebook had previously banned coordinated Iranian and Russian networks, while reporting traced an Iranian broadcaster's use of hundreds of fake accounts back to 2012. The relevance is not simply the political alignment of the campaign, but the repeated use of the same major social platforms for cross-border influence.

The case also predates Meta's later removal of hundreds of accounts tied to campaigns from multiple countries, indicating that platform enforcement has had to address a widening set of state-linked or state-aligned actors rather than a single source.

First-order effects

Twitter, Facebook and Instagram face renewed scrutiny over how accounts and content tied to a five-year pro-US operation operated across their services in the Middle East and Central Asia.
Stanford Internet Observatory's documentation gives platform trust-and-safety teams a concrete network pattern to assess alongside the Iranian-origin networks Facebook had already removed.

Second-order effects

Platforms' enforcement frameworks are pressured to apply consistently across campaigns aligned with different governments, rather than treating coordinated inauthentic behavior as a problem associated with one geopolitical bloc.
Researchers and investigators gain another cross-platform case for comparing how influence operations use Facebook, Instagram and Twitter, increasing the value of shared evidence and coordinated takedown work.

Third-order effects

If disclosures continue to identify campaigns from rival and allied actors alike, platform governance will be judged increasingly on whether enforcement is actor-neutral and evidence-based.
The longer-term contest shifts from isolated account removals toward identifying coordinated behavior that can migrate among major social platforms and regions.

The trend: Cross-platform influence operations are turning content moderation into a geopolitical governance problem in which platforms must enforce against coordinated behavior regardless of its sponsor's alignment.

Discussion

@noupside Renee DiResta on x
📢1/ Today Stanford Internet Observatory & @Graphika_NYC are releasing a report analyzing what we believe is the first major covert pro-US/Western influence operation suspended by Twitter and Meta. Each has just released a data set of accounts involved. https://t.co/U9lpphJaxp htt…
@alexstamos Alex Stamos on x
@arawnsley ... And Meta and Twitter's influence operations teams. They found the campaign and shared the data. The other platforms we mention, however. . .
@chrissgiles Christopher on x
💻📢 New Twitter takedown report. I analyzed a pro-Kremlin network that featured accounts - created in bulk on 23 January 2022 - that amplified narratives in the lead-up to Russia's invasion of Ukraine. https://cyber.fsi.stanford.edu/ ...
@noupside Renee DiResta on x
@Graphika_NYC The covert accounts had low engagement, illustrating the limitations of using inauthentic tactics to build influence online. The vast majority of posts and tweets reviewed received no more than a handful of likes or retweets. Avg was <1.
@shelbygrossman Shelby Grossman on x
2) Some accounts had profile photos made from doctored images of actors and models, some collages of multiple images! https://twitter.com/...
@stanfordio @stanfordio on x
📢 Today @stanfordio is releasing a report co-authored w/ @Graphika_NYC analyzing accounts suspended by Twitter and Meta. These accts participated in pro-US influence ops that targeted people in the Middle East and Central Asia from 2012-2022 🧵⤵️ 🔗https://cyber.fsi.stanford.edu / …
@kimzetter Kim Zetter on x
Covert influence ops pushing pro-US, pro-West messaging are apparently not as successful as Russia/China ops. “The covert accounts had low engagement...The vast majority of posts and tweets reviewed received no more than a handful of likes or retweets. Avg was <1.” https://twitte…
@arawnsley Adam Rawnsley on x
This is a BFD. We're used to seeing covert pro-Iran/Russia/China/Saudi/UAE/ Nicaragua/Philippines/etc social media influence campaigns get ID'ed and booted by the big platforms like Facebook/Twitter. First time we're seeing a pro-US campaign ID'ed & booted https://twitter.com/...
@mikedotwaves Mike Wawszczak on x
@karaswisher is it? seems like a failure https://twitter.com/...
@karaswisher Kara Swisher on x
@mikedotwaves It gives the Russians something to point to
@arawnsley Adam Rawnsley on x
Great work by @stanfordio & @Graphika_NYC
@sarahleah1 Sarah Leah Whitson on x
“The accounts behind the operation often posed as news outlets or took on the personas of people who didn't exist, posting content in at least seven languages, including Russian, Arabic and Urdu.” https://www.nytimes.com/...
@washingtonpost @washingtonpost on x
Facebook and Twitter disrupted at least two covert influence campaigns that targeted users in the Middle East and Asia with pro-western perspectives about Russia's invasion of Ukraine, according to a new report. https://www.washingtonpost.com/ ...
@sheeraf Sheera Frenkel on x
An influence campaign tried to use memes and false news reports to paint a rosy picture of America. As @noUpside points out, it's more like the influence campaigns run by China, than those run by Russia/Iran. https://www.nytimes.com/...
@mikedotwaves Mike Wawszczak on x
@karaswisher eh, they are well-versed in manufacturing things to point to if they can't find something off the shelf. US has a long history of this in the pre-social media age (propaganda radio stations, etc). This seems like the same thing, no? Just more poorly executed
@zachsdorfman Zach Dorfman on x
@KimZetter Imagine the RoI on that.
@arekfurt @arekfurt on x
I almost guarantee you Twitter and Facebook wouldn't burn information operations that might be from U.S. gov without checking with the U.S. gov first. Which, if true, raises interesting questions about what was really happening here. https://twitter.com/...
@joetidy Joe Tidy on x
Unusual and refreshing to see fingers pointing at the West for these influence operations as well as the Usual Suspects. https://twitter.com/...
@mmasnick Mike Masnick on x
So many thoughts on this... though the main one is that it's silly for whoever put this together (whether it was the US gov't or not) to try to basically mimic Russian trolls in reverse. There's no need to do that. You don't have to be sneaky. https://t.co/cfb4QKrcup
@pkafka Peter Kafka on x
All the trolling and sock-puppeting and influencing we're always worried about China and Russia doing on our socials? Guess who's also doing it: https://cyber.fsi.stanford.edu/ ... https://twitter.com/...
@martinsfp @martinsfp on x
Quelle surprise. https://twitter.com/...
@mikeolson Mike Olson on x
.@Graphika_NYC and @stanfordio have published an interesting new report on pro-Western covert influence ops on social media. Twitter and Meta took these accounts down for ToS violation. Good detail on tactics and methods. https://cyber.fsi.stanford.edu/ ... Came my way via @noUps…
@stanfordio @stanfordio on x
We believe this report describes the most extensive case of covert pro-Western influence operations on social media to be analyzed by open-source researchers.
@acookiecrumbles @acookiecrumbles on x
Reading the full report, and it's like, wouldn't you be better off creating dancing Memoji TikToks? (Yes, entirely flippant, and tbf they're probably doing that anyway, and just haven't been discovered yet) https://twitter.com/...
@noupside Renee DiResta on x
@Graphika_NYC We spent several weeks assessing their content and behavior. Top-level findings 🧵 First, content: there were several multi-year regionally inflected campaigns which advanced narratives promoting the interests of the US & its allies, opposing countries including 🇷🇺, …
@justinhendrix Justin Hendrix on x
“This information operation likely originated in the United States and targeted a range of countries in the Middle East and Central Asia” https://twitter.com/...
@annemariebridy @annemariebridy on x
“The vast majority of posts and tweets we reviewed received no more than a handful of likes or retweets, and only 19% of the covert assets we identified had more than 1,000 followers.” https://twitter.com/...
@noupside Renee DiResta on x
@Graphika_NYC The behaviors - TTPs, or tactics, techniques, and procedures - are also detailed under each cluster in the report. We see some typical things, like the creation of unattributed “grey” media properties, which often reworked attributable USG content from overt govt-fu…
@katieharbath Katie Harbath on x
This is a different twist - a report on accounts taken down that were pushing pro-US messaging elsewhere in the world. https://twitter.com/...
@anishasircar Anisha on x
“The assets identified by Twitter and Meta created fake personas with GAN-generated faces, posed as independent media outlets, leveraged memes and short-form videos, attempted to start hashtag campaigns, and launched online petitions: all tactics observed in past operations.” htt…
@marklittlenews Mark Little on x
Important research ... “illustrates the much wider range of actors engaged in active operations to influence online audiences.” https://twitter.com/...
@acookiecrumbles @acookiecrumbles on x
The thing is, this is all so boring, so predictable, so completely lacking in imagination that I don't even know why folks bother. https://cyber.fsi.stanford.edu/ ... https://twitter.com/...
@thegrugq Thaddeus E. Grugq on x
“Gentlemen, we cannot allow there to be a shitposting gap!” https://twitter.com/...
@thegrugq Thaddeus E. Grugq on x
lol. Some contractor made huge money on a garbage network that achieved nothing. https://twitter.com/...
@alexstamos Alex Stamos on x
Our team at @stanfordio, along with our friends at @Graphika_NYC, analyzed a takedown from Meta and Twitter, finding a covert influence operation targeted at a broad swath of the Middle East and Central Asia on eight platforms. So far a normal week, except... this one is pro-US. …
@jason_kint Jason Kint on x
Super interesting to see a report holding accountable a pro-Western influence operation on social media. To be clear, though, I have no issue with inauthentic accounts spreading cute kittens so maybe that should get an exemption. But seriously, great work. Will read in full. http…
@stanfordcyber @stanfordcyber on x
New joint report out this morning! From @stanfordio's team & @Graphika_NYC, an analysis of suspended pro-U.S. influence operations on Twitter, Facebook, and Instagram, targeting a range of countries in the Middle East and Central Asia. https://cyber.fsi.stanford.edu/ ...
@stanfordio @stanfordio on x
Two new reports out today: An investigation with @Graphika_NYC into a pro-US influence operation https://cyber.fsi.stanford.edu/ ... An investigation into a pro-Kremlin influence operation, led by our rockstar graduate research assistant @ChrisSGiles https://cyber.fsi.stanford.ed…
@shelbygrossman Shelby Grossman on x
📢 New joint report out from my team & @Graphika_NYC. We analyzed suspended pro-U.S. influence operations on Twitter, Facebook, and Instagram. The target? People in the Middle East and Central Asia. https://cyber.fsi.stanford.edu/ ...

Chronicles