While there’s plenty of blame to go around for the spread of Russian propaganda during the 2016 presidential election, Twitter and Facebook have bore the brunt of public criticism over the past two years. But according to a new report prepared for the Senate Select Committee on Intelligence (SSCI) released on Monday, the Russian state-supported Internet Research Agency (IRA) also used other platforms to great effect. While it’s difficult to quantify the impact of the attempted manipulation, it shows a clear pattern of obfuscation by the tech companies. It also shows that more insulated platforms like Instagram have leaned on their structural differences in hopes of avoiding a public reckoning.
Drawing upon a dataset of social media posts and metadata delivered to the SSCI by Facebook, Twitter, and Alphabet (Google’s parent company), researchers from cybersecurity company New Knowledge, Columbia University and Canfield Research found that engagement with IRA-created content on Instagram far outpaced engagement on Facebook. “Instagram was a significant front in the IRA’s influence operation, something that Facebook executives appear to have avoided mentioning in Congressional testimony,” the report says.
It’s not surprising that Facebook officials would try to downplay the scale of IRA activity on Instagram, which was purchased by Facebook in 2012. Shortly after the 2016 election, Mark Zuckerberg decried criticisms that fake news on Facebook could have influenced voters as “pretty crazy.” A year later, the company told Congress that around 126 million people were exposed to Russian propaganda on Facebook.
During the same testimony, a Facebook spokesperson said that only an estimated 20 million people had seen IRA content on Instagram. But they neglected an important point: the IRA actually created about 55,000 more posts on Instagram than on Facebook, and the total engagement (sum of likes, comments, shares) with Instagram posts came to about 187 million, over a hundred million more than on Facebook. Twelve of the IRA’s Instagram accounts reached over 100,000 followers, and about 40 percent of the accounts garnered more than 10,000. Yet until now, Instagram hasn’t really been included in the post-mortems of social media’s role in the election.
Obviously, it’s difficult to ascertain the scale of IRA’s operations and respond appropriately when the companies aren’t forthcoming about the activity on their platforms. But there are also structural differences between Facebook and Instagram that make it inherently fuzzier to notice the spread of propaganda and quantify its impact.
Virality was a key factor in what made IRA activity so discernable on Facebook. Because people can share content, posts traverse the platform quickly, crossing social groups and fomenting conflict. Instagram, on the other hand, doesn’t have a native sharing function and interactions tend to be much more siloed. This is a key difference that likely gave the IRA a wider (but not necessarily more impactful) reach on Facebook than on Instagram, and what would necessitate a higher number of posts in general. (It’s also possible that more people than estimated saw IRA content on Instagram through “regrams”, or repostings of original posts, but these are not traceable through any part of the service.)
While the lack of a sharing mechanism possibly made it more difficult to push content out to a large number of people, it also obscured the volume of propaganda on the platform. If Facebook is a public square, where information is visibly shared and different communities mingle, Instagram is an room full of cubicles, where interaction is constrained and it’s difficult to draw connections beyond individual profiles. And what better place to seed propaganda than one where others can’t observe what is happening?
Instagram’s role in pushing IRA causes stayed buried this long because of Facebook’s standard response to PR crises: say as little as possible to avoid bad press. In this strategy, the company isn’t alone. Twitter and Alphabet have been similarly cagey when it comes to their management of Russian propaganda, as the researchers note:
None of the platforms (Twitter, Facebook, and Alphabet) appears to have turned over complete sets of related data to SSCI. Some of what was turned over was in PDF form; other data sets contained extensive duplicates. Each lacked core components that would have provided a fuller and more actionable picture.
Regrettably, it appears that the platforms may have misrepresented or evaded in some of their statements to Congress; one platform claimed that no specific groups were targeted (this is only true if speaking strictly of ads), while another dissembled about whether or not the Internet Research Agency created content to discourage voting (it did). It is unclear whether these answers were the result of faulty or lacking analysis, or a more deliberate evasion.
The findings revealed in Monday’s report show that Instagram, and to a lesser extent, YouTube, is worthy of the same scrutiny afforded to Facebook and Twitter. It’s probably time the platform accepts some of the blame, too.