Facebook stopped just short of identifying Russia in its report, but also emphasized that it is “not in a position to make definitive attribution to the actors sponsoring this activity,” which it said represented only a small portion of the activity Facebook tracks on its platform.
Facebook acknowledged more broadly that it has a problem with what it calls “information operations,” government-run efforts to use Facebook to manipulate public opinion, distort domestic or foreign political sentiment, and influence the outcome of elections.
In many cases, such information operations are aimed at gaming Facebook’s algorithm, using tactics like the mass creation of fake accounts and the creation of groups populated by those accounts.
Fake accounts and misleading groups then carry out coordinated campaigns designed to amplify a message: They do this by simultaneously sharing and liking the same Facebook posts en masse, rapidly posting the same information across multiple groups at once, and spreading sensationalistic or heavily biased headlines as a way to distort facts and fit a narrative. These groups can be hard to detect because they often post legitimate and unrelated content, as well, “ostensibly to deflect from their real purpose,” Facebook says.
Facebook says it detected several “subtle and insidious” kinds of coordinated attempts to harm the reputation of “specific political targets” during the 2016 campaign, describing “malicious actors leveraging conventional and social media to share information stolen from other sources, such as email accounts, with the intent of harming the reputation of specific political targets.” Spokespeople for Facebook declined to answer my questions about which political target and stolen email information the Facebook report was referring to, but the intelligence report that Facebook links to describes “high confidence” in the assessment that Russian intelligence relayed hacked emails between senior Democratic officials to WikiLeaks to undermine Clinton. “Moscow most likely chose WikiLeaks because of its self-proclaimed reputation for authenticity,” the DNI report says. “Disclosures through WikiLeaks did not contain any evident forgeries.”
Without mentioning Russia or WikiLeaks, however, Facebook describes actors that create fake personas on Facebook as a way to direct people to the stolen data. “From there, organic proliferation of the messaging and data through authentic peer groups and networks was inevitable,” Facebook wrote.
At the same time, malicious actors would use fake Facebook accounts to “push narratives and themes that reinforced or expanded on some of the topics exposed from stolen data,” including attempts to seed stories with journalists and other third parties. Facebook didn’t describe the stories themselves. “We detect this activity by analyzing the inauthenticity of the account and its behaviors, and not the content the accounts are publishing,” its report said.