A group of NATO-affiliated researchers have found that Facebook, Google and Twitter are failing to detect and take down fake accounts and bot-driven engagement despite the industry’s efforts to clamp down on the issue.
The report comes several years after the 2016 presidential election, which drew new attention to the issue of inauthentic activity online as Russian trolls flooded U.S. networks to manipulate political discourse.
{mosads}Now, the NATO Strategic Communications Centre of Excellence has found that it’s still fairly easy for bad actors to buy fake accounts and “engagement” — meaning likes, retweets, views and more — on the world’s top social media platforms.
“We assess that Facebook, Instagram, Twitter and YouTube are still failing to adequately counter inauthentic behavior on their platforms,” the report reads.
The researchers spent 300 euros, or roughly $330, on batches of fake online engagement: 3,530 comments, 25,750 likes, 20,000 views and 5,100 followers. They identified 18,739 accounts manipulating social platforms.
Within four weeks, they found that 80 percent of the fake activity was still online. Even after they reported much of the inauthentic behavior to the platforms, they found that 95 percent of the reported content remained active several weeks later.
They concluded that some of the platforms are performing better than others in the fight against fake activity, however. They found that Twitter is “improving” its ability to block fake accounts when they’re being created, while YouTube is barely ever able to do so. They also found that Twitter and Facebook are getting better at identifying and removing manipulated accounts.
The researchers compared Facebook to a “fortress with formidable defenses facing the outside world, but qualified actors are still able to scale [its] walls,” while it found that Instagram — which Facebook owns — has “significant challenges with countering abuse.”
Facebook in a statement on Friday called inauthentic behavior a “challenge facing the entire industry.”
“We’re making massive investments to find and remove fake accounts and engagement every day,” a Facebook spokesperson said.
And Twitter’s head of site integrity, Yael Roth, highlighted the researchers’ comparably favorable conclusions about Twitter.
“Setting up fake accounts, paying for engagement, and deliberately manipulating our service are all prohibited behaviors,” Roth said. “We’re encouraged the report recognizes Twitter’s proactive work to stifle inauthentic activity at the source.”