Take it from a NATO group reporting from near the Russian border: U.S. social media remains a breeding ground for fraud, fake news and manipulation

Facebook, YouTube, Twitter and Instagram are failing to make good on efforts to eliminate fake news and clicks, according to the NATO Strategic Communication Centre of Excellence in Riga, Latvia.

The group paid 16 “manipulation service providers” — mainly in Russia — for fake clicks, likes and followers for 105 different posts on Facebook, Instagram, Twitter and YouTube and found after four weeks that 80% of the inauthentic engagements were still online, according to a report released today.

The platforms also failed to respond properly to user feedback. Three weeks after a sample of the fake accounts were reported, more than 95% were still active online, said the NATO Strategic Communications Center of Excellence. The independent organization is an advisor to the North Atlantic Treaty Organization.

“We see no sign that it is becoming more expensive or more difficult to conduct widespread social media manipulation,” the group wrote. “Facebook, Instagram, Twitter and Youtube are still failing to adequately counter inauthentic behavior on their platforms.”

The report adds fuel to criticism that the social media giants aren’t doing enough to combat online disinformation and manipulation. While the companies did make changes to their operations after Russia interfered in the U.S. 2016 presidential election, they are under pressure to make sure a repeat doesn’t happen when voters go to the polls in the future.

Social media manipulation could have an impact on European elections this year and the 2020 U.S. presidential election, Samantha Bradshaw, a researcher at Oxford University’s Oxford Internet Institute, told the New York Times. “Fake engagement — whether generated by automated or real accounts — can skew the perceived popularity of a candidate or issue,” she said.

“Facebook, Instagram, Twitter and Youtube are still failing to adequately counter inauthentic behavior on their platforms.”

To blunt the criticism, Facebook in October introduced a news section that will include stories from mainstream publications deemed to be reliable. It also instituted new policies — including the clear labeling of fake news and state-controlled media — to help users interpret what they read

The tech giants also have been turning to startups and working on in-house solutions. Twitter  acquired U.K.-based Fabula AI to help detect network manipulation, and Facebook launched new features to limit misinformation in April. Google has ranking algorithms and new user features to fight fake news, and last year pledged $300 million toward a new journalism initiative.

  • The NATO-affiliated group said the four companies varied widely in identifying inauthentic accounts. Facebook suspended 80% of the fakes, Twitter 66% and Instagram 50%. YouTube didn’t suspend any.  
  • Fact-checking efforts have shown to be ineffective because they rarely reach the people most exposed to the falsehoods, resulting in a “preaching to the choir” effect that has limited impact, according to a study from Alto Data Analytics. About 95% of fact-checking — the follow-up reports that either correct or confirm a story — isn’t read, the study said.