The trolls are unleashed. What are social media companies doing about it?

Ogre Monster 3D Illustration Isolated On White
Lesley Becker/Globe Staff/Adobe Stock

Not very long ago, newspaper opinion pages encountered a worrisome trend known as AstroTurf: letters to the editor that appeared to be spontaneous grass-roots appeals but were actually part of orchestrated campaigns. The submissions weren’t written by individuals but were pre-packaged form letters drafted by advocacy groups, sent in over different signatures to several newspapers at once. So editors began checking in personally with each writer, confirming not just the veracity of a letter but its originality, and the astroturf problem subsided.

All of that seems so innocent now, as we learn of a vast network of computer-generated trolls, bots, and hackers operating out of hostile foreign nations, tampering with US elections and manipulating Congress by inciting “public” groundswells that are little more than astroturf on Olympic steroids.

According to the indictment handed up last week by special counsel Robert Mueller, a phony Russian Internet agency conspired to undermine the 2016 election by creating hundreds of fake social media accounts that spread disinformation and incited passions on hot-button issues such as immigration and race. Roughly 126 million Americans — almost as many as the 138 million who voted — were exposed to the Russian-sponsored posts on Facebook alone. The Trump administration has made much of the fact that Mueller hasn’t determined whether the interference was enough to swing the election results. But the cyberwar against American democracy is appalling regardless of who is or is not helped.


And the conspirators haven’t stopped at the election. According to analyses by the German Marshall Fund’s Alliance for Securing Democracy, the decision last month to release a classified House memo about Mueller’s investigation was prompted at least in part by a building chorus of public demand for it, when in fact #ReleaseTheMemo was being pushed by hundreds of anonymous, automated Twitter accounts created just for that purpose. The drive to publish the memo, written by House Intelligence Committee chairman Devin Nunes, also included real individuals and Republican party interests, but it’s clear that an army of fake accounts has the power not just to spread propaganda but also to influence behavior at the highest levels of government.

Get Truth and Consequences in your inbox:
Michael A. Cohen takes on the absurdities and hypocrisies of the current political moment.
Thank you for signing up! Sign up for more newsletters here

With disinformation, hoaxes, and automated trolls reaching critical mass on social media, Facebook’s CEO Mark Zuckerberg has pledged to make changes. He will hire more “security officers,” he announced recently, and his vaunted algorithm will begin to introduce more “trusted sources’’ to users’ news feeds. All good, except the platform will rely on the users’ own choices to identify what constitutes a trusted source. “We decided that having the community determine which sources are broadly trusted would be most objective,” Zuckerberg wrote in a post to users. Why crowdsourcing news would be any more reliable than voting for the best cat video is hard to fathom. As tech writer Jason Cross commented drily on Zuckerberg’s post: “People don’t even know not to eat Tide pods.” Asking them to decide what news is trustworthy, he said, is “irresponsible.”

Indeed, an underlying revelation in Mueller’s indictment is the shocking lack of critical thinking among Americans who are duped by the sham posts, often written with telltale poor English grammar. Short of enrolling every voter in a news literacy course before the 2018 election, the country needs Facebook, Twitter, YouTube, and the rest to assert some editorial control over their rampaging platforms.

Zuckerberg laments that deciding what news sources can be trusted is “a hard question we’ve struggled with,” but measuring a news item’s credibility isn’t really so difficult. Surely the geniuses at Facebook can apply a few simple tests: Are the claims in the article attributed? Can they be verified? What is their source? Are the sources named, and are they themselves independent and reliable?

If an algorithm can’t be designed to identify these key factors, human alternatives are available: call them editors.

Renée Loth’s column appears regularly in the Globe.