Scientists Find That a Tiny Proportion of People Spread Almost All the Fake News, and They Turn Out to Be Exactly Who You’d Expect

"They are literally sitting at their computer pressing retweet."

Naomi Wolf Pipeline

A new study shows that a minuscule subset of "supersharers" spread the overwhelming majority of fake news on social media during the 2020 election cycle. The average supersharer profile, according to the research? Older, white, conservative, and incredibly online women in red states. Cue the gasp!

The study, published this week in the journal Science, analyzed data from the accounts of 660,000 verifiably real, US-based voters on the platform X, formerly known as Twitter.

Of these hundreds of thousands of American netizens, the researchers — a team comprising American and Israeli scientists — were able to determine that only about 2,000 users were responsible for sharing a whopping 80 percent of misinformation that spread during the 2020 election.

When the researchers examined the voter registration information attributed to the supersharers, a clear pattern emerged: they were disproportionately likely to be middle-aged-to-older white women with an average age of 58; they were also primarily Republican and lived in conservative states like Florida, Texas, and Arizona.

These users aren't just active, either. Per the journal's writeup, more than one in every 20 X users examined for the study were following these accounts, meaning that these supersharers are punching way above their weight in terms of reach. (The study builds on an earlier 2019 study from many of the same researchers, which found similar supersharer results when analyzing the 2016 election cycle.)

In a way, they could be likened to fake news influencers. Popular conspiracy websites like Infowars and Gateway Pundit publish fake news, which then makes its way to supersharers, who distribute it to the social media masses.

Final Boss

Though the researchers expected to find that the supersharers' many tweets were somehow automated, there was no clear timing pattern or other indicator suggesting that was the case. Instead, they found the opposite: that these folks are fully plugged into the misinformation IV, mainlining fake news and manually clicking retweet over, and over, and over again.

"That was a big surprise," study coauthor Briony Swire-Thompson, a psychologist at Northeastern University, told Science. "They are literally sitting at their computer pressing retweet."

To that end, it's also unlikely that the supersharing cohort in question was part of a coordinated disinformation effort. On the contrary, according to researchers, these users moreso represent a caustic breakdown in the way online fake news is created, shared, and consumed by a large faction of American voters. And though this study was about the 2020 election, as we all go kicking and screaming into November 2024, it's important to remember that not everyone exists in the same digital reality.

"It does not seem like supersharing is a one-off attempt to influence elections by tech-savvy individuals," Nir Grinberg, study co-author and computational social scientist at Israel's Ben-Gurion University of the Negev, told Science, "but rather a longer-term corrosive socio-technical process that contaminates the information ecosystem for some part of society."

More on fake news: Police Say AI-Generated Article about Local Murder Is "Entirely" Made Up