Scientists Reveal How Credible News Sources Are Exploited to Spread Misinformation

Mainstream Media Articles Weaponized to Spread Misinformation, Study Finds

A groundbreaking study published in Nature Human Behaviour reveals that online misinformation extends beyond fabricated stories from dubious websites to include factual reports from mainstream media outlets that are strategically repurposed to support false claims.

Researchers from Northeastern University, Duke University, Harvard, and the University of Maryland discovered that social media users who frequently share fake news also selectively share specific articles from reputable news sources—particularly those containing narratives commonly found in misinformation campaigns.

“Information doesn’t exist independently from how people use it,” explained lead researcher Pranav Goel from Northeastern University’s Network Science Institute. “Users don’t merely share individual facts. They share stories that advance their broader interests and political worldviews.”

The research team analyzed Twitter (now X) activity spanning from May 2018 to November 2021, focusing on accounts matched to a U.S. voter file to ensure they represented real individuals. They first identified users who regularly shared content from unreliable domains—websites classified as “fake news” based on NewsGuard ratings—and then observed what other content these same users shared.

This methodology allowed researchers to identify mainstream articles frequently “co-shared” with fake news, assigning them a “co-sharing score” based on how often they appeared alongside misinformation. The team then created a control group of stories from the same reputable outlets that weren’t popular among the misinformation-sharing crowd.

Using computational tools designed to extract narrative structures from text, researchers compiled a database of patterns prevalent in known fake news articles. When comparing mainstream articles popular with misinformation spreaders against the control group, a distinct pattern emerged: these co-shared mainstream articles were significantly more likely to contain narrative structures common in fake news—regardless of the outlet’s political leaning.

One prominent example involved a Washington Post headline stating “Vaccinated people now make up a majority of covid deaths.” While factually accurate at publication, the headline lacked crucial context explaining that with most of the population vaccinated, raw numbers would naturally skew toward vaccinated individuals despite the vaccine’s effectiveness. Misinformation spreaders seized on this headline to promote false narratives about vaccine inefficacy while citing a credible source.

Another strategy identified was the repurposing of archival news. A 2012 New York Times article headlined “As More Vote by Mail, Faulty Ballots Could Impact Elections” was widely circulated in 2020 to support unfounded claims of widespread mail-in ballot fraud during the presidential election.

“Clickbait-style headlines that oversimplify complex issues prove particularly vulnerable to weaponization,” noted co-author Jon Green from Duke University. “The body text might contain necessary nuance, but the headline alone serves the misleading narrative.”

The study found that audiences for these co-shared mainstream articles potentially reach much further than traditional fake news sites. Users sharing these repurposed mainstream articles averaged nearly twice as many followers as those exclusively sharing content from fake news domains, suggesting these articles function as bridges carrying misleading narratives into broader public discourse.

While acknowledging limitations—such as focusing on article text rather than the accompanying social media posts—the researchers performed manual checks confirming that users predominantly shared these articles to endorse their content rather than criticize it.

The findings raise important implications for journalism. The research suggests fact-checking article content alone isn’t sufficient; editors and reporters must consider how headlines and stories might be weaponized to support broader misleading narratives, even when the reporting itself is factually accurate.

“In today’s networked information environment, strictly true information can still result in a misinformed public if it’s strategically selected and presented without proper context,” said co-author David Lazer, who holds positions at both Northeastern and Harvard.

The study highlights a sophisticated evolution in misinformation tactics, where bad actors leverage credible sources to enhance the perceived legitimacy of false claims—a strategy potentially more effective than relying solely on obviously fabricated content from dubious websites.

As social media platforms and news organizations continue wrestling with the spread of misinformation, this research underscores the complexity of the challenge, suggesting that source credibility alone cannot serve as a reliable proxy for determining whether content contributes to public misunderstanding.