Truth-Tellers Beat Fake News on Twitter, USC Study Finds

When COVID-19 conspiracies flooded Twitter, a powerful counterforce emerged: everyday users armed with facts fighting back harder than the misinformation spreaders themselves.

USC researchers just proved what many suspected but couldn't quantify. People who debunked false health claims during the pandemic had significantly more followers and longer-established accounts than those pushing fake cures and conspiracy theories.

The team, led by Professor Emilio Ferrara and PhD student Eun Cheol Choi at USC's HUMANS Lab, trained artificial intelligence to analyze thousands of tweets from 2020 to 2021. They weren't just looking for misinformation itself but studying the people behind both sides of the information war.

What they found surprised them. Users countering false claims expressed more anger, sadness and disgust in their posts than misinformation spreaders did. Meanwhile, posts spreading fake news tended to be longer and expressed more surprise, while fact-checking responses stayed short and direct.

This flips the common assumption that misinformation is the most emotionally charged content on social media. Truth-tellers were actually angrier, and they had good reason to be.

The pandemic created what researchers called a "wild time" for public health misinformation. False claims about garlic water curing COVID, dog parasite drugs as treatments, and vaccines making people "Bluetooth connectable" spread rapidly across the platform.

But for every bogus claim, established users with real credibility pushed back. Their larger followings meant accurate information reached more people than the lies did.

The Bright Side

The research reveals something encouraging about social media's structure. Truth doesn't just have accuracy on its side; it has reach. People who've built trust over time naturally have bigger platforms to amplify correct information.

However, the study also uncovered a troubling technical problem. Content moderation systems on platforms like Twitter often automatically flag or downrank posts containing high levels of negative emotion. That means the angry, passionate fact-checkers fighting dangerous health misinformation might be getting suppressed by the very algorithms designed to protect users.

The researchers hope their findings will help social media companies redesign moderation systems to recognize the difference between harmful negativity and justified anger from people protecting public health. Their paper was accepted to ACM Hypertext 2026, giving these insights a platform in the academic community.

When fear and uncertainty fueled tens of thousands of unique false claims worldwide, ordinary people chose to stand up with facts, and they brought their communities with them.