A new study has found that falsehoods spread faster on Twitter than real news —and that the problem can’t be blamed on bots.
Fake news tends to have the real news beat in one area: novelty. That’s why Twitter users can’t help but share half-truths and unchecked rumors over the platform, say a trio of MIT researchers who conducted the study.
The findings were presented in the latest issue of Science and underscore an ongoing critique facing Twitter: that the social media service is potentially doing more harm than good by becoming a hotbed for fake news and propaganda.
The study from the MIT researchers attempted to delve deeper into the issue by examining how 126,000 stories —both real and false— were tweeted by over 3 million people.
“We found that falsehood diffused significantly farther, faster, deeper, and more broadly than the truth in all categories of information,” the authors wrote.
For instance, the true news-related tweets rarely reached over 1,000 people. In contrast, the top 1 percent of tweets carrying false news routinely spread from between 1000 to 100,000 people.
Fake political news was especially viral. These stories ended up reaching over 20,000 people faster than what other false news categories could do.
“When we estimated a model of the likelihood of retweeting, we found that falsehoods were 70 percent more likely to be retweeted than the truth,” the authors added.
Did follower counts have anything to do with the results? Not really. The people who shared the false news stories generally had fewer followers than people who tweeted fact-based stories, the study said. Fake news sharers were also less active on Twitter, usually weren’t verified, and had been on the platform for not as long.
To take in account of bots, the MIT team also used a computer algorithm to strip out suspected fake accounts from their research data. But even without the algorithm, their overall findings remained the same.
“We conclude that human behavior contributes more to the differential spread of falsity and truth than automated robots do,” they said. This suggests to fight misinformation, Twitter will have to do more than simply crack down on bots, the authors added.
The study doesn’t exactly paint the social media service in the best light. But Twitter actually supported the researchers by providing funding and access to the archived tweets.
The company’s CEO Jack Dorsey tweeted his support for the study, following his own public efforts to clean up the service. “Important findings within that [research] will help improve our work,” he said.
Among those projects is an effort to measure the health of conservations across the platform. The company has also been notifying users who interacted with Russian propaganda on Twitter during the 2016 presidential election.
This article originally appeared on PCMag.com.