A study published in the Proceedings of the National Academy of Scientists (PNAS) has found what we all suspected, that X’s algorithm prioritizes ragebait in order to juice engagement. The study also found that this content was served up more to people who identify as Democrats.
This is not a shock to anyone who has looked at X in the last couple of years. Currently, my feed is an endless cavalcade of right-leaning troll accounts popping off about how evil it is that a shop at the Mall of America sells hijabs (or something.)
One of the paper’s authors, Ziv Epstein, told 404 Media that they know this is “not the most surprising headline ever” but that they wanted to get to the why of it all. “X’s feed algorithm, like a lot of these social media algorithms, is optimized for engagement [but] it turns out that not all types of engagement are considered equally,” the Stanford researcher said.
The study involved following 715 American X users who installed a browser extension that collected their For You and Following feeds. Participants were asked to fill out a “values inventory” using something called the Schwartz Theory of Basic Values, which breaks down someone’s belief system into 19 points that correspond to qualities like “tolerance” and “dominance.” Volunteers also reported their political alignments.
Next, researchers tracked how these users engaged with posts and how those posts reflected their self-reported values. The study found that the algorithm was “more likely to amplify” content that didn’t align with those self-reported values. In other words, it didn’t take long until they started seeing stuff that angered them.
If a post infuriates someone, they are more likely to engage with it which, in turn, sends more outlandish posts their way. Epstein notes that replying (to a baseless or offensive post, for instance) is weighted much more heavily than liking, despite replies accounting for under seven percent of interactions on the platform.
“Replying is only a fraction of engagement, but there does seem to be some evidence that these algorithms are prioritizing and learning more from this kind of rarer form of engagement,” he said. “So it’s this feedback loop of outrage baiting. The algorithm learns that you get outraged and then continues to serve more content in that direction.”
However, the study also found that this ragebaiting occurred more for the users who identified themselves as Democrats. The researchers don’t yet have a reason for this, suggesting that it could be accounted for by the fact that there’s simply more right wing content on X or that Democrats tend to engage more with posts they disagree with.
Epstein said the study was designed to get people talking about social media apps and how they operate. “There are these social media algorithms that have enormous amounts of power in our lives, they shape the information that we consume and we have very little transparency into how they operate and what their implications are,” he said. “I think that has kind of important implications for civil society and just fighting some of the techno-feudalistic tendencies of platforms to control these algorithms.”
This is no longer true:
The largest contributor of seeing ragebait was the reply predictor and we were aware that angry replies were causing people to see more of that content.
So last month, we gave the reply predictor a 15x boost if it’s a friend’s post—and it reduced… https://t.co/dl8qVld4K2
— Nikita Bier (@nikitabier) August 18, 2026
X hasn’t formally responded to this research. However, the company’s former head of product Nikita Bier says the algorithm has been adjusted since the study took place and that ragebait content has been reduced “by order [sic] of magnitude.” My feed suggests otherwise, but at least the platform is aware of the issue and working on it.
