WASHINGTON (October 16, 2025) — A new study from researchers at the George Washington University finds that Twitter’s efforts to curb vaccine misinformation during the COVID-19 pandemic may have unintentionally amplified it.
The paper, “Explaining Twitter’s inability to effectively moderate content during the COVID-19 pandemic,” led by David Broniatowski, professor in the department of engineering management and systems engineering, examined how the platform’s design influenced the spread of misinformation.
The research team found that, despite Twitter’s removal of thousands of vaccine-skeptical accounts and posts, misinformation about vaccines continued to spread. In some cases, Twitter’s interventions may have even increased the engagement and virality of anti-vaccine content within groups of Twitter accounts that were already skeptical of vaccines. According to the researchers, that’s because Twitter was designed to facilitate the seamless spread of information online with few restrictions (i.e. any account can retweet content from other accounts). That architectural feature of the platform, combined with algorithms designed to recommend the most popular posts, meant that accounts could easily establish new pathways to prohibited content and thus circumvent Twitter’s moderation efforts.
“Our findings suggest that top-down attempts to control misinformation can backfire,” said Broniatowski. “It’s not just about bad actors or faulty algorithms. The architecture of the platform itself makes it difficult to contain the spread of information, incorrect or otherwise.”
The study highlights the need to rethink the overall design of social media platforms, rather than focus on reactive moderation, to balance communication and free expression.
The paper has been published in Scientific Reports. The John S. and James L. Knight Foundation funded the study through GW’s Institute for Data, Democracy & Politics (IDDP).
-GW-