In its early days, Twitter referred to itself as the “free speech wing of the free speech party.” The quote might not have been totally serious, but for years it rang true: its pseudonymous accounts and permissive rules gave the platform an anything-goes sensibility. A few years and high-profile harassment campaigns later, however, the company’s position on speech has noticeably shifted. Now, Twitter focuses on the “health” of conversations and increasingly considers the costs of unfettered speech, not just its benefits. The latest step of that evolution came on Tuesday night, when the company announced that it was taking steps to limit the influence of QAnon on its platform.

Part conspiracy theory, part cult, QAnon is a sprawling online community loosely organized around the belief that Donald Trump is waging a secret war against an elite cabal that engages in devil worship and pedophilia. Recently it has been linked to spreading dangerous coronavirus misinformation and coordinated harassment of individuals spuriously linked to child sex rings. Its adherents have even been motivated to enact real-world violence, including at least one alleged murder. And for years, it has used Twitter and other social media platforms to spread. Reddit, which was instrumental to the movement early on, banned the biggest communities over violent threats in 2018.

“This is an important marker that Twitter is recognizing how it is being manipulated,” said Joan Donovan, research director at Harvard’s Shorenstein Center on Media, Politics and Public Policy. Twitter, she explained, is crucial link between online discourse and real-world activity, because it attracts the attention of media and political elites who help amplify movements, often unintentionally. Twitter’s announcement, she said, suggests that the company has gotten more serious about the real-world consequences of activity on its platform. QAnon “grew out of a belief that Trump is a Messiah-like figure who is working very closely with the quote-unquote ‘Deep State’ to take down an elite ring of pedophiles,” she said. “If you believe that, and you are a passionate person, that might lead you to do all kinds of different things in anticipation of there being global unrest.”

The WIRED Guide to Online Conspiracy Theories

Everything you need to know about George Soros, Pizzagate, and the Berenstain Bears.

Twitter’s announcement has two main prongs. The first is account termination: the company says it will permanently ban anyone who tweets QAnon content and violates rules around coordinated harassment, running multiple accounts, or trying to evade previous suspensions. The second prong of the policy is about amplification. Twitter says it will stop recommending QAnon accounts, suppress the topic in search results, and block QAnon-related URLs from being shared.

The company told NBC News that it has already removed 7,000 accounts, and as of Wednesday morning, some prominent QAnon influencers had been taken down. As always, it remains to be seen how well the platform enforces the policy. Its newfound willingness to flag false or dangerous tweets from public officials, for example, including Trump, doesn’t seem to have been applied totally uniformly. And while Twitter has worked on automating moderation with AI, it still relies heavily on user reporting. The company has frequently been criticized for not enforcing existing policies consistently, and for not being responsive to targets of harassment campaigns, particularly if they aren’t high-profile public figures. The new policy also refers to “accounts associated with QAnon” without explicitly defining what that means.

Even if flawlessly implemented, Twitter’s new policy won’t make QAnon disappear. But Donovan predicted it would dramatically reduce the group’s ability to spread. “These movements mutate very, very quickly, and they’re not going to stay off Twitter,” she said. “But they’re going to have a hard time growing if they don’t solidify around a few keywords so that they can find one another again.”