Twitter may have a more relaxed stance on bans under Elon Musk, but it’s still willing to flag content that violates its rules. The social network will be label tweets that you believe are in violation of your hateful conduct policy. You’ll see a notice that Twitter is limiting the “visibility” of the problematic post with an opportunity to learn more.
The labels will cover more types of policy violations in the coming months. Users can initially provide “feedback” on a tag if they feel it was in error, but will have to wait until some time in the future to formally appeal the decisions. The company emphasizes that tags only apply to individual tweets, not an entire account.
We’re adding more transparency to the enforcement actions we take on Tweets. As a first step, you’ll soon begin to see labels on some Tweets identified as potentially violating our hateful conduct rules, letting you know we’ve limited their visibility. 🧵…
— Twitter Safety (@TwitterSafety) April 17, 2023
Twitter sees this as a reflection of its Musk-era “free speech, not free reach” philosophy, where it more often eschews outright bans in favor of limiting exposure to content. In theory, this prevents offensive speech from spreading without completely silencing users on the platform. It also promises greater transparency to address concerns of “shadow bans” that limit scope without someone’s knowledge. Musk has lifted the bans on several high-profile accounts, including that of former President Trump, though he was willing to temporarily suspend Ye despite lifting a permanent ban in November.
The move might not please some. Advocacy groups have criticized Twitter for not doing enough to protect frequent targets of hate, such as the LGBTQ community. While violators can still face bans if they are “bad actors” or break the law, they are more likely to stay on the service. Nor is it sure that this will satisfy advertisers who are abandoning Twitter for fear that their ads could run alongside objectionable material. And while it will address concerns about potential censorship, it won’t necessarily please those who see labels as tarnishing a post’s value.