NEW YORK - New research shows that Twitter fails to act on 99% of hate posted by Twitter Blue subscribers, suggesting that the platform is allowing them to break its rules with impunity and is even algorithmically boosting their toxic tweets.
Researchers collected tweets promoting hate from 100 Twitter Blue subscribers. The tweets were reported to the platform using Twitter’s own tools for flagging hateful conduct.
Four days after reporting the tweets, researchers found that Twitter had failed to act on 99% of the posts and 100% of the accounts remained active. In the one instance that Twitter removed a hateful tweet, the account from which it was tweeted remains active.
Twitter failed to act on tweets containing racist, homophobic, neo-Nazi, antisemitic or conspiracy content posted in the last month, including tweets claiming that:
- “The black culture has done more damage [than] the klan ever did”
- “The Jewish Mafia wants to replace us all with brown people”
- “Trannies are pedophiles”
- “Diversity is a codeword for White Genocide”
- “Hitler was right”, accompanied by a montage of the former dictator
- Black people belong “locked in cages at the zoo”
- LGBTQ+ rights activists need “IRON IN THEIR DIET. Preferably from a #AFiringSquad”
Twitter failed to act despite the fact that the tweets surfaced by the Center clearly violate the platform’s policies against hateful conduct, which bans abuse on the basis of race, ethnicity, national origin, sexual orientation and gender identity.
Twitter’s lack of responsibility for removing hate speech is unsurprising. Previous studies from CCDH have looked into anti-Jewish and anti-Muslim hate on Twitter, finding that the platform had failed to act on 89% and 97% of posts, respectively.
The research not only raises concerns that Twitter Blue users can get away with tweeting hateful content, but that they also benefit from algorithmic amplification as a result of their paid subscriptions.
Twitter says on its website that Twitter Blue users are given “prioritized rankings in conversations and search”, stating that “people on Twitter will see a slight preference for replies from Blue verified accounts over other replies.”
Elon Musk has recently tweeted to confirm this, announcing on 25 April that “verified accounts are now prioritized”. Anyone with $8 dollars a month and a verified phone number can sign up for the Blue tick, threatening the credibility of accounts and accuracy of content.
Researchers found that tweets from Twitter Blue users in the sample often appeared to have been given priority in threads. In one example of a thread containing almost 100 tweets from non-verified users, the top-ranked reply was from a Twitter Blue user calling for violence against migrants.