Twitter is promising to crack down on abusive accounts on its service, moving to limit accounts engaging in abusive behavior, and bringing in new filters to allow users to control what they see on the social network.
According to a new report from Recode, the company said it is using algorithms to detect abusive behavior, giving users new filtering tools and being more transparent in the reporting process.
The company confirmed that it’s taking punitive actions against harassers, which were spotted in the wild earlier in February. When Twitter’s algorithms discover an abusive tweeter, the company will temporarily reduce that person’s reach. Anyone who doesn’t follow that user won’t be able to see their tweets for a set cool down period.
“We aim to only act on accounts when we’re confident…[are]..abusive. Since these tools are new we will sometimes make mistakes, but know that we are actively working to improve and iterate on them everyday,” the company said in a blog post.
Twitter declined to give more details on what exact behaviors might trigger the algorithm and temporarily ban an account’s reach, because they don’t want people to find workarounds. A spokeswoman confirmed that it could potentially reflect harassment trends beyond obvious trigger words.
Filtering options on the notifications page are also improving — you will be able remove accounts that don’t have a profile photo, or those that have unverified email addresses or phone numbers. Twitter users can also now mute words or conversations right from their timeline for a set amount of time.
The new filtering could help get rid of the bad egg problem in people’s mentions. Harassers and spammers are frequently called “Twitter eggs” because the of the default white egg avatar. “Twitter egg” is now used as an insult referencing this phenomenon.