Continuing to address hate speech and online abuse by introducing tools that offer people more control over what they see on Twitter, the company has today announced that it will now start blocking the creation of new abusive accounts, while also bringing forward safer search results, and collapsing potentially abusive or low-quality Tweets. The new features will be rolling out over the next few days.
Twitter says that its new safety features will prevent banned users from rejoining the service via new accounts. According to Recode, Twitter will use a combination of human reviewers and machine learning technology to keep suspended users from resurfacing. The new “safe search” feature will remove tweets with inappropriate words, phrases or images from search results. Users can toggle this off and on in settings.
Twitter will start hiding inappropriate responses to tweets so they don’t appear in user conversations. Tweets containing words, phrases or images deemed inappropriate by the company will be hidden from conversation threads, and users will need to click a new “show less relevant replies” button in order to see them. Twitter will use machine learning technology (a.k.a. algorithms) to automatically hide certain responses.
The company also said that it is working on giving users a more direct way to report hateful conduct.