In a blog post, Twitter's VP of engineering Ed Ho claims abuse on the platform is "significantly less" than it was six months ago. Ho attributes this recovery to Twitter's improved response rate. The company is now taking action on ten times the number of abusive accounts every day compared to the same time last year -- when it was in snooze mode.
By far Twitter's biggest recent product update on harassment has been its quality filter. Ho claims the feature has led to a drop in unwanted interactions. As evidence, he cites that blocks after @mentions from people you don't follow are down 40 percent.
Ho also touts Twitter's new systems for targeting harassment. In February, the platform started placing a temporary timeout on abusive accounts. These too are doing the job, with profiles placed under the ban generating 25 percent fewer abuse reports. Additionally, the majority of these accounts (65 percent) are only restricted once, which indicates the punishment may actually have an impact on future behavior.
The Center for Democracy and Technology's Emma Llansó spoke to Engadget about the platform's reforms. Her non-profit is part of Twitter's Trust and Safety Council, an advisory group made up of safety advocates, researchers, and academics. "We're particularly interested in the work Twitter has been doing to help users better understand the policies that govern the site," said Llansó. In reference to the timeouts, she added: "Letting someone know they've crossed a line helps them avoid the same behavior in the future, without shutting them down or silencing them completely."
Twitter's failings make it an easy punching bag, and there are signs that its abuse problem still exists. But, its relatively small audience loves it, despite the harmful aspects of its culture. If the company can carry on policing its free expression train correctly, its toxicity could continue to subside.