Twitter Inc. has been struggling for years to better protect its users from online abuse, and today the company announced a suite of new tools aimed at doing just that.

Ed Ho, vice president of engineering at Twitter, explained in a blog post that the company believes that abuse can prevent users from speaking their minds, which is obviously a bad thing for a social network that relies on people blurting out whatever comes to mind.

“We stand for freedom of expression and people being able to see all sides of any topic,” Ho said. “That’s put in jeopardy when abuse and harassment stifle and silence those voices. We won’t tolerate it and we’re launching new efforts to stop it.”

Chief Executive Jack Dorsey (pictured) has been under pressure from users to reduce rampant harassment on Twitter. When he rejoined as CEO in 2015, he named that job as a key priority. But so far, despite the banning of some high-profile Twitter users such as Milo Yiannopoulos, the controversial Breitbart News writer, Twitter’s moves haven’t yet solved the problem.

One way that Twitter hopes to prevent abuse is to make it harder for trolls to create new accounts after getting banned the first time for posting abusive content or harassing other users. Ho did not explain how exactly this new feature works, but he said that the company has been working to better identify new accounts created by users that have already been suspended. Twitter is also working on shutting down accounts that are created solely for the purpose of harassing others, but again, Ho did not explain how exactly the social media platform is going to do this.

Aside from tackling trolls directly, Twitter is also introducing new features that will make it easy for users to hide or avoid abusive content on the sight. This includes the creation of a new safe search feature, which automatically omits results that include “potentially sensitive content and Tweets” from abusive accounts that have been blocked or muted. Ho notes that safe search is an opt-out feature, so users who do not want their searches to be filtered can disable safe search in their account settings.

Finally, Twitter is also taking a Reddit-like approach to replies by automatically collapsing “low-quality replies.” Much like Reddit comments that have received a significant number of down votes, Twitter will hide reply threads that have been identified as abusive. As with the safe search feature, users can still opt to view potentially abusive reply threads if they want to, but they will be hidden by default.

While fighting abuse and harassment is certainly noble, Twitter’s walking a tightrope. Its anti-abuse tactics have been criticized in the past for stifling the same freedom that the company says it values.

The company also seemingly relies mostly on its users to report abuse. That leaves the system vulnerable to brigading, by which members of a community try to silence specific users by mass-reporting them.

Photo: Esten Hurtle (@esten) for Twitter