Twitter's New 'Safety Mode' Aims to Crack Down on 'Potentially Harmful Language'

Twitter released a statement today announcing the rollout of its new "Safety Mode," saying that a select feedback group would now have access to the feature that aims to "reduce disruptive interactions."

In the statement, Twitter described the new feature as being a tool that places a temporary 7-day block on accounts that use "potentially harmful language." The site describes such language as insults or hateful remarks but added that the block also applies to accounts that send "repetitive and uninvited replies or mentions."

When turned on, Safety Mode uses the website's "systems" to "assess the likelihood" of negative engagement within the interaction, honing in on the tweet's content and the author's relationship with the replier. Twitter said the technology takes relationships into account, meaning the tool will not block accounts the user follows or frequently interacts with.

"We want you to enjoy healthy conversations, so this test is one way we're limiting overwhelming and unwelcome interactions that can interrupt those conversations," the company said. "Our goal is to better protect the individual on the receiving end of Tweets by reducing the prevalence and visibility of harmful remarks."

Twitter has recently taken steps to punish high-profile figures who violate the company's code of conduct.

Earlier this month, the company placed Georgia Representative Marjorie Taylor Greene on a 7-day read-only ban after she violated the site's "misleading information" policy when she stated that COVID-19 vaccines were "failing" and that the FDA "should not approve" them. This followed a 12-hour ban in July when Greene made a false claim that COVID-19 does not affect young people.

The micro-blogging service has received flak from members of the right after it made the decision to place a permanent suspension on former President Donald Trump due to the risk of "further incitement of violence" following tweets he made the day of the January 6 Capitol insurrection.

It remains unclear how Safety Mode may impact accounts like Trump's and Greene's.

Users who employ the tool will receive notifications before each Safety Mode period ends providing a recap of the cycle. The company said blocks employed through the system can be undone anytime within the user's settings as the company works to detect its identification capabilities.

"We'll observe how Safety Mode is working and incorporate improvements and adjustments before bringing it to everyone on Twitter," the company wrote. "Stay tuned for more updates as we continue to build on our work to empower people with the tools they need to feel more comfortable participating in the public conversation."