Twitter Expands Beta for Safety Mode Autoblocking Feature

Twitter is greatly expanding the size of the beta for its Safety Mode feature, which aims to limit “unwelcome interactions” in your feed.

Previously available only to a small feedback group, the feature will now be available to about 50% of users in the US, UK, Canada, Australia, Ireland and New Zealand, Twitter said Tuesday. The feature, which launched in limited beta in September, imposes a seven-day block on accounts that use what Twitter calls potentially harmful language, including insults or repeated replies and mentions.

The beta’s expansion is intended to help “collect more insights on how Safety Mode is working and explore ways to incorporate additional improvements,” Twitter said.

Twitter, which has about 217 million active daily users, has long been under pressure to do more to combat harassment. The social media platform has been criticized for being a “toxic place,” especially for women. A study by Amnesty International and Element AI in 2018 found that female journalists and politicians received “abusive” or “problematic” tweets every 30 seconds on average. 

Twitter also said it’s introducing a companion feature called Proactive Safety Mode prompts that it says will help reduce the burden on users in identifying harmful interactions.

“Since the initial rollout of the Safety Mode beta in September, we’ve learned that some people want help identifying unwelcome interactions,” Twitter said. “For this reason, our technology will now proactively identify potentially harmful or uninvited replies and prompt people in the beta to consider enabling Safety Mode.”