Source: Twitter
Twitter is introducing a new feature that temporarily autoblocks accounts that send harmful or abusive tweets to you. Safety Mode launched in beta with a "small feedback group" on iOS, Android, and Twitter.com with English-language settings enabled. Accounts are blocked for seven days, but you can undo any autoblocks that were wrongfully flagged.
If you have access to it, you can enable the feature in your Settings. Twitter's system will determine the likelihood of a negative engagement by looking at the tweet's content and the relationship between the account and the replier. The tech considers existing relationships you have, so you will not be automatically blocked if you follow and frequently interact with an account.
Twitter plans to roll this feature out to more testers "in the coming months." It joins other recently launched Twitter safety features like hiding replies, limiting who can reply to posts, and warning before sending a potentially harmful reply.