Instagram is rolling out a new tool to automatically filter out offending DMs


Instagram today announced a new tool that allows users to automatically filter direct message requests with offensive words, phrases and emojis. The tool is aimed at celebrities and public figures who receive a large number of unwanted malicious DMs.

The update builds on Instagram’s work to combat hate speech on the platform. In February, the company said it would begin disabling the accounts of users who have sent multiple harassing messages. In 2018, the company expanded its Offensive Comments filter to automatically block comments that attack a person’s appearance or character.

The message request filter can be turned on or off in a new section of the app called ‘hidden words’. When enabled, offensive messages will be pushed to a separate folder. DMs in this folder are hidden so users can browse messages without having to read what they say. When a user touches a message, he can read, delete or report it.

The company says it has partnered with “leading anti-discrimination and bullying organizations” to create a preset list of offensive expressions. Users can also customize their own list based on what they don’t want to see.

Instagram said the new feature will be rolling out to “different countries” in the coming weeks, although it didn’t specify which countries they are. The company plans to expand into more countries in the coming months.

As part of today’s announcement, Instagram also said it is rolling out a tool that allows people to preemptively block new accounts for harassment. Now when a user blocks someone in the app, they have the option to block new accounts that person creates. The company declined to specify how it does this, just saying it uses a “variety of signals” to detect new accounts from those same users. The feature will be released worldwide in the coming weeks.