Instagram reveals tool for filtering abusive DMs

21 Apr 2021

Image: Instagram

The filter detects abusive and offensive words and phrases in unwanted messages from people that a user does not follow.

Facebook-owned Instagram is rolling out a new feature to curtail abuse on its platform.

The tool allows users to filter out abusive messages, sent through DMs, from other users that they do not follow.

When a user flicks the feature on, it automatically scans DM requests for common offensive words, phrases and emojis, or words and phrases that can be selected by the user. Instagram has created a bank of words and phrases for blocking after consulting with anti-discrimination and anti-bullying organisations.

Any messages detected as abusive will be moved to a separate folder.

“If you choose to open the folder, the message text will be covered so you’re not confronted with offensive language, unless you tap to uncover it. You then have the option to accept the message request, delete it, or report it,” Instagram said in its announcement.

According to the company, the filtering takes place on a person’s device with no message content sent back to its server, unless the user chooses to submit a report about the abuse.

Ireland is one of the first markets that the tool is being launched in over the coming weeks, along with the UK, France, Germany, Australia, New Zealand and Canada. The feature will be found in Instagram’s privacy settings under a section called ‘hidden words’.

Instagram said the feature was introduced to help people, especially those with large followings, to avoid any abusive messages.

In a recent high-profile case, former professional footballer Ian Wright received abusive, racist messages on Instagram from an Irishman, who ultimately avoided a criminal conviction after being brought before the courts.

Instagram said that it is also adding features to prevent an abusive user from creating new accounts to try contacting people again – though it is not clear how this will work technically.

“This is in addition to our harassment policies, which already prohibit people from repeatedly contacting someone who doesn’t want to hear from them,” it said. “We also don’t allow recidivism, which means if someone’s account is disabled for breaking our rules, we would remove any new accounts they create whenever we become aware of it.”

Jonathan Keane is a freelance business and technology journalist based in Dublin

editorial@siliconrepublic.com