In one of the first announcements since Instagram's co-founders departed their company several weeks ago, the social media platform is launching new anti-bullying tools. New head of Instagram Adam Mosseri announced in a blog post that the app is now using machine learning to detect harassment in photos and captions, immediately sending those posts for review.
Additionally, the company is expanding its bullying comment filter to live videos. As a quick refresher: the comment filter is also enabled by machine learning. In May 2018, Instagram rolled out the option for users to toggle on a setting that hides bullying comments (as identified by the AI) by default.
The updates come shortly after some high-profile instances of toxicity on the platform, including pop singer Ariana Grande's fans harassing her fiance, comedian Pete Davidson, to the point of his temporary departure from the app. With these changes, Instagram is trying to automate the overwhelming task of moderation — but it may be a drop in the bucket when it comes to stemming the tide of hateful comments on an app known for portraying idealized images and lifestyles.
Still, Instagram is attempting to promote better behavior with a new "spreading kindness" campaign featuring teen dancer and actor Maddie Ziegler (of Sia music video fame), which encourages users to tag friends and spread a new camera effect with positive, uplifting sentiments.
The bullying comment filter on live videos is now available globally, while bullying detection in photos and captions is rolling out over the coming weeks.