Instagram stock photo
[Photo by: Pexels]

Instagram is using AI to find bullying in posts

Instagram is now using artificial intelligence to root out bullying in photos and captions, the Facebook-owned photo and video-sharing app announced Tuesday. The same technology is also being applied to Instagram live videos.

To roll out officially in the coming weeks, the new tech gives the ‘gram a better handle on in-app bullying. It goes above and beyond someone simply reporting an inappropriate image or comment, explains new Instagram head Adam Mosseri.

Read more: Hayley Williams blasts Mike Huckabee for Taylor Swift tweet

“Online bullying is complex,” he says, “and we know we have more work to do to further limit bullying and spread kindness on Instagram. That’s why today we’re announcing our latest tools to help combat bullying, including a new way to identify and report bullying in photos. We’re also introducing a camera effect to help spread kindness in Stories.”

And the latest change is just in time for October’s National Bullying Prevention Month in the U.S. and Anti-Bullying Week in the U.K., where it’s also being implemented. It’s also the first new product announcement under new head Mosseri.

But how does it work? The technology that enables such detection was surveyed last year by Wired, which highlighted the company’s inner attempts at machine-learning comment filters. But it’s not a perfect solution for online harassment.

Still, artificial intelligence will assist in the effort to “proactively detect bullying” on the platform, stepping in with machine learning where human detection often fails, as reported by The Verge. A fun new “kindness” camera effect accompanies the latest change, the latest face filter option a collaboration with teen dancer Maddie Ziegler.

What do you think of Instagram’s attempts to curb online bullying? Do you agree with the use of artificial intelligence in wiping out inappropriate comments or photos on the social media app? Sound off in the comments with your thoughts.

Watch more: 10 surprising cover songs