Instagram develops AI-Powered tool to detect offensive captions

Instagram recently announced that it would be rolling out a new feature that will notify people if the caption included in their post may be considered offensive.

The company says the notification is intended to give users a chance to pause and reconsider their words before posting. The feature still relies on an AI algorithm that can recognize different forms of bullying on the platform even in the comment section.

Instagram has increased its efforts on fighting online bullying on the platform and trying to make it a haven. The new tool, according to Instagram, will also help users be educated on what is not allowed on the platform. Plus, this will help users know when their accounts are at risk of breaking the set rules.

The new feature is currently under testing in a few select countries, and will be expanded globally “in the coming months.”

Earlier this year, the company launched a feature that notifies people of offensive comments on posts before they’re posted.

“Results have been promising, and we’ve found that these types of nudges can encourage people to reconsider their words when given a chance,” the company said in a blog post.

Earlier this month, Instagram rolled out age restrictions to prevent underage people from joining the platform.

Just like on Facebook, Instagram is also fighting against misinformation. The company acknowledges this challenge, and in May, it began working with third-party fact-checkers exclusively in the US to help identify, review, and label false information.

Now, Instagram is expanding its reach to more users globally, thus allowing more third-party fact-checkers to assess and rate any photo or video-based information posted on the platform.

If the content is marked as false or partly false, Instagram will reduce its distribution, consequently removing it from the Explore tab as well as hashtag pages. Furthermore, inaccurate content will also be labeled so and everyone across Instagram – in feed, profile, stories, and direct messages – will be able to it that way.

Content flagged as false or partly false on Facebook will also be automatically labeled as false on Instagram and vice versa.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.