Instagram is Using AI to Block Offensive Comments and Spam

Artificial Intelligence (AI) is on the rise, both in public awareness and from companies trying to capitalize on the technology.

Which is why it isn’t too surprising that even Instagram, the giant social network focused on sharing photos and videos, is even using it. Today, Instagram has officially announced a pair of new AI-based features baked into the platform that will hopefully make the social network a bit easier to navigate, at least when it comes to avoiding any offensive comments and/or spam.

The service actually started skimming comments back in September of last year, but with this new AI push, the feature is expanding quite a bit. Before, the feature would only remove comments that had specific words or phrases that had already been specifically identified as offensive. Users could also add words or phrases to the feature as well.

The feature is using machine learning to better identify comments on a general spectrum that appear to be offensive. It actually gives the system some ability to identify a reply’s context, which Instagram says should help reduce false positives and catch more bad comments. Interestingly enough, when the system does flag and block a comment, it is removed from sight from all Instagram users — except for the person who posted the offending comment.

The more advanced comment filter is now turned on by default, too. Previously users had to go in and turn it on if they wanted their comments skimmed. Users will be able to turn it off and on when they want, and they will also still be able to add their own custom phrases and words to the filter as well.

In addition to that, Instagram is officially unveiling a feature that it has actually had in place for months now. It’s an AI-based spam filter. It’s available in nine different languages, including English.

Both of these new features are available now.

[via Instagram Blog]