Instagram's new test lets you mute specific words from suggested posts

The company's push toward more recommendations has been deeply unpopular.

NurPhoto via Getty Images

Instagram is giving users more ways to tweak their suggested posts amid a backlash to the app’s aggressive shift toward recommendations. The app is testing a new option that will allow users to use keywords and emoji to mute certain topics from appearing in suggested posts.

The change will block posts in which the users’ keywords, which can include emoji as well as words and phrases, appear in the caption or hashtag for a post. “You can use this feature to stop seeing content that’s not interesting to you,” Instagram writes in a blog post. Users can customize their filter words from the app's settings. The company notes that people can also opt to snooze all recommendations entirely.

Instagram is also testing a new way to weed out unwanted posts from the app’s Explore section. With the change, users can select multiple posts at a time and mark them all as “not interested.” This will hide those posts, and block out similar recommendations in the future, according to the company.

Instagram will allow users to mute specific words and emoji from suggested posts.

While both new options require a bit of extra work, the changes could bring some relief for users’ who have been frustrated by the quality of Instagram’s recommendations as the app has taken increasingly aggressive steps to become more like TikTok. Instagram’s top exec Adam Mosseri said last month the company would tone down the number of recommended posts and halt its experiment with a full-screen feed. Both changes have been deeply unpopular, prompting viral memes criticizing the company’s efforts to copy TikTok.

Regardless of criticism, Meta’s leaders have been clear that they intend to shift both Facebook and Instagram’s feeds from mostly friend content to more posts from AI-driven recommendations. But the new controls could help the company eventually improve the quality of those suggestions, which might make them more palatable to users in the long run.