Social media is where the internet gathers. As a result, social media can be a mess.
Sometimes, people just can’t deal with certain topics. Whether it's news about war, celebrity drama, plot twists or spoilers, or personal triggers like death, loss, or breakups. Muting words helps them protect their peace. With the feature, users can mute political keywords and influencers they dislike, or even entire topics like "crypto" or "Taylor Swift" just to keep their feed sane and relevant.
Threads has this in the form of a content filter.
And this time, Meta is giving it an upgrade.
The social media from meta is stepping out of Instagram’s shadow by launching its own standalone word-blocking feature.
Announced by Instagram head Adam Mosseri, the new Hidden Words setting on Threads now lets users block specific words, phrases, and even emojis in customizable batches—with the option to set time limits.
Previously, Threads’ content filters were tied directly to Instagram, meaning whatever users muted on one app applied to the other.
Now, users can tailor filters independently on Threads—shaping their experience across feeds, profiles, search, and replies.
A handy addition is the ability to temporarily snooze words for up to 30 days—perfect for dodging spoilers or avoiding a flash-flood of drama without going full mute. Instagram doesn’t yet offer this snooze option, though rival platform X (formerly Twitter) has long allowed timed muting.
It's worth noting that Threads also has a feature to mark something as a spoiler.
When enabled, photos are automatically blurred in the feed, and text is grayed out—clicking on either reveals the hidden content. It’s a sleek way to avoid ruining surprises while still letting users share freely.
Netflix and Marvel are among the first major studios to adopt the feature for their posts, signaling a big shift toward more responsible spoiler etiquette.
Other platforms like Reddit, Instagram, and Bluesky already offer similar spoiler controls.
Meanwhile, X still lacks a true spoiler tag—though users can mark media as sensitive, which creates a blurred preview until manually revealed.