“Did you know sx isn’t supposed to hurt?” read the words on an Instagram reel posted by the sexual health and wellness company The Pelvic People in early 2025.

At first glance, the missing “e” looks like a typo. But it’s not. A scroll through the Pelvic People’s Instagram page and others like it on the social media platform reveals videos littered with words like “s*x,” “lub3,” and “c00chie.” 

Commonly referred to as “algospeak,” this popular form of online lingo exists to sidestep algorithms employed by tech companies like Meta, TikTok, and X that penalize or even remove posts for discussing sensitive topics like sex, mental health, or substance use. 

Posts perform better with “a little bit of censorship,” said Emily Tran, social media manager at the Pelvic Health and Rehabilitation Center, who uses the center’s Instagram, @pelvichealth, to share evidence-based information about sexual health. 

“If I were to write something about female anatomy and use the correct terms, [the algorithm] might flag it and say that it’s inappropriate,” Tran added. 

The algorithm would then reduce the post’s visibility, leading to a decline in engagement—something Tran said happened to the account before she started using algospeak. 

Creators who discuss sensitive topics often see this kind of self-censorship as essential to posting freely without losing engagement. But, experts said in interviews with Rewire News Group, censored language like algospeak can feed the stigma around these topics and even pose threats to sexual health. And recent research has cast doubt on whether using algospeak actually improves their social media visibility. 

Researchers at the University of St. Cyril and Methodius in Slovakia gathered the most recent 50 Instagram posts from nine sex education accounts that post in English, Slovak, or Czech. They analyzed the likes, comments, and shares on a total of 450 posts to understand overall engagement.

The resulting study, which was published in the journal Media Literacy and Academic Research in June 2025, found no statistically significant decrease in likes and shares between posts that did not use algospeak compared with those that did.

Researchers also found........

© Rewire.News