Detect Hidden Profanity
How to keep users safe from high-risk, manipulative content
Request a Demo
Send us your details in the form below, and we'll be in touch within 24 hrs.
Users who are determined to post high-risk content will always search for ways to manipulate the system.
Today’s internet users are savvy, so you can count on a percentage of your users to bypass traditional content filters using manipulations like l33t speak and creative misspellings. With Community Sift, you will always be one step ahead of the trolls. Our system was designed to catch the various ways that users manipulate language to sidestep community guidelines. Here is just a sampling of the sneaky content that we detect:
- l337 5p34k (1 h473 y0u, y0u 5h0uld ju57 k1ll y0ur53lf)
- Deliberate misspellings in context (you’re so dumq vs go to the dumq)
- Hidden caps across and within words (nice pASS, KickIntoLessLemon You)
- Reversed words and inverted letters (you kcus, so ylgu)
Our team leverages powerful built-in pattern recognition that allows us to monitor billions of lines of User Generated Content every day. We can detect emerging manipulations as they happen, and prevent them from ever affecting your community. We would love to chat about the importance of protecting your users from threatening content.