Built around you

Hate is personal. Your protection should be too.

Built around your identity and the specific abuse you face, not a generic filter.

Sign-up takes a few minutes. No card needed.

Why generic filters don't work

Most automated comment moderation works from a list of banned words. If the word is on the list, the comment is flagged. If it isn't, it goes through. That catches the most explicit abuse. It misses almost everything else.

Racial abuse doesn't always use slurs. Antisemitic abuse arrives framed as political opinion. Misogynistic abuse shows up as commentary on appearance or credibility. Homophobic and transphobic targeting uses dog whistles and coded language that reads as innocuous to a keyword system. The people sending this abuse know exactly how the systems work, and they write to stay just inside the line.

Fendr can understand that a phrase is a dog whistle for a specific community, that a comment is targeted harassment on your account specifically, that context makes something abusive here that wouldn't be somewhere else, because it's built around you, not around a generic content policy.

What "built around you" means

When you set up Fendr, you tell it who you are: your identity, your background, the communities you belong to, the abuse you typically face, the context of your account. Fendr builds a protection profile specific to you, and that profile shapes how it evaluates every comment.

  • Identity-aware, Understands the specific forms of hate that target your community, including coded language, dog whistles, and abuse framed as debate or opinion.
  • Context-aware, What's abusive on your account may not be abusive on someone else's. Fendr's evaluation is specific to you.
  • Adapts over time, As your account evolves and the nature of abuse you face changes, Fendr's protection adapts with it.

What this catches that generic filters miss

  • Coded antisemitic language that reads as political commentary.
  • Islamophobic abuse framed as security or cultural concern.
  • Transphobic comments presented as debate about biology.
  • Racial dog whistles and in-group references.
  • Misogynistic "criticism" of appearance or credibility.
  • Targeted harassment that uses no banned words at all.

See it working on your account.

Free for 14 days. No card required. Plans from €49/month.

14-day free trial · cancel anytime · no card to start

Personalised Hate Detection Built Around You | Fendr