SaferChat AI

🛡️ SaferChatAI — AI Moderation That Actually Gets It

> Harassment, hate speech, and toxic behaviour don't follow keyword rules.
> SaferChatAI understands context — not just catches words.

---

THE PROBLEM

Keyword filters are blind. They miss typos, coded language, and anything that's technically not a slur but absolutely is harassment.

"Fuk off, your setup is embarrassing lmao, typical behaviour from people like you"

❌ AutoMod / MEE6 — No keyword match. Passes silently.
✅ SaferChatAI — Harassment flagged. Full report sent to your mod team.

"not being racist but those communities are just more aggressive, it's just facts lol"

❌ AutoMod / MEE6 — Polite language. No slurs detected.
✅ SaferChatAI — Coded hate speech identified and flagged.

---

SETUP IN 2 MINUTES

/config → pick mod channel → done

No config files. No complex rules. Monitoring starts immediately.

---

FEATURES

🔍 Context-aware detection — Reads the full conversation, not just single messages. Catches what keyword filters miss, including typos and slang.

🛡️ You stay in control — always — Never auto-bans. Every report gets one-click buttons: Delete · Warn · Timeout · False Positive · Ban. Your mods decide.

💙 Crisis detection — Self-harm language triggers a private, compassionate DM with crisis resources. Your mod team gets an urgent alert.

⚡ Instant word filter — Build your blocklist. Blocked words vanish before anyone sees them. Zero delay.

🔗 Link protection — 5 configurable modes. Scam detection, phishing blocking, custom domain lists.

📊 Moderation dashboard — /statistics shows 30-day breakdown: incidents by category, top offenders, false positive rate.

🎯 Learns from your team — Every false positive your mods mark makes SaferChatAI more accurate to your server's culture.

🌍 Multilingual — Works in English, German, and more — catches harassment regardless of language or intentional misspellings.

---

COMMANDS

/config — Setup: mod channel, auto-delete, word filter, link protection, exemptions
/advanced — Per-category sensitivity (Strict · Standard · Relaxed · Off)
/mod history @user — Incident & action history
/mod warn @user — Official warning via DM
/statistics — 30-day moderation dashboard
/help — Full command guide

---

WORKS ALONGSIDE YOUR EXISTING SETUP

SaferChatAI runs next to AutoMod, MEE6, and Carl-bot — not instead of them. It's the context layer on top of your existing rules.

---

FREE DURING BETA

✅ No credit card · ✅ No trial expiry · ✅ No feature gating · ✅ Remove any time
✅ Beta users get a permanent discount when paid plans launch

saferchat.org · Built in Innsbruck, Austria 🇦🇹