From cb3f36e44d662cff0c20dbe72b8d91c5b8f7f93c Mon Sep 17 00:00:00 2001 From: Fuwn Date: Thu, 25 Sep 2025 16:48:28 -0700 Subject: fix(gateway:aiModeration): Improve hate speech analysis --- packages/gateway/src/listeners/aiModeration.ts | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'packages/gateway') diff --git a/packages/gateway/src/listeners/aiModeration.ts b/packages/gateway/src/listeners/aiModeration.ts index 4f722b0..57bf4c7 100644 --- a/packages/gateway/src/listeners/aiModeration.ts +++ b/packages/gateway/src/listeners/aiModeration.ts @@ -125,6 +125,14 @@ SLUR GUIDELINES: - Context matters: Consider if the word is being used in a hateful manner vs. casual/reclaimed usage - When in doubt, err on the side of caution for hard slurs but be more lenient with soft slurs +HATE SPEECH GUIDELINES: +- CASUAL SLANG: Words like "hoes", "bitches", "sluts", "faggot", "nigga" (soft R) in casual conversation are generally acceptable +- PROHIBITED: "nigger" (hard R) is still prohibited and should be flagged +- ACTUAL HATE SPEECH: Targeted harassment, threats, or demeaning language directed at specific individuals +- Context matters: Casual use of most slurs = generally acceptable, targeted harassment = hate speech +- Only flag if the language is genuinely harmful or targeted harassment +- Most slurs and offensive language in casual conversation should NOT be flagged + RULE #6 CLARIFICATION: - Rule #6 (zero-tolerance) ONLY applies to underage/loli/cunning content - General adult sexual content is ALLOWED in NSFW channels -- cgit v1.2.3