TRUST & SAFETY ENGINE
Comprehensive trust and safety breakdown detailing strict guardrails against harassment, personal attacks, and hate speech, ensuring roasts stay light-hearted, safe, and consensual.
We built BURNBOT for ruthless laughs, not harm. Here is how our live moderation architecture keeps the cellar fun, consensual, and sharp.
RULE OF THE CLUB
Spicy banter, zero cruelty.
Every AI response is evaluated through safety classifiers before it reaches your screen. If an output crosses line into hate speech or doxxing, it gets incinerated before delivery.
Protected characteristics are strictly off-limits.
BURNBOT uses multi-layer semantic filters to prevent roasts targeting race, ethnicity, religion, disability, gender identity, or sexual orientation. Any attempt to bait discriminatory insults triggers an instant system refusal.
Enforcement: Automatic prompt block + safety log flag.