Common Filter Types.md

Filter Type	What It Does
Harmful Language Filters	Detect and block hate speech, abusive language or profanity. Sensitivity thresholds can be tuned to balance safety with the risk of false positives.
PII Filters	Identify personally identifiable information, such as phone numbers, emails or account numbers, and prevent it from being exposed.
Advanced Safety Filters	Use more comprehensive models to flag issues like jailbreak attempts, bias, hallucinated responses or violent, unethical content.