Rebuff

A self-hardening prompt injection detector

Description:

Rebuff.ai is a multi-layered defense system designed to protect AI applications from prompt injection (PI) attacks. It offers four layers of defense: heuristics, LLM-based detection, Vectordb, and Canary tokens. The system can detect PI on user input and canary word leakage in completion using an AI model such as OpenAI's GPT-3.

Price:
GitHub
Share on FacebookShare on TwitterShare on LinkedIn
View
Rebuff
and other useful AI tools on Daily Zaps
#DailyZaps
#DailyZaps
Daily Zaps A.I. Hub

Explore Similar AI Tools

Copyright 2023. All Rights Reserved