From 008b8c1c83a1b2f9e81d3fbcb2be3d673f46b8af Mon Sep 17 00:00:00 2001 From: Omar Santos Date: Sun, 26 Jan 2025 21:11:33 -0500 Subject: [PATCH] Update ai_security_tools.md --- ai_research/ai_security_tools.md | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/ai_research/ai_security_tools.md b/ai_research/ai_security_tools.md index 2c5e913..e7c04d9 100644 --- a/ai_research/ai_security_tools.md +++ b/ai_research/ai_security_tools.md @@ -40,8 +40,12 @@ _Products that intercept prompts and responses and apply security or privacy rul - [Skyflow LLM Privacy Vault](https://www.skyflow.com/product/llm-privacy-vault) - Redacts PII from prompts flowing to LLMs. - [Guardrails AI](https://www.guardrailsai.com) - Guardrails runs Input/Output Guards in your application that detect, quantify and mitigate the presence of specific types of risks. [![code](https://img.shields.io/github/license/guardrails-ai/guardrails)](https://github.com/guardrails-ai/guardrails/) +## OWASP Security Resources +- [OWASP's List of AI Security Tools](https://owaspai.org/docs/5_testing/#open-source-tools-for-predictive-ai-red-teaming) + ## AI Red Teaming Datasets - [AttaQ Dataset](https://huggingface.co/datasets/ibm/AttaQ) - a red teaming dataset consisting of 1402 carefully crafted adversarial questions ## AI Red Teaming Guidance +- [OWASP AI Testing Guidance](https://owaspai.org/docs/5_testing/) - [HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal](https://arxiv.org/pdf/2402.04249)