From 37c18dd79e37e7c10999295ad84239b5c88fb50e Mon Sep 17 00:00:00 2001 From: Matteo Meucci Date: Sun, 23 Nov 2025 17:42:29 +0100 Subject: [PATCH] Update AITG-MOD-07_Testing_for_Goal_Alignment.md --- .../tests/AITG-MOD-07_Testing_for_Goal_Alignment.md | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-) diff --git a/Document/content/tests/AITG-MOD-07_Testing_for_Goal_Alignment.md b/Document/content/tests/AITG-MOD-07_Testing_for_Goal_Alignment.md index 49f055f..6a6b315 100644 --- a/Document/content/tests/AITG-MOD-07_Testing_for_Goal_Alignment.md +++ b/Document/content/tests/AITG-MOD-07_Testing_for_Goal_Alignment.md @@ -31,14 +31,10 @@ This test evaluates vulnerabilities associated with AI model goal misalignment, - **Output Filtering and Moderation**: As a final layer of defense, pass the model's output through a separate moderation API or filter that can catch any remaining misaligned or harmful content before it reaches the user. ### Suggested Tools -- **Promptfoo**: An open-source tool for evaluating LLM output quality and testing for regressions. Excellent for creating test suites to check for goal alignment against a set of predefined criteria. - - Tool Link: [Promptfoo on GitHub](https://github.com/promptfoo/promptfoo) -- **Garak**: An open-source framework for LLM vulnerability scanning, including probes specifically designed to test for goal misalignment and ethical boundary violations. - - Tool Link: [Garak on GitHub](https://github.com/leondz/garak) -- **NVIDIA NeMo Guardrails**: An open-source toolkit for adding programmable guardrails to LLM applications, helping to enforce alignment and prevent unwanted behaviors. - - Tool Link: [NeMo Guardrails on GitHub](https://github.com/NVIDIA/NeMo-Guardrails) -- **Microsoft Guidance**: A tool for controlling LLMs, ensuring that outputs strictly adhere to predefined guidelines and formats. - - Tool Link: [Guidance on GitHub](https://github.com/microsoft/guidance) +- **Microsoft Guidance**: A tool for controlling LLMs, ensuring that outputs strictly adhere to predefined guidelines and formats - [Guidance on GitHub](https://github.com/microsoft/guidance) +- **Promptfoo**: An open-source tool for evaluating LLM output quality and testing for regressions. Good for creating test suites to check for goal alignment against a set of predefined criteria - [Promptfoo on GitHub](https://github.com/promptfoo/promptfoo) +- **Garak**: Including probes specifically designed to test for goal misalignment and ethical boundary violations - [Garak on GitHub](https://github.com/leondz/garak) +- **NVIDIA NeMo Guardrails**: An open-source toolkit for adding programmable guardrails to LLM applications, helping to enforce alignment and prevent unwanted behaviors - [NeMo Guardrails on GitHub](https://github.com/NVIDIA/NeMo-Guardrails) ### References - Askell, Amanda, et al. "A General Language Assistant as a Laboratory for Alignment." Anthropic, 2021. [Link](https://arxiv.org/abs/2112.00861) (Constitutional AI)