No Free Lunch with Guardrails

As large language models (LLMs) and generative AI become widely adopted, guardrails have emerged as a key tool to ensure their safe use. However, adding guardrails isn't without tradeoffs; stronger security measures can reduce usability, while more flexible systems may leave gaps for adversarial attacks. In this work, we explore whether current guardrails effectively prevent misuse while maintaining practical utility. We introduce a framework to evaluate these tradeoffs, measuring how different guardrails balance risk, security, and usability, and build an efficient guardrail. Our findings confirm that there is no free lunch with guardrails; strengthening security often comes at the cost of usability. To address this, we propose a blueprint for designing better guardrails that minimize risk while maintaining usability. We evaluate various industry guardrails, including Azure Content Safety, Bedrock Guardrails, OpenAI's Moderation API, Guardrails AI, Nemo Guardrails, and Enkrypt AI guardrails. Additionally, we assess how LLMs like GPT-4o, Gemini 2.0-Flash, Claude 3.5-Sonnet, and Mistral Large-Latest respond under different system prompts, including simple prompts, detailed prompts, and detailed prompts with chain-of-thought (CoT) reasoning. Our study provides a clear comparison of how different guardrails perform, highlighting the challenges in balancing security and usability.

SAGE-RT: Synthetic Alignment data Generation for Safety Evaluation and Red Teaming

We introduce Synthetic Alignment data Generation for Safety Evaluation and Red Teaming (SAGE-RT or SAGE) a novel pipeline for generating synthetic alignment and red-teaming data. Existing methods fall short in creating nuanced and diverse datasets, providing necessary control over the data generation and validation processes, or require large amount of manually generated seed data. SAGE addresses these limitations by using a detailed taxonomy to produce safety-alignment and red-teaming data across a wide range of topics. We generated 51,000 diverse and in-depth prompt-response pairs, encompassing over 1,500 topics of harmfulness and covering variations of the most frequent types of jailbreaking prompts faced by large language models (LLMs). We show that the red-teaming data generated through SAGE jailbreaks state-of-the-art LLMs in more than 27 out of 32 sub-categories, and in more than 58 out of 279 leaf-categories (sub-sub categories). The attack success rate for GPT-4o, GPT-3.5-turbo is 100% over the sub-categories of harmfulness. Our approach avoids the pitfalls of synthetic safety-training data generation such as mode collapse and lack of nuance in the generation pipeline by ensuring a detailed coverage of harmful topics using iterative expansion of the topics and conditioning the outputs on the generated raw-text. This method can be used to generate red-teaming and alignment data for LLM Safety completely synthetically to make LLMs safer or for red-teaming the models over a diverse range of topics.