Automated Red Teaming with GOAT: the Generative Offensive Agent Tester

Maya Pavlova, Erik Brinkman, Krithika Iyer, Vítor Albiero, Joanna Bitton, Hailey Nguyen, Cristian Canton Ferrer, Ivan Evtimov, Aaron Grattafiori

Track: Long Paper Track (up to 9 pages)

Keywords: red teaming, adversarial machine learning, adversarial examples, attacks on language models

TL;DR: We propose an automated agentic red teaming system that simulates plain language adversarial conversations while leveraging multiple adversarial prompting techniques to identify key vulnerabilities in LLMs.

Abstract: Red teaming assess how large language models (LLMs) can produce content that violates norms, policies, and rules set forth during their safety training. However, most existing automated methods in literature are not representative of the way common users exploit the multi-turn conversational nature of AI models. While manual testing addresses this gap, it is an inefficient and often expensive process. To address these limitations, we introduce the Generative Offensive Agent Tester (GOAT), an automated agentic red teaming system that simulates plain language adversarial conversations while leveraging multiple adversarial prompting techniques to identify vuLnerabilities in LLMs. We instantiate GOAT with 7 red teaming attacks by prompting a general purpose model in a way that encourages reasoning through the choices of methods available, the current target model’s response, and the next steps. Our approach is designed to be extensible and efficient, allowing human testers to focus on exploring new areas of risk while automation covers the scaled adversarial stress-testing of known risk territory. We present the design and evaluation of GOAT, demonstrating its effectiveness in identifying vulnerabilities in state-of-the-art LLMs, with an ASR@10 of 96% against smaller models such as Llama 3.1 8B, and 91% against Llama 3.1 70B and 94% for GPT-4o when evaluated against larger models on the JailbreakBench dataset.

Submission Number: 32