Novel Abuse Red Teamer

The Intelligence and Investigations team at OpenAI is dedicated to ensuring the safe, responsible deployment of AI by rapidly detecting and mitigating abuse. As a Novel Abuse Red Teamer, you'll advance our pre- and post-deployment testing efforts, ensuring our models and products remain resilient against evolving, real-world adversarial threats.

You will design, execute, and refine innovative adversarial testing and simulation protocols that leverage state-of-the-art tools and a hands-on red team approach. You will simulate threat actor methodologies and conduct rigorous application testing to uncover novel abuse vectors.

Key responsibilities include:

• Designing and executing high-fidelity, realistic attack simulations that emulate sophisticated threat actor methodologies
• Analyzing testing outcomes and generating actionable insights to refine detection mechanisms
• Serving as a critical technical bridge between security, product, and policy teams
• Developing and integrating automation tools to streamline testing workflows
• Collaborating closely with internal stakeholders to align testing strategies with evolving product needs and security objectives

The ideal candidate will have experience in application security, penetration testing, or red team operations, with a proven track record of executing adversarial testing and simulation protocols. Strong technical background in web application security and hands-on expertise with tools such as Burp Suite, Metasploit, and similar are required.

OpenAI offers a salary range of $158.4K - $320K, a hybrid work model, and relocation assistance to new employees.