SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
OpenAI's Safety Systems team is hiring a Model Policy Manager to shape model behavior for U.S. government use, with a focus on national security applications. You will define nuanced policies and translate them into training and evaluation criteria, helping frontier models navigate high-stakes scenarios while preserving their usefulness and capabilities.
In this role, you will develop model policies that guide safe and useful behavior across real-world environments. You'll build evaluations, identify policy gaps and model failures, and use findings to improve policies and training. You'll work cross-functionally with research, engineering, and domain experts to support safe, reliable deployment of frontier models in security-critical settings.
The Safety Systems team manages the complete lifecycle of safety efforts for OpenAI's frontier models, ensuring responsible deployment and positive societal impact. The work spans diverse research and engineering initiatives—from system-level safeguards and model training to evaluation and red-teaming—all aimed at mitigating misuse, misalignment, and maintaining high safety standards. The Model Policy team specifically designs policies that define safe model behavior and ensure models operate reliably in real-world environments.
REQUIREMENTS:
- Relevant experience in AI safety, policy, or risk assessment
- Strong judgment and ability to translate complex safety questions into clear, practical policies
- Technical fluency to work hands-on with model data and evaluations
- Motivation aligned with OpenAI's mission and responsible AI use in safety-critical settings
- Active TS/SCI clearance or equivalent (preferred but not required)
