Model Policy Manager, Agentic Safety
OpenAI · San Francisco, CA
Posted Nov 16, 2026Apply by 11/25/26
OpenAI is hiring a Model Policy Manager, Agentic Safety who will shape how OpenAI understands and addresses risks from model misalignment as models become more autonomous, investigating failures such as models persisting toward the wrong objective, taking unsafe shortcuts or circumventing constraints, and turning findings into behavioral policies, evaluation criteria, monitoring and safeguards. The position pays $207,000 to $335,000 plus equity and is based in San Francisco on a hybrid schedule, three days a week in the office. Position is located in San Francisco, CA.
Model Policy Manager, Agentic Safety, OpenAI (Safety Systems, Model Policy team). San Francisco, CA. Compensation: $207K - $335K + Offers Equity. No application deadline stated.
About the role: Within Safety Systems, the Model Policy team works to ensure increasingly capable models behave safely and reliably in real-world environments. In this role, you will shape how OpenAI understands and addresses real-world risks that emerge from model misalignment as models become more autonomous and operate over longer horizons. You will investigate how misaligned behavior emerges across extended trajectories, including when models persist toward the wrong objective, take unsafe shortcuts, lose track of instructions, exploit weaknesses in their environment, or circumvent constraints, and translate these insights into behavioral policies, evaluations, monitoring, and safeguards.
Responsibilities: identify vulnerabilities that emerge as models interact with tools, data, and external systems, and translate them into model- and system-level safeguards; develop threat models and empirical frameworks for understanding harmful outcomes from misaligned behavior; turn findings into policy frameworks, evaluation criteria, online measurement and safeguards; develop human data campaigns and gold sets; partner with research, engineering, security, and product teams to shape model and system safety; inform deployment decisions, system cards, and safeguards reports; and build monitoring approaches that detect regressions and emerging risks after deployment.
This role is based in OpenAI's San Francisco office, on a hybrid model of three days in the office per week with optional work from home on Thursdays and Fridays. Relocation support is offered.
Apply →Applications are handled by the employer.