OpenAI · posted · checked
Product Manager, Frontier Risks Safety
$347k – $445kHybrid · San FranciscoSenior
Who can apply
Hybrid · San Francisco
Location: San Francisco
About the Team
The Frontier Risks Safety team sets out safety requirements and executes planning for frontier model launches, including biological and cyber risks and government models. We work across research, engineering, and policy to assess risks, define and validate safeguards, and drive safety readiness throughout the model development and deployment process. We also help shape frontier trusted-access programs, ensuring that access to advanced capabilities is supported by appropriate safeguards, clear requirements, and rigorous evaluation.
About the Role
As a Product Manager on the Frontier Risks Safety team, you will own product strategy, requirements, and delivery for initiatives that mitigate frontier AI risks and enable responsible access to advanced capabilities.
You will partner with researchers, engineers, policy teams, and independent experts to identify the most important problems to solve, translate research and risk assessments into product requirements, and prioritize what to build. You will guide work from discovery through deployment and iteration, making tradeoffs explicit and measuring whether the resulting safeguards and workflows address the risks they are intended to mitigate.
We’re looking for people who combine strong technical judgment, experience building and shipping complex technical products, and a commitment to AI safety. You should be comfortable working through uncertainty, engaging deeply with research, and adapting direction as capabilities, evidence, and user needs evolve.
This role is based in San Francisco, CA. We use a hybrid work model of 3 days in the office per week and offer relocation assistance to new employees.
In this role, you will:
Set product strategy and priorities. Develop a shared understanding of frontier risks, deployment needs, and gaps in safeguards. Own the roadmap for mitigations, prioritizing investments based on risk, evidence, technical feasibility, and the value the underlying capabilities enable for users.
Drive frontier trusted-access programs. Ensure defenders and scientists have access to the frontier capabilities they need to strengthen security and accelerate scientific discovery.
Turn safety needs into product requirements. Work with model teams, safety researchers, engineers, and policy partners to translate threat models, evaluation findings, and user needs into clear requirements, acceptance criteria, and measurable outcomes.
Drive safeguards from concept to deployment. Partner with research and engineering teams to turn promising safety approaches into effective mitigations across models and products. Resolve product and technical tradeoffs, manage dependencies, and drive delivery of improvements that close material safety gaps.