Researcher, Agent Safety, Oversight and System Mitigations
OpenAI · San Francisco
About this role
**About the Team** The Agent Safety team ensures increasingly capable AI agents act safely, exercise sound judgment, and remain aligned with user intent. Work spans training, measurements, and oversight to reduce severe unintended outcomes while preserving effective autonomous action. **About the Role** Focus on oversight and system-level mitigations enabling safe autonomous agent operation in real environments. Design and build practical oversight systems, study long-term agent supervision and constraint approaches, and develop mitigations grounded in security principles. **Location:** San Francisco, CA (Hybrid - 3 days/week in office) **Key Responsibilities** • Design, build, and evaluate system-level controls for agent actions with sandboxing and permission boundaries • Collaborate with engineering teams to productionize AI controls • Red-team agentic systems to measure control effectiveness against data exfiltration and unsafe tool use • Optimize safety-productivity tradeoffs through measurement and iteration **Ideal Candidate** • Strong systems or security instincts with concrete reasoning about isolation, permissions, and failure modes • Ability to translate ambiguous safety questions into threat models, experiments, and practical mitigations • Experience building robust experimental infrastructure and rigorous evaluations • Deep interest in frontier AI alignment, safety, and control Background in AI control or security welcome but not required.
Listing freshness
CronJobs last confirmed this listing 19h ago. If its source stops confirming the opening for seven days, this page is removed from active inventory.