jobs
Researcher, Frontier Robustness
About the Team The Frontier Robustness team advances the science of securing frontier AI systems. We develop new methods to keep increasingly capable models contained, observable, and under human…
En inglés
- Company
- OpenAI
- Location
- San Francisco
- Status
- Open
- Posted
- 2026-11-16 21:19 UTC
OpenAI's Frontier Robustness team is hiring a researcher to study how frontier models can be contained, monitored and kept under human control. The work covers threat models, adversarial tests, sandbox escapes, unauthorized communication between agents and access to sensitive systems. The researcher designs experiments and evaluations, reproduces vulnerabilities, prototypes defenses, red-teams safeguards and works with security and infrastructure teams to deploy improvements. The posting asks for hands-on LLM research, independent coding, rigorous evaluation design and practical security experience in sandboxing, vulnerability research or monitoring.
Original job posting