Researcher, Recursive Self-Improvement Safety
About The Team
Preparedness is a Safety Research team at OpenAI, focusing on mitigating AI threats that could scale to extreme severity. Our work involves tracking and predicting misalignment propensities and capabilities of frontier AI systems, implementing mitigation strategies, and coordinating efforts to keep misuse safeguards, alignment tools, and security measures aligned with future threats.
About The Role
We are hiring strong technical executors to support preparations for recursive self-improvement. This role requires reasoning about future problems that might not exist now, emphasizing strategic thinking and execution. Key responsibilities include scalable oversight, automated auditing, rigorous monitorability, model behavior science, coordination and verification, and AI R&D risk measurement. The role demands careful consideration of future threats and translating open-ended objectives into actionable tasks.
- Carefully consider future problems and prepare for them.
- Execute quickly, building scrappy prototypes, and iteratively improve them.
- Secure buy-in from other staff at OpenAI and communicate work clearly.
- Collaborate with or manage other staff as needed.
Qualifications
To succeed, you should be an exceptional technical executor with strong strategic and research taste. Passionate about mitigating risks associated with recursive self-improvement and driven by a desire to positively impact the future of AI development. Bonus points for prior work in ML research, AI alignment, or AI verification.
OpenAI
OpenAI is dedicated to ensuring that general-purpose artificial intelligence benefits all of humanity. We push the boundaries of AI systems and seek to safely deploy them through our products. We are an equal opportunity employer and comply with anti-discrimination laws. For additional information, please see our policies.
Compensation
The compensation range for this role is $295K - $445K.