Researcher for Recursive AI Safety & Alignment

openai; General Trias; On-site; PHP 1,200,000 - 1,800,000; Full time

Job summary

OpenAI Preparedness is seeking strong technical executors to anticipate and mitigate future misalignment risks in AI development. This role spans pre-deployment risk assessment, control measures, RSI-relevant training interventions, and translating insights into formal institutional practices and external communications.

The work is urgent and fast-paced, involving design, testing, and deployment of scalable safety solutions, with collaboration across engineering and policy teams to guide

Qualifications

• Exceptional technical executor.
• Strong strategic and research taste.
• Passionate about mitigating risks associated with recursive self-improvement.

Responsibilities

• Carefully consider the problems OpenAI might face in the future and how to prepare for them.
• Turn an open-ended objective like “prepare for future misalignment threats” into a concrete direction (e.g. “stress-test monitors for scheming”) and prioritize the work.
• Execute quickly, build scrappy prototypes, and iteratively improve them into established safety components.
• Secure buy-in from other staff at OpenAI when necessary and communicate your work clearly.
• Collaborate with or manage other staff as needed to scale efforts quickly.

Job description

OpenAI Preparedness is seeking strong technical executors to anticipate and mitigate future misalignment risks in AI development. This role spans pre-deployment risk assessment, control measures, RSI-relevant training interventions, and translating insights into formal institutional practices and external communications.

The work is urgent and fast-paced, involving design, testing, and deployment of scalable safety solutions, with collaboration across engineering and policy teams to guide