Frontier AI Safety Researcher

OpenAI, Inc.; City Of London; On-site; GBP 89,000 - 142,000; Full time

Job summary

OpenAI seeks exceptional researchers to push the frontier of safety mitigations and derisk frontier models by developing novel techniques from interpretability, control, and alignment. You will help define how a safe AI system should look in the future at OpenAI and play a pivotal role in building and deploying safe AGI.

This role requires deep technical expertise and collaboration across domains such as misalignment, cybersecurity, and biology to create a scalable, end-to-end safety stack.

Qualifications

• 2+ years in AI safety, especially in RLHF, interpretability, or control.
• Ph.D. or related degree in CS/ML requested.
• 4+ years of research engineering experience desired.

Responsibilities

• Identify emerging AI safety risks and mitigations.
• Develop evaluations to assess risk impact with domain experts.
• Define research directions for safer, more aligned AI systems.
• Contribute to best practices for AI safety across OpenAI and industry.
• Design red-teaming pipelines to test end-to-end safety robustness.

Skills

AI safety

RLHF

Interpretability

Robustness

Control

Education

PhD or equivalent in CS/ML

Tools

Python

Job description

OpenAI seeks exceptional researchers to push the frontier of safety mitigations and derisk frontier models by developing novel techniques from interpretability, control, and alignment. You will help define how a safe AI system should look in the future at OpenAI and play a pivotal role in building and deploying safe AGI.

This role requires deep technical expertise and collaboration across domains such as misalignment, cybersecurity, and biology to create a scalable, end-to-end safety stack.