Research Scientist, AI Safety

Job Summary

To drive innovative AI safety research, the full-time Research Scientist, AI Safety will accelerate the team's agenda by developing and evaluating white-box methods, publishing findings, and engaging with the AI alignment community in a hybrid work environment.

Key responsibilities

• Develop and evaluate methods leveraging model internals to enhance AI safety
• Publish research findings and actively participate in the AI alignment community
• Propose new research directions and collaborate with national AI safety institutes and frontier model developers

Required qualifications

• Hands-on experience applying at least one white-box method to a real model
• A track record in AI safety through publications or relevant projects
• Experience in evaluations, AI control, red-teaming, or reinforcement learning
• A PhD or several years of research experience in computer science, machine learning, or a related field
• Previous experience in applied ML for fields such as biology, chemistry, or robotics