Lead Researcher, AI Offense-Defense Dynamics

Core responsibilities:

• Lead research on offense-defense dynamics of advanced AI systems, examining how AI technologies influence societal safety or amplify risks.
• Develop quantitative system dynamics models capturing technological, social, and institutional factors affecting AI risk landscapes.
• Expand our offense/defense dynamics taxonomy, creating metrics to predict whether AI features favor offensive or defensive applications.
• Design analytical models to identify policy intervention points that could shift offense-defense balances toward safer outcomes.

This role creates positive change by advancing understanding of AI offense-defense dynamics to inform policies and interventions that promote safer AI development and reduce societal risks from advanced AI systems.

Required skills, experience, qualifications:

• Experience in AI safety, risk management, or related research fields.
• Strong quantitative modeling and system dynamics analysis skills.
• Ability to develop taxonomies and metrics for complex technological systems.
• Policy analysis and intervention design capabilities.

No compensation, perks, or culture highlights mentioned in the posting.

The Center for AI Risk Management and Alignment focuses on managing risks and ensuring alignment of advanced AI systems with societal safety and beneficial outcomes.