Head of Policy Design, Societal Harms

Highlights

• Lead policy design for child safety, well-being, manipulation and elections.
• Define what Claude can and cannot do and build mitigations with research.
• Partner with product to evolve safety layers as models and risks advance.

What they offer

• Competitive compensation and benefits
• Equity donation matching
• Generous vacation and parental leave
• Flexible working hours

Ideal candidate

Experienced policy leader with deep consumer harms expertise and fluency in frontier AI development who thrives collaborating across teams and making sound calls in ambiguous, high-stakes decisions.

Requirements

• Experience leading teams including managing managers in AI safety or product policy
• Deep applied familiarity with consumer harms like child safety, well-being and election integrity
• Track record of cross-team collaboration with distributed ownership
• Working understanding of frontier model training, evaluations and launch
• Experience translating policy into enforceable, measurable mechanisms
• Sound judgment in ambiguous, high-consequence decisions
• Bachelor's degree or equivalent combination of education and experience