Model Policy, Frontier Cyber Risk
Core
Define and enforce AI safety policies for high-risk cybersecurity capabilities, translating threat models into behavioral specifications and evaluation criteria for frontier AI systems.
Role type
Senior IC AI Safety Policy Engineer (Cybersecurity)
Builds
Policy frameworks, threat models, taxonomies, evaluation criteria, and behavioral specifications for AI models in cybersecurity contexts.
Domain
AI Safety + Cybersecurity (Offensive/Defensive Security, Threat Intelligence, Malware Analysis)
Deliverable
production ML models | product features | dashboards & analysis | research | client delivery | infrastructure | physical/clinical work
Required skills
Offensive security, defensive security, vulnerability research, malware analysis, incident response, threat intelligence, application security, exploit development, infrastructure security, cloud security
Preferred skills
Judgment on AI impact on cyber threat landscape, distinguishing legitimate security use cases from harmful assistance, building threat models for adversarial systems, translating technical expertise into policy frameworks, using empirical evidence for policy decisions, systems thinking across policy/evaluations/training/monitoring, cross-functional collaboration, written communication of complex security concepts
Technologies
N/A
Responsibilities
Design and maintain model policies for cybersecurity and frontier-risk domains; Translate cybersecurity threat models into behavioral specifications and evaluation criteria; Define boundaries between legitimate security research and harmful activity; Build policy artifacts for training, evaluation, deployment, and monitoring systems; Partner with researchers and engineers to operationalize policies; Analyze red-teaming results and model failures to improve policy quality; Identify emerging cyber capability areas for misuse; Contribute to system cards and safety reports
Seniority
Senior, hands-on IC