Technical AI Policy Researcher, Frontier Risk - Trust and Safety
Core
Define, measure, and mitigate safety and fairness risks in frontier GenAI models through policy frameworks, model risk assessments, and upstream policy solutions.
Role type
Senior IC technical AI policy researcher (frontier risk)
Builds
Multimodal GenAI policies, behavioral specifications, evaluation criteria, grading guidance, and system-level safeguards for frontier AI systems.
Domain
Generative AI safety, frontier risk, AI governance, and regulatory compliance
Deliverable
production ML models
Required skills
Frontier risk research, AI evaluations, red-teaming, AI governance, LLM/multimodal model behavior analysis, policy artifact creation, end-to-end policy workflow design, regulatory compliance expertise
Preferred skills
High efficiency on-device architectures, multimodal understanding, V&V of AI systems, RAG, global south cultural context, entertainment media technology sector knowledge
Technologies
LLMs, multimodal models, genmedia models, red-teaming frameworks
Responsibilities
Design multimodal GenAI policies for safety-relevant domains including agentic modalities and loss of control; Translate risk models into behavioral specifications and evaluation criteria; Define boundaries between beneficial AI use and potential harm; Build policy artifacts supporting model training, evaluation, and deployment; Design end-to-end policy workflows from pre-launch evaluation to post-launch monitoring; Use red-teaming and deployment data to improve policy quality; Identify emerging capability areas creating new safety challenges; Monitor post-launch model activity for policy gaps; Champion research to strengthen policy defensibility; Combine long-horizon safety research with hands-on deployment work; Support regulatory teams on AI compliance initiatives
Seniority
Senior, hands-on IC
