Lead a cross-functional team defining actionable technical best practices for frontier AI safety, shaping long-term priorities, engaging developers and standards bodies, and driving adoption across the industry.
Develop authoritative technical best-practice documents and accessible explainers on frontier AI safety, working with researchers and stakeholders to make guidance accurate, evidence-based and useful to developers.
Lead and contribute to rigorous research on how frontier AI systems influence people, designing studies and evaluations, publishing findings, and helping shape research priorities on high-stakes AI safety questions.
Test and exploit vulnerabilities across AI evaluation sandboxes, cloud, identity and application infrastructure, while building agent-assisted red-teaming tools and helping teams harden systems.
Build and secure tools, platforms, and infrastructure for frontier AI security research, including evaluation systems, model hosting, secure access, and researcher workflows.
Research Director expression of interest for leading advanced AI security research, shaping research priorities, and informing how frontier AI is governed and deployed.
Evaluate frontier AI capabilities and risks in cybersecurity, autonomy, and AI R&D by building realistic assessments, testing models before deployment, and investigating novel model behaviours.
Contribute scientific and technical research to evaluate and mitigate chemical and biological risks from frontier AI, developing assessments, benchmarks, safeguards and recommendations for government and industry.
Build and own reliable Python platforms, tooling, and infrastructure that enable chemical and biological security researchers to run AI experiments, evaluate safeguards, and deliver results.
Page 1 · 9 results
