AI Policy Jobs

Interdisciplinary role focused on abuse monitoring, misuse investigation, and safety enforcement for OpenAI products, with policy and engineering collaboration to build risk controls and mitigations.

Role focused on model policy for frontier cyber risk, including threat modeling, behavioral specifications, red-teaming analysis, and mitigations for harmful AI-enabled cyber activity.

Frontier AI safety/policy role focused on chemical and biological risk, model policies, and risk taxonomies for training and evaluation.

Safety Systems role focused on defining and maintaining model policies, evaluation criteria, and safeguards for frontier AI model behavior in high-risk contexts.

Full-stack engineer role on OpenAI's Safety Systems team building internal tools and platforms to improve model safety, risk measurement, safeguards, and policy workflows for production AI systems.

Safety operations role focused on child safety, content enforcement, and improving Grok's defenses through policy enforcement, risk assessment, and LLM red-teaming.

Research Scientist role on Scale Labs' Safety Post Training team developing post-training and interpretability methods to improve frontier AI safety, robustness, alignment, and safety evaluations, with some policy-facing collaboration.

Research Scientist role focused on frontier AI risk evaluations, including building evaluation harnesses and datasets, testing dangerous capabilities, and publishing findings for policymakers.

Technical program manager driving security programs for Anthropic's Responsible Scaling Policy readiness, coordinating security, compliance, infrastructure, research, and policy work to protect sensitive AI assets.

The Technical CBRN-E Threat Investigator will detect and investigate the misuse of AI systems for CBRN-E threats, focusing on developing detection capabilities and collaborating with policy teams.

Policy analyst for Anthropic’s fraud and scams safeguards, responsible for enforcement policies, threat modeling, and classifier-guided detection on AI products and APIs.

The Safeguards Enforcement Analyst will ensure AI models meet safety and policy standards through evaluations and mitigations, collaborating with cross-functional teams.

Technical research role on Anthropic's Frontier Red Team focused on evaluating and defending against advanced AI-enabled cyber threats, including autonomous capability evaluations, safety defenses, and policy-relevant demonstrations.

4-month Anthropic Fellows research role in Economics & Policy focused on empirical research about AI's economic effects, societal impacts, and policy-relevant recommendations.