Safeguards Enforcement Analyst, Chem & Explosives Harms

As an Enforcement Analyst focused on Chem & Explosives Harms, you will play a critical role in protecting against the misuse of AI systems for chemical and explosives harms. You will enforce our Usage Policy with a specific focus on detecting and mitigating these risks, investigating potential violations, and help continuously strengthen our safeguards. The work sits at the intersection of chemical and explosives threat analysis and platform enforcement: you will read real model interactions and make fast, well-reasoned calls about whether activity is benign research or a credible attempt at harm.

• Enforce Usage Policies with a specific focus on detecting and mitigating potential chemical and explosives risks and harmful use of AI systems.
• Take ownership of enforcement monitoring workflows for the chemical and explosives harm area, improving end-to-end detection, investigation, triage, and escalation processes.
• Monitor and analyze platform activity to identify emerging patterns related to chemical and explosives threats (within the broader CBRNE landscape) that may require policy updates or enforcement action.
• Design and architect automated enforcement systems and review workflows that scale effectively while maintaining high accuracy across a technically complex content surface.
• Conduct thorough investigations of potential violations, gathering and documenting evidence to support enforcement decisions.
• Proactively surface trends and propose improvements to detection methods and review workflows without waiting to be directed.
• Partner with Engineering and Data Science teams to optimize detection models and automated enforcement systems for policy violations.
• Partner with Policy and Threat Intelligence teams to understand potential exploits and contribute to risk-assessment frameworks, and partner with engineers iterating on safety systems.
• Provide enforcement-grounded feedback on policy gaps, and handle escalations and time-sensitive situations related to chemical and explosives policy violations.

• Hold a degree in a chemistry-related field (e.g., chemistry, chemical engineering, materials science) and/or relevant professional experience in a related field.
• Possess experience in Trust & Safety, content moderation, or policy enforcement at platform scale, including working with generative AI tools to refine and optimize content review and enforcement workflows.
• Possess experience in utilizing AI tools to develop data dashboards for metrics collection in support of continuous improvement efforts.
• Can analyze complex, ambiguous situations and make well-reasoned, defensible decisions under time pressure.
• Are proactive and self-directed: you spot trends, dig in, and ship improvements on your own initiative.
• Communicate clearly in writing and can translate technical chemistry concepts for diverse audiences, including both technical and non-technical stakeholders.

• Subject matter expertise in chemical and explosives defense, security, WMD/CBRNE non-proliferation, or threat-intelligence.
• An understanding of where real-world chemical and explosives risk actually lies — adversary intent, acquisition and weaponization pathways, and dual-use concerns, including familiarity with cross-platform threat analysis and open-source intelligence (OSINT) techniques relevant to weapons of mass destruction.
• Proficiency in SQL and/or other data analysis tools to draw insights from large datasets and monitor enforcement workflow health.
• Familiarity with the Chemical Weapons Convention (CWC), scheduled chemicals, precursor controls, and energetic-materials safety frameworks.

Annual Salary: $245,000—$285,000 USD. Anthropic offers competitive compensation and benefits, optional equity donation matching, generous vacation and parental leave, flexible working hours, and a lovely office space in which to collaborate with colleagues.

Apply via the provided Greenhouse link. Anthropic recruiters only contact you from @anthropic.com email addresses. Be cautious of emails from other domains. Legitimate Anthropic recruiters will never ask for money, fees, or banking information before your first day.