Safeguards Enforcement Analyst – Conventional Weapons

About the role

Anthropic is seeking a Safeguards Enforcement Analyst focused on Conventional Weapons to help protect its AI systems from misuse that could facilitate real‑world harm. You will design and operate detection and enforcement workflows, evaluate model behavior, and stay ahead of emerging threats involving weapons and dangerous technology.

Key responsibilities

• Design and architect automated enforcement systems and review workflows that scale while maintaining high accuracy.
• Develop and maintain evaluation suites that measure model performance on weapon‑related policy areas, surface regressions, and inform policy and model improvements.
• Partner with Engineering and Data Science to optimize detection and automated enforcement for potential policy violations.
• Review flagged content, make enforcement decisions, and identify policy gaps, especially for novel or technically sophisticated misuse attempts.
• Provide structured feedback to the Safeguards policy team on enforcement ambiguities and emerging weapon trends.
• Create and update enforcement guidelines and reviewer documentation to ensure consistent decision‑making.
• Monitor regulatory changes, emerging weapons trends, and AI policy best practices, applying them to workflows and evaluations.
• Identify and escalate emerging misuse patterns, novel attack vectors, and signs of coordinated violent activity.

Required profile

• Deep, applied expertise in weapons systems with the ability to translate complex technical evidence into enforcement decisions.
• Experience in policy enforcement, threat intelligence, counter‑terrorism, government, or a related field involving harmful content or dangerous technology.
• Proven track record of building and scaling policy enforcement or content‑review workflows.
• Proficiency in SQL or other data‑analysis tools to extract insights from large datasets and monitor workflow health.
• Ability to identify emerging risks and communicate findings to cross‑functional stakeholders.

Required skills

• SQL