Safeguards Enforcement Analyst, Cyber Harm

Anthropic, an AI safety-focused company, is hiring a full-time Safeguards Enforcement Analyst specializing in Cyber Harm for positions in San Francisco, New York City, or Washington, DC—with remote work options available.

About the Role

This position protects Anthropic's AI systems from malicious cyber operations. You will analyze flagged accounts and interactions to prevent threat actors from exploiting our technology for cyberattacks, malware engineering, and similar harmful activities. Your work bridges safety policy and technical threat detection, ensuring our users can trust and build confidently on top of Anthropic's platforms.

Core Duties

• Assess flagged content and user accounts to render sound, fully documented policy compliance decisions
• Identify and counter attempts to weaponize AI systems for cyber operations, including attack infrastructure development and exploit creation
• Route intricate, uncertain, or serious cases to the appropriate teams for specialized handling
• Share insights with our policy development group about policy limitations discovered during enforcement work
• Collaborate with engineers and data scientists to flag detection gaps and share quality metrics to improve system effectiveness
• Maintain consistency and high accuracy rates across review operations
• Monitor emerging enforcement trends, threat actor methods, and evolving cyber risk landscapes to guide daily decisions

What You Bring

• A track record in cybersecurity involving offensive methodologies, exploit development, malware analysis, or security research
• Prior involvement in content review, abuse investigation, or policy-driven enforcement at meaningful scale
• Hands-on competency with SQL and/or Python for analytical work and threat detection
• Capability to identify emerging threats and communicate findings across Product, Policy, Engineering, and Legal teams
• Working knowledge of generative AI systems, including prompt engineering for review and enforcement
• Preferred: Experience in trust and safety operations, cyber investigation, or threat analysis within technology or AI companies; understanding of large language model capabilities and potential misuse vectors; familiarity with abuse monitoring platforms; knowledge of large-scale policy implementation and content review challenges; exposure to government partnerships or regulated sector work

Why Join Anthropic

Anthropic pursues the development of AI systems that are trustworthy, interpretable, and responsive to human oversight. Our multidisciplinary team of researchers, engineers, policy specialists, and leaders collaborates intensively on AI safety initiatives, prioritizing transformative impact over incremental advances. We embrace empirical, science-led approaches and foster an environment where strong communication and diverse perspectives drive our most important work.

How to Apply

To apply, complete your application directly on this page, or you'll be redirected to the employer's application platform to finish submitting there.