Enforcement Analyst, Agentic Alignment

Details

Anthropic is an AI safety and research company building reliable, interpretable, and steerable AI systems. The Enforcement Analyst focused on Agentic Alignment will identify and investigate harmful or unexpected model behavior and make enforcement decisions. The role works cross-functionally with Policy, Threat Intelligence, Engineering, Research, and Transparency teams.

This position may involve exposure to explicit sexual, violent, or psychologically disturbing content. It also includes on-call responsibility across the Policy and Enforcement teams.

Responsibilities

• Review and investigate potential instances of agentic misalignment using transcripts, tool-use traces, and code.
• Triage cases by severity and urgency and share them through the appropriate channels.
• Help operationalize review pipelines for agentic cases by translating detection outputs and investigation leads into scalable, structured review workflows.
• Make enforcement decisions under the Usage Policy, applying it consistently to novel agentic scenarios and across surfaces, and document the reasoning behind each decision.
• Assess cyber-related agentic activity, including scanning, exploitation attempts, credential misuse, and malicious code, to determine its potential for real-world harm.
• Write clear, well-evidenced case reports for audiences ranging from enforcement peers to senior leadership.
• Identify patterns and trends across cases and feed enforcement findings back to policy, detection, and research teams to improve policies, classifiers, and model behavior.
• Maintain rigorous documentation of enforcement decisions, review logic, and precedents.
• Collaborate with Threat Intelligence, Research, Engineering, and Policy teams to keep detection coverage aligned with evolving agentic misuse and misaligned behavior.
• Use AI systems, including Claude, to accelerate investigations by summarizing long transcripts, analyzing code, querying data, and building lightweight tooling.
• Contribute to regular reports on agentic harm trends and enforcement outcomes.

Requirements

• Ability to read and understand code, such as Python, JavaScript, or shell scripts, well enough to determine what happened and why it matters.
• Background in security, cyber threat intelligence, or a similar technical field that provides an understanding of how cyber harm occurs and how to assess its severity.
• Experience in trust and safety, enforcement, investigations, or a related field, translating ambiguous policy questions and facts into defensible enforcement decisions and clear written findings.
• Experience conducting structured investigations from the first signal through the final finding and determining when evidence is sufficient to act.
• Experience supporting or contributing to formal enforcement actions, including case documentation, evidence packaging, and escalation coordination.
• Strong clear and concise writing skills, with the ability to present complex findings to Legal, Policy, and Engineering stakeholders, including non-technical audiences.
• Sound judgment about when a case requires escalation and resilience when working with difficult content.

Preferred Qualifications

• Demonstrated use of AI systems to accelerate work and interest in developing additional applications for them.
• Comfort working with large volumes of data, including long transcripts, logs, traces, and complex multi-table datasets, and surfacing meaningful behavioral patterns.
• Curiosity about AI agent behavior and willingness to investigate uncategorized behavior and build approaches where established playbooks do not exist.
• Familiarity with AI safety, alignment, and current debates about the risks of increasingly autonomous AI systems.

Education and Logistics

• Minimum education: Bachelor’s degree or an equivalent combination of education, training, and experience.
• Required field of study: A field relevant to the role, demonstrated through coursework, training, or professional experience.
• Minimum years of experience: Requirements correlate with the internal job level.
• Location-based hybrid policy: Staff are expected to work from one of Anthropic’s offices at least 25% of the time, although some roles may require more office time.
• Anthropic explicitly states that it sponsors visas, although sponsorship may not be available for every role or candidate, and provides immigration lawyer support when making an offer.

Benefits

Anthropic offers competitive compensation and benefits, optional equity donation matching, generous vacation and parental leave, flexible working hours, and office space for collaboration.