Safeguards Enforcement Analyst, Chem & Explosives Harms

Anthropic is hiring a Safeguards Enforcement Analyst focused on Chemical and Explosives Harms for a full-time, remote-friendly position based in San Francisco, New York, or Washington, DC. In this role, you'll be the technical authority guiding AI safety enforcement against weapons-related risks, investigating model interactions that could facilitate chemical or explosive harm.

The Opportunity

This position merges technical threat analysis with platform safety operations. You'll examine conversations between users and our AI systems to identify and investigate activities that could support chemical synthesis, precursor diversion, weapons engineering, or explosive development. You must rapidly differentiate between benign scientific inquiry and genuine weaponization attempts, making high-stakes judgment calls with incomplete information. Your enforcement work will prevent AI-enabled proliferation threats and strengthen our core safeguard architecture.

You'll own the complete enforcement process for this harm category: building detection systems, conducting investigations, improving workflows, and driving continuous optimization. You'll partner with Policy, Threat Intelligence, Data Science, and Engineering teams to scale automated risk identification and stay ahead of emerging threats within the broader chemical, biological, radiological, and nuclear (CBRNE) landscape.

Day-to-Day Responsibilities

• Investigate platform interactions and evaluate potential violations of chemical and explosives policies
• Design, build, and enhance the detection systems and review workflows specific to chemical and explosives threats
• Create scalable enforcement frameworks that balance precision with coverage across technically demanding content areas
• Identify emerging threat trends and recommend policy modifications or enforcement strategy adjustments
• Collaborate with data science and engineering teams to evaluate and improve detection models and automated systems
• Work with policy and threat intelligence teams to close knowledge gaps, map threat vectors, and strengthen risk assessment processes
• Respond to urgent enforcement situations and policy violations requiring immediate attention

Requirements

• Bachelor's degree in chemistry, chemical engineering, materials science, or equivalent professional background in a related field
• Track record in trust and safety, content moderation, or policy enforcement at platform scale, with demonstrated experience using AI tools to improve content review and enforcement workflows
• Experience designing or maintaining AI-powered dashboards or analytics systems for tracking safety metrics
• Ability to navigate complex, contradictory information and make sound, justified decisions under time constraints
• Proactive, independent approach to work: you spot problems, dig deeper, and deliver solutions without being prompted
• Strong writing ability and skill translating specialized chemistry topics for mixed audiences of technical and non-technical colleagues

Working at Anthropic

Anthropic is a public benefit corporation committed to AI safety and beneficial AI development. This role involves regular exposure to sensitive and potentially disturbing content, and carries shared on-call duties with the broader enforcement team. We maintain a 25% minimum office presence across our San Francisco, New York, or Washington, DC locations; certain roles may require more on-site time. We offer visa sponsorship and encourage applicants from all backgrounds.

Next Steps

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