Safeguards Policy Analyst, Cyber Harms

Anthropic’s mission is to create reliable, interpretable, and steerable AI systems. We want AI to be safe and beneficial for our users and for society as a whole. Our team is a quickly growing group of committed researchers, engineers, policy experts, and business leaders working together to build beneficial AI systems.

We're looking for an analyst to support the team's cyber product policy work: the usage policy language, help-center and enforcement guidance, and launch policy notes that say what Anthropic permits and prohibits for cyber-relevant use, kept consistent with the constitution and our access tiers. You'll analyze the constitution and usage policies to check that enforcement matches them, work with the threat intelligence and enforcement teams to make sure our safety standards are met, and coordinate cyber policy inputs to model releases and regulatory requirements alongside the Senior Cyber Policy Lead. You'll also help maintain our controlled-access framework and help keep our cyber-related policy commitments current as standards shift.

You don't need to write code in this role, but you should be comfortable engaging deeply with technical material — probes, classifiers, and trusted-access programs — and translating it into policy that holds up under scrutiny.

• Contribute to the team's cyber product policy artifacts (usage policy language, help-center and enforcement guidance, launch policy notes) and help keep them consistent with the constitution and access tiers

• Analyze Anthropic's constitution and cyber-related usage policies, and check that enforcement decisions and safeguards match what they say; maintain a running gap log and propose text fixes

• Work with the threat intelligence and enforcement teams to ensure cyber safety standards are met; review a sample of enforcement decisions against policy text on a regular cadence and report drift

• Coordinate cyber policy inputs to model releases and to regulatory requirements: prepare the cyber policy section of launch/model-card reviews and regulator pre-briefs, in support of the Senior Cyber Policy Lead

• Support Anthropic's access-requirements policy and contribute to consolidation of the controlled-access framework across existing and emerging access programs

• Help keep Anthropic's cyber-related policy commitments current with shifting industry and regulatory standards, and ensure they map cleanly to our technical safeguards

• Draft inputs to reporting for partners and regulators

• Help translate technical evaluation and safeguard work into policy positions and internal guidance

• Engage closely with technical teams to understand the capabilities of probes and classifiers relevant to access policy

• Demonstrated ability to write clear policy analysis (professional or academic)

• Familiarity with how AI developers or platforms enforce usage policies, or with cybersecurity policy, standards, or regulatory frameworks (e.g., usage policies/AUPs, trust and safety enforcement, NIST CSF, CVD)

• Ability to read and interpret technical security material, such as vulnerability reports or threat assessments

• Ability to work across teams (threat intelligence, enforcement, engineering, policy) and communicate clearly in writing

• Experience in cybersecurity policy, including familiarity with coordinated vulnerability disclosure

• Exposure to government information-sharing and incident-notification