The posting, in Anthropic's own words
archived Oct 19, 2026About Anthropic
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.
About the role
As a Program Manager on Anthropic's Safeguards team, you'll help make sure we meet our obligations under AI safety laws and regulations around the world. Working closely with Legal, you'll turn new and emerging requirements into programs, controls, and deliverables. This requires a detailed understanding of how our models and products are built, evaluated and safeguarded. Your work will support Anthropic's commitment to developing AI safely and beneficially as the regulatory landscape evolves. The responsibilities listed below are some examples of areas that someone in this role could own, depending on their experience and the team's needs.
Read the full posting ↓
Key responsibilities
Manage and drive complex safety initiatives that identify, assess and mitigate risks from frontier models, from initial risk assessment through to the deployment of safeguards Partner with the Safeguards incident management team to strengthen incident response and reporting processes, including the policies that determine how incidents are escalated and when they must be reported externally Conduct post-incident reviews, and work with the subject matter experts for each harm area to incorporate the findings into model and product safeguards Own the safety strategy and deliverables for AI systems used in high-risk settings, including risk assessments, technical documentation, and the information that deploying customers need to use those systems safely Define the safety and risk controls that our models and products require, ensure that each control has an owner and a documented procedure, and confirm that it operates as intended Partner closely with the owners of each risk area to develop a deep understanding of AI risks, safety evaluations and mitigation strategies, and use that understanding to prepare safety documentation, transparency reports and regulatory submissions
Minimum qualifications
Experience in trust and safety, AI safety or a related field, including designing and implementing safeguards against misuse or harm Understanding of AI safety, including the principal risk areas and the evaluations and safeguards used to address them, well enough to engage substantively with the technical experts responsible for them Experience running risk assessment or harm mitigation programs from identification through to resolution Experience translating policy or regulatory requirements into practical operational processes Ability to drive complex, cross-functional programs to completion in a fast-moving environment, including with technical teams outside your own discipline Clear written communication, including documentation for technical, legal and external audiences Strong attention to detail, and the judgment to recognize when work is at risk and raise it early Commitment to the safe development and deployment of AI