About the Role
As Technical Policy Lead for Government and Third-Party Safety Partnerships, you'll advance Anthropic's safety partnerships with governments and the broader third-party ecosystem. This is a hybrid role that is technically grounded but drives policy impact – strengthening state capacity, building a wider ecosystem for AI safety, and developing shared standards, in order to advance Anthropic’s mission to ensure the world safely makes the transition through transformative AI.
You'll serve as the critical bridge between Anthropic's Policy, Frontier Red Team, Alignment, Safeguards, and Legal teams on one side, and government safety institutes and third-party evaluators on the other. You'll manage complex multi-stakeholder engagements spanning pre-deployment evaluations, collaborative safety research, information-sharing agreements, and regulatory harmonization efforts. As AI capabilities advance, these partnerships will be essential for providing trusted external validation of our safety efforts.
Working within our International Policy team, you'll have significant autonomy to shape these partnerships. We're looking for someone who can be both a trusted partner to external counterparts and a thoughtful steward of Anthropic's resources and priorities - someone who brings clear judgment about where to focus our efforts for maximum impact. This role exists to ensure our government and third-party safety partnerships are strategically coherent, technically sound, and structured in ways that scale globally.
In this role you will:
-
Identify collaboration opportunities that advance state capacity on AI safety, strengthen the broader safety ecosystem, and support Anthropic policy objectives for ASL-4+ readiness and standards harmonization
-
Develop frameworks for engaging with emerging government safety organizations, to promoting a coherent global approach that leverages each country's unique capabilities
-
Evaluate and prioritize inbound requests from external safety organizations, ensuring Anthropic's engagement is focused on high-impact collaborations
-
Own strategic alignment as well as day-to-day coordination with US CAISI, UK AISI, and other safety organizations relevant to policy, serving as the primary operational point of contact
-
Manage research and testing collaborations covering pre/post-deployment evaluations, safeguards testing, and alignment research
-
Coordinate between internal teams, in particular Policy, Frontier Red Team, Alignment, Safeguards, Legal, and Communications to facilitate external safety partnerships
-
Support Legal in negotiating and maintaining collaboration agreements and MOUs
-
Manage technical access provisions in accordance with agreement terms
-
Track external safety partnerships and provide timely updates to internal stakeholders
-
Position external safety partnerships to provide decision-relevant information for internal safety teams and processes, such as in safety cases, and enable trusted independent evaluation and validation of our models and processes
-
Represent Anthropic at external conferences and speaking engagements
You may be a good fit if you:
-
Have direct experience with government or third-party safety organizations. You may have previously worked with or at US CAISI, UK AISI, METR, Apollo, or equivalent organizations in research, technical program management, or operations roles. You understand how these organizations operate from the inside.
-
Have a technical background in AI safety-relevant areas. You should have sufficient technical experience to engage credibly with evaluation methodologies, understand alignment and safeguards concepts, and partner with internal technical staff to solve problems. You should also have familiarity with safety standards such as our Responsible Scaling policy and the EU Code of Practice.
-
Excel at multi-stakeholder coordination. You should have a proven track record managing complex programs that span research, policy, legal, and communications teams with differing priorities.
-
Are comfortable making judgment calls in ambiguous situations. You can synthesize incomplete information from multiple teams, identify blockers, and drive decisions when strategic direction is evolving.
-
Exercise sound judgment on prioritization. You can focus effort where it matters most and manage stakeholder expectations effectively, with clarity and directness.
-
Communicate effectively across contexts. You are equally comfortable briefing executives on partnership strategy, explaining technical evaluation plans to policy teams, and managing day-to-day coordination with external researchers and engineers.
-
Understand government contexts. You are familiar with how government agencies operate, including considerations around information security, inter-agency coordination, and policy development processes.
This role can be based in San Francisco, Washington DC, or London.
