Anthropic Fellows Program 2027: AI Safety and Security Research Fellowship

The Anthropic Fellows Program offers a four-month, full-time research fellowship for researchers and engineers to work on AI safety and security projects with mentorship and financial support. Fellows conduct empirical AI research, often producing public outputs such as academic papers, and may have access to research funding and computing resources. The program supports both AI safety and AI security tracks and connects participants with a broader research community.

Weekly stipends; direct mentorship from Anthropic researchers; research computing funding; access to shared workspaces; opportunities for future positions; involvement in AI safety and security research communities.

Applicants should have a strong technical background in computer science, mathematics, physics, or related disciplines, and be available for full-time participation. Applicants must have full-time work authorization in the United States, United Kingdom, or Canada and be able to reside in their authorized country throughout the programme.

Submit an application through the Constellation-managed process linked from the official apply page; include a CV and relevant research experience.