AI Safety Fellow

• In this fellowship, you'll conduct empirical AI safety research for four months with the goal of producing a publishable paper.
• Work with mentors from Anthropic's research team on projects aligned with their research priorities.
• Use external infrastructure like open-source models and APIs to implement and test ideas.
• Collaborate within the broader AI safety research community in shared workspaces.
• Produce public research outputs that advance understanding in areas like scalable oversight or mechanistic interpretability.

• In this fellowship, you'll conduct empirical AI safety research for four months with the goal of producing a publishable paper.
• Work with mentors from Anthropic's research team on projects aligned with their research priorities.
• Use external infrastructure like open-source models and APIs to implement and test ideas.
• Collaborate within the broader AI safety research community in shared workspaces.
• Produce public research outputs that advance understanding in areas like scalable oversight or mechanistic interpretability.

Anthropic is a frontier AI research and product company, with teams working on alignment, policy, and security. We post specific opportunities at Anthropic that we think may be high impact. We do not necessarily recommend working at other positions at Anthropic. You can read concerns about doing harm by working at a frontier AI company in our career review on the topic.