Anthropic opens residency paid up to $500,000 a year equivalent, no ML background required; applications close Nov. 22

Anthropic has opened applications for a paid research residency that does not require a PhD or machine-learning experience. Pay is equivalent to $350,000-$500,000 a year, and applications close Nov. 22.

According to Anthropic's alignment blog, the program targets "established technical leaders" such as professors in quantitative fields, technical founders and staff-level engineers. "You don't need a PhD, prior ML experience, or published papers," it says, adding it does not expect residents to arrive as ML experts. Placements run six weeks, three months or six months, with pay prorated to length. Residents get a mentor, regular one-on-ones, self-study time and a dedicated compute budget to work with frontier models. San Francisco is preferred; shorter placements can be remote.

Research areas are AI safety and alignment (scalable oversight, adversarial robustness and control, interpretability and more), AI security and frontier red-teaming, and ML systems and reinforcement learning. The aim is to move experts from other fields into full-time AI safety research. Crypto Briefing reported offers are expected around January. The number of slots was not disclosed.

The news drew about 2,400 likes on X and about 1,500 on a Polymarket repost. The original post framed it as the talent war going from "we're hiring" to "we'll pay professors to pivot into AI safety." Applications are split between US and UK job pages, and Anthropic said applicants may also be considered for its Fellows Program.