Matt Clifford has been forced to step down as chair of the UK government's Advanced Research and Invention Agency (ARIA) after taking a full-time role at Anthropic, following conflict-of-interest objections from senior members of Parliament1,2.
Clifford announced last week that he was joining Anthropic in a full-time capacity, leading the company's engagement with governments outside the United States, including the UK. He had, however, retained his leadership position at ARIA, the government body that funds frontier science and technology projects.
The dual role drew sharp criticism. Senior MPs called the arrangement a "clear conflict of interest," given that Clifford would simultaneously oversee public research funding and represent a commercial AI company in its dealings with the same government.
The pressure proved decisive. Clifford was forced to stand down from the ARIA chair role.
Anthropic, the San Francisco-based company behind the Claude chatbot, has been expanding its government-facing operations. Clifford's hire was part of that push, positioning him as the company's lead interlocutor with non-US governments.
ANALYSIS The episode exposes the tension between the UK's desire to attract AI-industry talent into policy roles and the governance constraints that apply when those individuals hold commercial positions at the companies their agencies may fund or regulate. ARIA's mandate to back frontier technology projects sits directly in Anthropic's competitive domain, making the overlap difficult to manage through recusal alone.
For Anthropic, the outcome removes a high-profile government liaison from a position of public authority but does not appear to affect Clifford's commercial role at the company. The net result is a cleaner separation between his policy influence and his corporate responsibilities.