Anthropic adds physicist Christopher Stubbs to its Long-Term Benefit Trust

Anthropic announced on October 21 that Harvard physicist Christopher Stubbs has joined its Long-Term Benefit Trust, the independent body built into the AI company’s governance structure. Stubbs is the first physicist to serve on the Trust. The appointment brings experience in experimental science, university leadership and national-security advice to a body responsible for considering AI’s long-term effects.

The Trust has authority to appoint a majority of Anthropic’s board and advises directors and company leaders on critical decisions, particularly those involving AI risks and societal impacts, according to the announcement. Stubbs joins chair Neil Buddy Shah, national-security specialist Richard Fontaine and former Federal Reserve chair Ben Bernanke. Anthropic says trustees are independent of its management and investors. The announcement appoints Stubbs as a trustee, a role distinct from serving as a company director.

Anthropic first set out the arrangement publicly in September 2023. Its design gave the Trust a special class of shares carrying rights to elect and remove board members, with those rights phasing in through time- and funding-related milestones. The structure was intended to give a body insulated from financial interests in Anthropic a formal place in decisions whose consequences could extend beyond the company’s customers and shareholders.

The original explanation reserved ordinary commercial management for the board and executive team. It described the Trust’s focus as longer-range issues, including incentives to evaluate future models for catastrophic risks and maintain strong security. Anthropic presented this as an experiment in corporate governance, with mechanisms for adjusting the arrangement. The appointment extends that existing structure rather than creating a new operating team or announcing a change in Claude’s capabilities.

Stubbs is Harvard’s Samuel C. Moncher Professor of Physics and of Astronomy. His research crosses cosmology, particle physics and gravitation, and he was part of the team that helped discover the universe’s accelerating expansion. He joined Harvard in 2003, chaired its physics department from 2007 to 2010 and served as Faculty of Arts and Sciences dean of science from 2018 to 2024. He has also worked on the Rubin Observatory and belongs to the JASON scientific advisory group.

His involvement with AI has included education as well as research. Harvard reported in March 2024 that he and Logan McCarty were teaching an undergraduate General Education course on generative AI’s technical, ethical, legal and philosophical implications. The course required no computer-science background and combined using the tools with examining their limitations and effects. That earlier work placed questions about how to use AI responsibly inside the university’s broader educational program.

The Trust appointment follows a separate October 18 announcement about Anthropic’s public-benefit funding. The company proposed dedicating 5% of its shares to a wholly owned fund, subject to stockholder approval, with potential increases tied to valuation milestones and board discretion. The proposed vehicle would allow different forms of giving, including support for nonprofit work and responses to labor-market disruption. Its financial role is separate from the Trust’s governance rights.

Anthropic’s appointment notice also highlights Stubbs’s work on reproducibility in AI-assisted science, where a model can answer the same question differently on different runs. The company links his selection to experience leading complex collaborations across academia, government and national laboratories. Within the Trust, that scientific perspective joins existing expertise in national security, economic policy and public-benefit institutions.