Prof. Alan Rozenshtein Takes Leave to Join Anthropic’s Rule of Law Team

Alan Rozenshtein, associate professor of law, has taken a leave of absence from the University of Minnesota Law School to join Anthropic’s Rule of Law team as a member of its technical staff within the Anthropic Institute.

The Rule of Law team brings together experts and scholars of law and ethics to examine what increasingly powerful artificial intelligence could mean for executive power, courts, democratic institutions, elections, and public deliberation—and to share that research with the broader public. Rozenshtein’s work at Anthropic will build directly on his scholarship at Minnesota Law, where he has studied the intersection of technology, constitutional law, national security, and government power.

Rozenshtein’s work has examined how legal institutions should respond as technology changes the way government and private companies exercise power. More recently, his research has turned directly to artificial intelligence, including his article “Untangling AI Openness” (Wisconsin Law Review 2026), which examines what it means for AI systems to be “open” and the competing interests—including innovation, safety, national security, and public oversight—that shape debates over AI policy.

“The opportunity for Alan to bring his expertise in law and artificial intelligence directly to Anthropic’s Rule of Law team is a testament to the enormous significance of his scholarship,” says Dean William McGeveran. “ We are proud of his contributions to our faculty and excited to see how the new insights and context he gains will inform his scholarship when he returns.”

As governments and AI companies negotiate the boundaries of AI’s use in national security and public life, the legal and ethical issues have become increasingly urgent. Rozenshtein has been closely engaged with the legal issues surrounding Anthropic itself. In early 2026, he analyzed the Pentagon’s designation of Anthropic as a national security “supply-chain risk” after the company refused government demands to permit the use of Claude for mass surveillance of Americans and fully autonomous weapons. Rozenshtein questioned the legal basis for the designation, noting that the statute the Pentagon appeared likely to rely on was intended for foreign companies and had apparently never been used in this way against a U.S. company.

His move to Anthropic brings this scholarship into direct conversation with the development and governance of powerful AI. At a moment when AI is raising fundamental questions about the separation of powers, the authority of the executive branch, the role of courts, the resilience of democratic institutions, and the future of public discourse, Rozenshtein sits at the center of a rapidly developing field: understanding how law and legal institutions can help shape the relationship between artificial intelligence, government, and society.