Abstract
The rapid adoption of artificial intelligence (AI) and machine learning (ML)
technologies by federal agencies presents novel constitutional questions that
existing legal doctrine has not yet adequately addressed. This paper
examines the emerging constitutional dimensions of the 2026 conflict
between Anthropic and the Department of Defense (DOD), in which the
Pentagon demanded unrestricted use of Anthropic's AI model Claude for
domestic mass surveillance and autonomous weapons systems, ultimately
designating Anthropic a supply chain risk when the company refused.
Drawing on three salient constitutional frameworks, specifically the Fourth
Amendment doctrine established in Kyllo v. United States (2001) and
Carpenter v. United States (2018), the executive power limits articulated in
Youngstown Sheet and Tube Co. v. Sawyer (1952), and constitutional
scholar John Hart Ely's process-based theory of judicial review as
developed in Democracy and Distrust (1980), this paper further maps the
constitutional terrain of government AI adoption. Rather than prescribing
specific outcomes, this analysis identifies structural constitutional tensions
that AI-powered surveillance creates and argues that existing doctrinal
frameworks, while instructive, are severely insufficient to address the scale
and opacity of AI-enabled government power.
Recommended Citation
To, Alan
()
"Constitutional Inferences on Emerging Technologies: Analyzing Government Overreach, the Fourth Amendment, and Democratic Process in the Age of Artificial Intelligence,"
McNair Research Journal SJSU: Vol. 22
, Article 7.
https://scholarworks.sjsu.edu/mcnair/vol22/iss1/7