Publications: Daniel Castro

To protect individuals while preserving the open information ecosystem that supports innovation, policymakers should focus on outputs rather than training inputs, encourage transparency norms for autonomous AI agents, and create a safe harbor for responsible use of publicly available data.

Center for Data Innovation Director Daniel Castro sent a letter to Maryland Senate Finance Committee Chair Pamela G. Beidle, Vice Chair Antonio L. Hayes, and members of the committee in opposition to Senate Bill 889.

In Tech Policy Press, Daniel Castro argues that a dispute between the U.S. Department of Defense and Anthropic over military AI use underscores the need for Congress—not executive pressure or private contracts—to set clear statutory guardrails for deploying AI in national defense.

As Daniel Castro writes in Government Technology, a new Information Technology and Innovation Foundation analysis finds a sharp “digital marriage divide,” with only 10 states offering largely end-to-end online processes while many still rely on paper forms and in-person visits. Castro argues the barriers are legal and administrative—not technological—and calls for reforms such as permitting electronic signatures to modernize marriage services

European calls for “payment sovereignty” misdiagnose the problem: Visa and Mastercard lead through competition, not coercion, and a state-backed alternative would entrench protectionism instead of enabling regulatory reforms that would let European firms scale and compete globally.

States have moved many public services online, but the legal steps to get married remain largely paper-based. Modernizing marriage licensing, recording, and certificates would reduce costs, save time, and make the major life event easier for American families.

In a letter to the Financial Times, Daniel Castro argues that Europe’s push for “digital sovereignty,” exemplified by France replacing Zoom and Teams with a domestic platform, risks fragmenting the transatlantic digital ecosystem and weakening security and efficiency, and that true resilience comes from interoperable systems, shared rules, and cooperation among allied countries.

Governments are entering a critical phase in the adoption of AI. It is already contributing to everyday public sector work, and the question is no longer whether to adopt AI, but how to do so both effectively and responsibly. The Public Sector AI Adoption Index 2026 focuses on the human side of AI adoption, examining how it is experienced by public servants every day.

The TikTok deal shows that targeted structural safeguards can address data security risks without banning foreign apps outright. It also highlights unresolved challenges around reciprocity, uneven enforcement, and how governments should handle other Chinese tech platforms going forward.

The Grok bikini controversy highlights real harms from AI misuse, but it also shows that the right response is enforcing existing laws, holding bad actors accountable, and pursuing tech-neutral, proportionate policies—rather than rushing into broad, AI-specific regulation that risks undermining free expression and innovation.