Transatlantic Lessons for the AI Value Chain
Listen to this article
Audio is generated automatically and may contain minor inaccuracies.
Executive Summary
AI is reshaping economies at unprecedented speed and scale. Governments in the United States and the EU have justified massive investments in the field in the name of competitiveness, technological leadership, and strategic rivalry with China. Public opinion, however, remains unconvinced. Underlying the skepticism is a sense that citizens have little control over who benefits from AI and who bears its costs. Governments have endorsed “public interest AI” in principle, but these commitments have not consistently translated into deployment or policy choices, contributing to growing mistrust of the technology. As AI moves from digital interfaces into physical infrastructure, public services and industry, public acceptance becomes a condition for its long-term economic and political viability.
This report examines how the AI value chain can be better aligned with the public interest. It draws on GMF’s May 2026 Transatlantic Tech Exchange (TTX), during which 10 members of the European Parliament met with US federal and state policymakers, industry, investors, academics and civil society in Washington, DC and Boston. The report focuses on three nodes where public-interest tensions are sharpest and transatlantic cooperation most promising: data centers, data, and innovation policy.
On data centers, policymakers should:
- make transparency the baseline for local decisions and require projects to disclose estimated and actual impact on a common checklist of items. The process should be reinforced by shared learning through a transatlantic forum of cities and regions.
- require data centers to deliver lasting value to host communities through enforceable community benefit agreements, backed by baseline requirements throughout a facility’s lifecycle.
- make AI infrastructure growth conditional on energy availability and efficiency. New data center demand should be tied to additional clean power and efficiency standards.
On data, policymakers should:
- encourage transatlantic interoperability, portability and standardization, including shared definitions of “anonymized” and “personal” data
- invest in transatlantic public datasets of nonpersonal data, building on initiatives such as the National AI Research Resource and the EU’s Open Data Directive
- promote research on technical pathways for trusted data sharing, including federated learning and privacy-enhancing technologies
On innovation policy, policymakers should:
- use public funding, procurement, and approvals to fill gaps left by private capital, with support conditional on measurable public-interest outcomes
- strengthen existing innovation ecosystems while broadening participation to under-resourced regions
- align public and private capital with public-interest objectives, with better tracking of AI investment flows.