Development of an Artificial Intelligence (AI) Action Plan

TO:
Networking and Information Technology Research and Development (NITRD)
National Coordination Office (NCO), National Science Foundation (NSF)
2415 Eisenhower Avenue
Alexandria, VA 22314

DATE: March 14, 2025

SUBJECT: “Development of an AI Action Plan”

FROM: Bipartisan Policy Center

The Bipartisan Policy Center appreciates the opportunity to offer insights in response to the National Science Foundation’s (NSF) request for information on the Development of an Artificial Intelligence (AI) Action Plan. BPC brings together diverse perspectives to craft data-driven, pragmatic policy solutions.

For more than 20 years, BPC and Bipartisan Policy Center Action have been consistently achieving results on complex issues, demonstrating that bipartisan cooperation is not only possible but essential for lasting solutions. This letter reflects BPC’s broad policy expertise on AI-related issues such as energy resiliency, innovation, national competitiveness, and workforce development.

These recommendations emphasize BPC’s commitment to advancing AI innovation that strengthens U.S. global leadership, supports critical energy and infrastructure needs, safeguards privacy, and enhances security and economic competitiveness. We look forward to working with the administration to develop and implement a forward-looking AI Action Plan that meets the demands and opportunities of this rapidly evolving landscape.

I. Data Centers, Energy Consumption and Efficiency, Hardware and Chips

BPC’s AI and Energy Project is actively exploring bipartisan policy opportunities at the intersection of AI, energy, and infrastructure. We are committed to forward-looking solutions that meet AI’s growing electricity demands and accelerate AI infrastructure development in the United States.

In February 2025, BPC published Electricity Demand Growth and Data Centers: A Guide for the Perplexed, which examines the accuracy of data center electricity demand projections and their implications for policymakers. The report determines that data centers will likely account for up to 25% of expected electricity load growth nationally through 2030, with the remaining 75% being driven by onshoring of manufacturing and increasing electrification of transportation, heating, and industrial processes. It is also important to address future uncertainties from advancements or changes in computing efficiency, manufacturing production capacity, and demand for AI products and services.

BPC recommends the following actions:

• Advance efficiencies in AI chip technology and data centers. Direct the Department of Energy (DOE) and other federal agencies to invest in advancements that improve the energy efficiency of data centers and next-generation chips.
• Promote low-carbon power procurement for data centers. Foster DOE partnerships with data center operators to explore advanced nuclear, geothermal, carbon capture, and other clean firm power sources.
• Incentivize utility and data center expansion. Establish grants and targeted tax credits to accelerate investment in generation capacity, transmission upgrades, and cybersecurity enhancements.
• Develop standardized metrics for data centers. Task NIST, DOE, and industry with developing clear, standardized frameworks for measuring data center computing efficiency and power usage.
• Streamline key infrastructure buildout support for AI Energy Zones. Develop a strategy to streamline approvals in federally designated areas to enable rapid buildout of data centers powered by clean firm energy.
• Prioritize affordability and reliability in AI energy infrastructure development. Establish guardrails that consider cost impacts and affordability for all customer types.

II. Research and Development, Innovation, and Competition

DOE has long demonstrated the unique expertise and capabilities necessary to study, use, and develop innovative AI applications. BPC recognizes, identifies, and encourages DOE’s continued R&D investments and contributions to a robust national AI strategy.

BPC recommends the following actions:

• Centralize DOE AI strategy and execution. Establish a central DOE office to coordinate and execute a DOE AI oversight strategy.
• Advance AI applications under DOE’s broad mission space. Leverage DOE’s expertise to solve large-scale problems in science, energy, and security.
• Coordinate federal AI research agendas across DOE, NSF, and NIST. Unite efforts to reduce duplication and accelerate breakthroughs.
• Increase researcher access to DOE’s AI facilities. Expand NSF’s NAIRR and similar initiatives to more researchers.
• Leverage DOE national laboratory partnerships. Strengthen collaboration pathways using national lab assets.
• Establish public-private test beds for AI-driven electric grid solutions. Create collaborative environments to de-risk emerging technologies.
• Expand AI R&D for real-time electric grid management. Direct DOE to explore AI tools for reliability and predictive operations.
• Support long-term AI R&D programs that fill private sector gaps. Maintain DOE investment in underfunded AI sectors.
• Encourage open government data initiatives for AI research. Expand and enhance public datasets and ensure they are AI-ready.

III. National Security, Cybersecurity, and Data Privacy

• Deploy AI for critical infrastructure security and threat detection. Instruct DOE and DHS to develop AI anomaly detection tools.
• Enforce robust data safeguards in the AI lifecycle. Integrate privacy and security practices at every AI development stage.
• Develop a federal privacy framework tailored to AI systems. Propose legislation that addresses the unique demands of AI data use.

IV. Technical and Safety Standards, Procurement

• Empower NIST to develop voluntary AI standards and certification programs. Fund NIST to develop and implement industry standards.
• Standardize federal AI procurement processes and guidelines. Establish consistent rules, benchmarks, and accountability structures.
• Institute clear documentation and transparency requirements for government AI. Require agencies to document design, inputs, and performance of AI used in public services.

V. Education and the Workforce

The adoption of AI can drive innovation and economic growth. However, misinformation and mistrust remain significant barriers. Building an AI-literate population is essential for national competitiveness, public confidence, and workforce preparedness.

BPC recommends the following actions:

• Expand upskilling programs for the federal workforce. Offer micro-credentials, short courses, and rotations to build AI competency.
• Establish training and grant programs for state and local readiness. Support municipalities in responsibly adopting AI in public services.
• Promote AI modules within core curricula at colleges and universities. Seed grants to develop and scale AI modules across disciplines.
• Support existing DOE programs that are training the national AI workforce. Continue investing in programs like the AI Summer Institute and Computational Graduate Fellowship.
• Attract and retain foreign-born AI talent. Expedite green card processing for AI-related occupations facing labor shortages.

CONCLUSION

BPC stands ready to collaborate with OSTP and other federal partners in shaping a comprehensive AI Action Plan that promotes technological innovation, strengthens U.S. infrastructure resilience, ensures public trust, and supports a skilled workforce. We appreciate your consideration of these recommendations and look forward to continued engagement on these critical issues. If you have any questions or need further information, please contact our team.

This document is approved for public dissemination. The document contains no business-proprietary or confidential information. Document contents may be reused by the government in developing the AI Action Plan and associated documents without attribution.

Sincerely,

Tanya Das
Director, AI and Energy Policy
Bipartisan Policy Center
[email protected]

Jack Malde
Associate Director
Bipartisan Policy Center
[email protected]

John Fogarty
Associate Director
Bipartisan Policy Center
[email protected]

Sarah Turyahikayo
Senior Policy Analyst
Bipartisan Policy Center
[email protected]