President Trump’s Artificial Intelligence Action Plan, issued July 23, 2025, pursues American global dominance in AI.  He demands AI developers and business managers choose: either join the American government-driven AI wave (a new more autocratic form of “public-private partnership”) or be isolated as a non-ally or adversary.  What’s your strategy?  .

His federal government action plan for artificial intelligence addresses innovation, infrastructure and international diplomacy.  His plan seeks minimal regulation, significant federal investment, workforce development and energy production.  The plan will transform outsourced technology-driven business services.  We will explore your business transformation to integrate, or not, into Trump’s new vision for “public-private partnership.”

A U.S. Constitutional Perspective. At least nine key federal agencies will establish standards, fund AI research, continuously evaluate risks and update regulations.  These include the Departments of Commerce, Energy, Treasury, Labor, Education, Defense, Homeland Security, Justice and State and the Federal Trade Commission, Federal Communications Commission, the Office of Management and Budget, Bureau of Labor Statistics, National Science  Foundation, Bureau of Economic Analysis, National Science and Technology Council, National Institute of Standards and Technology (NIST), U.S. Trade and Development Agency, Export-Import Bank and International Development Finance Corporation.

The AI Plan assumes the Executive Branch (including the President and federal agencies) has due congressional authorization to implement the details.  The AI Plan lists many existing long-standing federal laws, implying such authorization exists and that the President can implement the AI Plan without further legislation.  This raises potential constitutional challenges to new regulations under Supreme Court decisions (such as Loper Bright, 2024) identified in Trump’s April 9, 2025 Executive Order “directing the repeal of unlawful regulations.”

The AI Plan calls for repeal of “onerous” federal regulations that would stifle AI innovation.  It rejects the more cautious approach of President Biden’s October 2023 Executive Order.  Given Trump’s failure in March 2025 to get Congress to enact a 10-year moratorium on State regulations of AI, the AI Plan mandates that federal funding of AI projects should be awarded only to states with minimal regulations, that do not hinder AI development or deployment. 

Innovation.  The AI Plan proposes federal “financial support” for private enterprises to participate in AI research and development and electricity generation.  It favors open-source and open-weight tools that do not generate subscription revenues for co-developers.  It seeks a robust supply-chain for American semiconductors, robotics and drones, engineering, materials sciences, chemistry, biology and neuroscience, with AI standards in specific domains such as healthcare, energy, agriculture and education. Federal data will be rebuilt as confidential datasets to enable secure AI use cases accessing such data.  

Such support would target more efficient financial markets for energy and include AI in federally-funded R&D programs.  The government would also provide access to “regulatory sandboxes” in government research centers where private AI developers could develop AI tools in exchange for open sharing of data and results.  In new “public-private partnerships,” government, academia and private companies would collaborate to develop adoption of AI standards in specific domains, such as healthcare, energy, agriculture and education.   For employment, a federal AI Workforce Hub will prioritize AI retraining programs.  For manufacturing, federal agencies will invest in developing and growing innovative manufacturing technologies and supply-chain challenges, such as in semiconductors, robotics and drones.

To accelerate basic scientific experimentation, federal agencies will invest in cloud-enabled laboratories (public, private and academic) in engineering, materials sciences, chemistry, biology and neuroscience.  All employees whose work could benefit from using AI will have access to training.   For government procurement, new rules will identify AI models compliant with U.S. privacy, data governance and transparency laws.  

For cybersecurity, federal agencies will promote “security by design” and provide guidance on remediating AI-specific vulnerabilities.  In the judicial system, a new federal rule of evidence may be adopted for forensic authentication of evidence to prevent “deepfake” false documentation. 

Infrastructure.  The AI Plan will facilitate construction of new energy generation, data centers and domestic manufacturing of chips to support AI.   It will enable power plant construction on federal lands and “optimize” the national power grid.

For the use of AI in safety-critical or homeland security applications, AI tools must be “secure-by-design.”   Information on data threats will be shared across federal agencies.  Federal agencies will promote “security by design” and share guidance to the private sector on remediating AI-specific vulnerabilities.  

International Diplomacy.  The AI Plan proposes an enduring American-led global alliance with non-rival foreign countries as customers and trusted suppliers.  It promotes controlled exportation (and no re-export) of full-stack AI packages, including AI models, software, hardware, semiconductors, applications and standards.  The AI Plan invites international diplomatic and standards-setting bodies, such as the UN, ICANN and OECD, to advocate for light regulation and against Chinese influence.  But the AI Plan does not ask the European Union to simplify its heavy regulation within Europe under its AI Act. 

Strategies for AI Developers and Users.  

Choose Your Ideal Clients.  The Trump AI Plan seeks to apply AI to government, energy grid, military and defense and infrastructure cybersecurity.  This opens the door to developing AI tools for unique governmental use cases, subject to extensive governmental governance and ownership rules. 

Choose Your Quagmire: Which Regulatory Framework and Marketplace: U.S. vs. Rest of World.  The American AI Plan avoids much the European AI Act’s quagmire of complexity, recordkeeping, high-risk governance, watermarking and penalties.  Yet the American AI Plan may adopt its own quagmire of “public-private” governance rules due to federal funding.  New Mission:  By segregating product lines according to the AI rules in different markets, AI developers might pursue both U.S. and European markets, complying with their different regulations.

Choose Your Ideal Target Customer: Government or Private Industry.  Trump wants to create an AI procurement toolbox to increase federal governmental use of AI to serve the public.  He proposes to build an “AI Evaluation Ecosystem” to assess the performance and reliability of AI systems.  He proposes to publish governmental guidelines for Federal agencies to evaluate AI systems’ operation and compliance with law, to evaluate potential use in governmental affairs.  In short, AI developers must choose whether to serve the government and, if so, decide to comply with emerging evaluation and procurement rules.

Choose Your Control and Ownership Levels: Build Your Own Proprietary AI; Start Small under Model Context Protocol.  The Plan invites anyone, including small business, to make proprietary AI tools, with minimum regulation.  Even small proprietary tools can link to the universe of external data sources and tools (such as the Internet) using the AI-lingo “Model Context Protocol” (MCP) to link one AI tool to the context-specific data of another AI tool.  Thus, many smaller businesses may wish to develop interconnectable proprietary AI tools.  This will enable them to retain control over intellectual property, privacy, cybersecurity and competitive advantage, at the cost of accessing third-party context-specific data.  As a precaution, since AI tools are developing inter-AI communications abilities, your development plan should perhaps require anonymization or pseudonymization of data before joining or linking to external AI tools.  In short, you can build AI tools that serve both government and private marketplaces, so long as you insulate your building blocks and retain your trade secrets.

Who Owns What?  The Trump AI Plan is vague as to ownership and licensing of intellectual property rights under federally-funded research.  Developers could lose substantial commercial value to the extent they share innovations in return for federal funds.  See Federal Acquisition Regulation (FAR) 52.227-14 and in the Defense Federal Acquisition Regulation Supplement (DFARS) 252.227-7013 and 252.227-7014.  

De-Risk Now: No Hidden Bias.  Clean up your AI Tools.  To satisfy the Plan, AI developers should remove ideological or other bias, including diversity, equity and inclusion.

De-Risk Now: No Foreign Influence.  AI developers should anticipate restrictions on the integration of any Chinese content or tools in their toolkit.  The plan seeks to integrate foreign customers into U.S. AI tools, subject to U.S.-approved security agreements and standards, and prevent our allies from becoming dependent on “foreign adversary technology.”  Diplomacy about AI will seek to advocate to foreign governments and standards-setting bodies to adopt the “American governance approaches” and American values and counter “authoritarian influence.” 

De-Risk by Compliance with Non-AI Laws. While American AI regulation might be light, generic compliance laws on employment, personal privacy, data protection, cybersecurity, civil rights and business will still apply to actions taken in reliance on AI tools.  

De-Risk by Getting Consents for Your Chatbots: Get Visitor Consents to Training your AI-driven Chatbots.  Since AI tools may impact a large number of individuals, non-compliance may invite consumer class action litigations.  By using a chatbot to respond to your website visitors’ questions, you could be exposing your company to claims of non-disclosure, non-consent and invasion of privacy in class actions based on U.S. data privacy laws.  Disclose to your visitors that your chatbot is automated, that it may gather your questions and comments as training materials and that may include personally identifiable information that will remain confidential. 

Will AI Grow your Firm’s Value?  AI offers opportunities to grow value. In human resources, AI will lead to job displacement and new job roles that focus on human reasoning and personal interactions with others.  It’s time to reevaluate future needs and train for upskilling.  

Owners and Boards: Choose Your Path: Growth or Exit Plan?   The AI Revolution is here.  Business owners, boards and executives must redesign their business models to accommodate AI tools.  Adaptation will require substantial investment.   Many human-intensive industries, particularly offshore outsourcing and operational support, might choose to not make that investment.  Such industries (such as call centers, back-office transactional processing, manufacturing, and customer service) are candidates for AI transformation.  Private equity investors have already begun to make acquisitions to acquire and transform portfolio companies that will integrate AI tools and redefine human roles.   

If you wish to explore any of these avenues for resiliency, growth, intellectual capital, contracting or developing an exit strategy for business transmission, please contact William Bierce.

Posted: September 17, 2025