Privacy & Information Security Law Blog

On July 23, 2025, the Trump Administration published an AI Action Plan and three Executive Orders on AI.  The AI Action Plan sets forth a series of policy goals for execution by the federal government with respect to AI.  The Action Plan is accompanied by three Executive Orders: “Accelerating Federal Permitting of Data Center Infrastructure,” “Preventing Woke AI in the Federal Government”, and “Promoting the Export of the American AI Technology Stack.”

AI Action Plan

The AI Action Plan, titled “Winning the Race: America’s AI Action Plan,” articulates policy recommendations by the Trump Administration to ensure American leadership and success in the global AI landscape.  The AI Action Plan consists of three pillars: (1) Accelerate AI Innovation, (2) Build American AI Infrastructure and (3) Lead in International AI Diplomacy and Security. 

Key recommendations from the 28-page AI Action Plan include:

• “Remove Red Tape and Onerous AI Regulation”: The AI Action Plan recommends (1) launching a request for information (RFI) regarding current federal regulations that “hinder AI innovation and adoption” to gather stakeholder feedback; (2) revising and repealing rules, orders and guidance that “unnecessarily hinder AI development or deployment”; (3) evaluating whether state AI regulations “interfere with the FCC’s ability to carry out its obligations and authorities”; and (4) reviewing FTC investigations, consent orders and injunctions to modify or set-aside any that “unduly burden” AI innovation.
• “Ensure that Frontier AI Protects Free Speech and American Values”: The AI Action Plan calls for ensuring that “free speech flourishes in the era of AI and that AI procured by the government objectively reflects truth rather than social engineering agendas.” As part of this recommendation, the Trump Administration proposes revising the NIST AI Risk Management Framework to eliminate references to misinformation, diversity, equity and inclusion (“DEI”) and climate change, and updating federal procurement guidelines to ensure that the government “only contracts with frontier large language model (LLM) developers who ensure that their systems are objective and free from top-down ideological bias.”
• “Encourage Open-Source and Open-Weight AI”: The AI Action Plan recommends partnering with leading technology companies to increase the research community’s access to private sector computing, models, data, and software resources.
• “Enable AI Adoption”: The AI Action Plan recommends a coordinated federal push to foster a culture of experimentation and rapid AI deployment across industries. This includes establishing regulatory sandboxes and AI Centers of Excellence to allow researchers, startups, and companies to test AI technologies in real-world settings. The plan also calls for domain-specific initiatives in areas like healthcare, agriculture, and energy to develop national standards and measure AI’s impact on productivity.
• Build World-Class Scientific Datasets”: The AI Action Plan calls for the creation of “the world’s largest and highest quality AI-ready scientific datasets” while upholding privacy and civil liberties. The plan includes setting data quality standards for the use of biological, materials science, chemical, physical, and other scientific data modalities in AI model training, among other items.
• “Combat Synthetic Media in the Legal System”: The AI Action Plan highlights the misuse of deepfakes in judicial processes as a risk of AI, and recommends expanding NIST’s deepfake detection efforts into formal forensic guidelines and benchmarks and issuing guidance to help adjudication agencies adopt a deepfake standard.
• “Bolstering Critical Infrastructure Cybersecurity”: The AI Action Plan calls for (1) the establishment of an AI Information Sharing and Analysis Center (AI-ISAC) to promote the sharing of AI-security threat information and intelligence across U.S. critical infrastructure sectors; (2) DHS to issue and maintain guidance to private sector entities on remediating and responding to AI-specific vulnerabilities and threats; and (3) ensuring collaborative and consolidated knowledge sharing of known AI vulnerabilities from within federal agencies to the private sector.
• “Promoting Secure-by-Design AI Technologies and Applications”: The AI Action Plan recommends refining the Department of Defense’s (“DOD’s”) responsible AI and generative AI frameworks, roadmaps and toolkits, and publishing an IC standard on AI Assurance.

Other key features of the AI Action Plan include recommendations to empower American workers in the age of AI, support next-generation manufacturing, invest in AI-enabled science, and train a skilled workforce for AI infrastructure, among other items.

To read about the recommended actions in more detail, see the AI Action Plan.

Executive Orders

The White House also released three Executive Orders on July 23, 2025 focused on implementing key components of President Trump’s AI Action Plan. 

“Accelerating Federal Permitting of Data Center Infrastructure” (“Data Center EO”)

The Data Center EO states as its policy and purpose the pursuit of “bold, large-scale industrial plans to vault the United States further into the lead on critical manufacturing processes and technologies that are essential to national security, economic prosperity, and scientific leadership,” including the development of AI data centers and related infrastructure.  A stated priority of the Trump Administration is to “facilitate the rapid and efficient buildout of this infrastructure by easing Federal regulatory burdens.”  As part of this initiative, the Data Center EO indicates that the Trump Administration will use “federally owned land and resources for the expeditious” development of data centers.

To encourage investment in data center projects, the Data Center EO directs the Secretary of Commerce, in consultation with other relevant agencies, to launch an initiative to provide financial support for “Qualifying Projects,” which are defined to include projects for the construction of data centers and their necessary components, such as energy infrastructure, energy sources (e.g., natural gas, coal, nuclear, geothermal), semiconductor manufacturing, networking equipment, and data storage components.  The Data Center EO includes various criteria for a project to be eligible as a Qualifying Project, such as a project sponsor committing at least $500 million in capital expenditures, an incremental electric load addition of at least 100 MW, or a project that protects national security.  The financial support provided to project sponsors may include loans and loan guarantees, grants, tax incentives, and offtake agreements.

The Data Center EO also sets forth measures to expedite and facilitate environmental reviews under the National Environmental Policy Act (NEPA), and to streamline permitting reviews by developing or modifying regulations under the Clean Air Act, Clean Water Act, Comprehensive Environmental Response, Compensation and Liability Act (CERCLA), and Toxic Substances Control Act.

To further facilitate data center projects, the Data Center EO makes federal lands available in certain circumstances.  The Data Center EO directs the Interior and Energy Departments to consult with industry and the Department of Commerce to offer appropriate authorizations for sites consistent with applicable federal laws.  The Secretary of Defense is also directed to identify suitable sites on military installations and to competitively lease available lands to support DOD AI objectives.

“Preventing Woke AI in the Federal Government” (“Woke AI EO”)

The Woke AI EO begins by claiming that certain AI models distort the quality and accuracy of their output based on the inclusion of ideological biases or social agendas in the model.  The Woke AI EO further claims that DEI ideologies favor preferred AI model outcomes over accurate ones.  It acknowledges that the federal government should be reluctant to regulate AI model functionalities in the private sector, but also states that in the procurement context, the federal government “has the obligation not to procure models that sacrifice truthfulness and accuracy to ideological agendas.”

In order to achieve these objectives, the Woke AI EO sets forth two “Unbiased AI Principles” and directs federal government agency heads to procure only LLMs developed in accordance with such “Unbiased AI Principles.”  The first principle is titled “truth-seeking.” The Woke AI EO states that LLMs must be “truthful in responding to user prompts seeking factual information or analysis” and must “prioritize historical accuracy, scientific inquiry, and objectivity...”  The second principle is titled “ideological neutrality.”  The Woke AI EO states that LLMs must be “neutral, nonpartisan tools that do not manipulate responses in favor of ideological dogmas such as DEI,” and that developers must not “intentionally encode partisan or ideological judgments into an LLM's outputs unless those judgments are prompted by or otherwise readily accessible to the end user.”

The Woke AI EO provides measures for implementation of the “Unbiased AI Principles.”  These measures include the issuance of guidance to federal agencies to implement the Principles, required contract terms for federally procured LLMs, and the adoption of procedures to ensure that federally procured LLMs comply with the Principles.  As an added incentive for AI providers to comply with the Principles, the Woke AI EO provides that decommissioning costs will be charged to the AI provider in the event of termination by a federal agency for the AI provider’s noncompliance.

“Promoting the Export of the American AI Technology Stack” (“American AI EO”)

The purpose of the American AI EO is to ensure that the United States leads in developing AI technologies and that American AI technologies, standards and governance models are adopted worldwide to strengthen relationships with U.S. allies.  The American AI EO seeks to establish a coordinated national effort to support the American AI industry by “promoting the export of full-stack American AI technology packages.”

The American AI EO directs the Secretary of Commerce to establish and implement an American Al Exports Program with 90 days of the EO to “support the development and deployment of United States full-stack export packages.”  The Secretary of Commerce is directed to issue a public call for proposals from industry-led consortia.  Each proposal must describe a full-stack AI technology package, encompassing AI-optimized computer hardware, data center storage, cloud services and networking, data pipelines, AI models, and information security measures.  Each proposal also must identify specific target countries for export, describe a business and operational model including which entities will build, own and operate data centers, identify requested federal incentives, and comply with U.S. export control and other applicable laws. The Secretary of Commerce in consultation with other federal agency heads will evaluate and select proposals and support the selected AI export packages by facilitating priority access to federal financing tools.

The Secretary of State is directed to coordinate U.S. participation in multilateral initiatives and country-specific partnerships for AI deployment and support partner countries in fostering pro-innovation environments appropriate for the deployment of American AI systems.  The American AI EO directs the Economic Diplomacy Action Group (EDAG) to coordinate mobilization of federal financing tools in support of priority Al export packages.  EDAG members will offer available federal financing tools to support the priority AI export packages, including loans and loan guarantees, equity investments, co-financing, political risk insurance, credit guarantees, and technical assistance and feasibility studies.