AI Trends For 2026 - The Federal Government’s Use and Regulation of AI

Throughout 2025, the Trump administration has tried to shape the direction of national AI policy through a series of executive orders and agency directives. In July 2025, President Trump’s issued the AI Action Plan along with three executive orders: Preventing Woke AI in the Federal Government, Accelerating Federal Permitting of Data Center Infrastructure, and Promoting the Export of the American AI Technology Stack.* These actions aim to reduce (and potentially even eliminate) perceived obstacles to AI development, which the administration believes will cause the U.S. to fall behind its competitors. For companies that develop or deploy AI, these initiatives signal a federal push toward faster infrastructure approvals and a more government-aligned approach to AI exports.

In early December 2025, President Trump signed Ensuring a National Policy Framework for Artificial Intelligence, an executive order seeking to preempt state-level AI laws and regulations with a yet-to-be-developed uniform AI policy framework. The order also calls for this uniform AI policy framework to be drafted in the form of a legislative recommendation. Lastly, the order establishes a new AI Litigation Task Force within the Department of Justice and proposes to deny broadband grant funding to states deemed to have non-compliant laws. Companies operating across multiple states should begin assessing which compliance obligations are tied to state rules and prepare for potential realignment once a federal standard emerges. The new Task Force also suggests a heightened enforcement environment, so those companies should review marketing materials, technical claims, and procurement certifications now to ensure defensibility.

As part of the effort to prevent federal agencies from procuring “woke AI,” the Office of Management and Budget issued a memorandum to ensure that AI technology purchased by the federal government produces “truthful” outputs that do not “manipulate responses in favor of ideological dogmas.” Large language models (LLMs) are to “prioritize historical accuracy, scientific inquiry, and objectivity, and shall acknowledge uncertainty where reliable information is incomplete or contradictory.” The December 11, 2025 memorandum directs federal agencies to update their internal policies and procedures by March 11, 2026 to ensure agencies only purchase “unbiased” AI and LLM software and modify existing contracts where appropriate. These directives impose significant new documentation and transparency requirements on developers, resellers, deployers, operators, and integrators of AI systems. Vendors will need to provide detailed LLM training process summaries and identified risks and mitigations to enable federal agencies to evaluate compliance. Prospective contractors should assemble a “federal-ready AI documentation package” now to reduce procurement friction and strengthen competitiveness. Contractors should also review their existing federal agreements to identify potentially material amendments and plan for renegotiation timelines.

* We discussed these executive orders and the Action Plan in detail in our blog post.