On January 23, 2025, the Trump Administration issued Executive Order No. 14,179, “Removing Barriers to American Leadership in Artificial Intelligence.” This Order followed the January 20 rescission of the Biden Administration’s policies protecting Americans from technology-based human rights harms, including discriminatory mistreatment and data breaches. Because the new Administration’s artificial intelligence (AI) policy is still taking shape, it is useful to understand the protections that were in place when President Biden left office.

In its final two years, the Biden Administration issued several carefully crafted statements to guide the development and use of AI systems. These included the Blueprint for an AI Bill of Rights, issued in October 2022 by the White House Office of Science and Technology Policy (OSTP); the Artificial Intelligence Risk Management Framework (AI RMF), issued in January 2023 by the National Institute of Standards and Technology (NIST); and the Risk Management Profile for Artificial Intelligence and Human Rights, issued in June 2024 by the U.S. State Department’s Bureau of Cyberspace and Digital Policy. Each statement identified several risks to individuals posed at different stages of the AI development and deployment lifecycle, including the risks that data gathered for training purposes would invade protected privacy interests; that discriminatory biases in the training data would be perpetuated by the AI system in operation; that AI systems would be deployed without sufficient testing to ensure safe operation or would be used beyond their intended scope; and that individuals would not be informed when an AI system was used to make decisions affecting them and would not be able to challenge algorithmic decisions.

The United States’ duty to reduce many identified risks arose explicitly from the shared global obligation to respect human rights. At the same time, all the reports, particularly OSTP’s Blueprint, recognized a U.S. national interest in protecting Americans against incursions on their rights from AI systems trained on U.S. data sources. For these reasons, the reports offered design and implementation guidance to ensure that AI systems were, as NIST’s AI RMF put it, “valid and reliable, safe, secure and resilient, accountable and transparent, explainable and interpretable, privacy-enhanced, and fair with harmful bias managed.”

Building on his Administration’s prior statements, President Biden issued Executive Order No. 14,110, “Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence,” on October 30, 2023. Although this very lengthy directive was addressed only to federal executive agencies, it proposed leveraging government purchasing and regulatory decisions to guide future AI development by private entities along the lines suggested above.

Shortly after his inauguration, President Trump rescinded E.O. 14,110, and issued Executive Order No. 14,179, “Removing Barriers to American Leadership in Artificial Intelligence.” This two-page Order asserts that the prior Administration’s policies “act[ed] as barriers to American AI innovation” and prevented the nation from “act[ing] decisively to retain global leadership in artificial intelligence.” Instead, the new Order’s stated objective is “to sustain and enhance America’s global AI dominance in order to promote human flourishing, economic competitiveness, and national security.” It establishes a 180-day deadline for a group of Executive-branch officials to submit recommendations to the President on accomplishing that goal.

The new Order’s substantive impact on AI development is yet unclear, although more specifics may emerge following the 180-day Executive study. The language of the Order, however, offers a few hints of the current Administration’s likely direction:

• Although the Biden Administration’s pronouncements focused consistently on protecting the interests of persons affected by the development and use of AI technologies, the new Order signals, by its references to “innovation” and “America’s global AI dominance,” that the interests of AI producers and developers will now be prioritized.
• Although the new Order asserts that past policies dampened innovation in AI systems, it offers no explanation or supporting examples. The assertion seems difficult to square with recent history, more accurately characterized by the law straining to catch up long after AI developers have surged ahead. Because the new Order opposes prior policies whose innovation-dampening effects are assumed rather than shown, one may presume that the Administration prefers more laissez-faire This expectation may be contradicted, however, by the Order’s assertion (similarly unsupported) that past AI technologies reflected “ideological bias or engineered social agendas.” The likelihood that the Administration will seek to influence the output of AI systems to mesh with its own political views is apparent.

In short, the Administration’s cursory statements to date make predicting its eventual substantive policies difficult.

The nationalistic tone taken in many of the new Administration’s public statements is also on display in the new Order’s reference to “America’s global AI dominance.” By rescinding the Biden Administration’s detailed protections for U.S. citizens, however, the new Order’s likely near-term impact may be just the opposite. Instead, by withdrawing American regulation from the field, the Order boosts the importance of foreign AI regulations, most prominently the 2024 EU Artificial Intelligence Act.