The First Hundred Days of Governing AI Chips

In 2025, the United States has entered a critical new phase in how it governs artificial intelligence (AI). Driven by growing concerns over national security, economic competitiveness, and the rapid spread of powerful AI systems, lawmakers in Washington have responded with a wave of new policies, regulations, and executive actions. The federal government keeps moving beyond voluntary guidelines and launching a legal framework to manage the risks and unlock the potential of AI across sectors. Do these changes mark a turning point where the era of light-touch AI oversight is over and a more structured, accountable, and strategic approach is beginning to take shape?

[The Big Picture of Federal AI Laws and Policies]

Recent federal statutes and regulatory instruments revealed a policy narrative that consistently encourages AI to flourish: Lawmakers have woven pro-innovation language into funding formulas, competition rules, and sector-specific oversight regimes, and they repeatedly affirmed advancement in AI as a national economic imperative.[1] At the same time, the administration adopted a carefully articulated, risk-based approach that leaves ample space for rapid iteration.[2] This liberal architecture fosters a sense of predictability for tech companies to chart long-term research roadmaps and preserves the policy agility needed to respond when technology edges into ethically sensitive domains.[3] By integrating education, workforce, and infrastructure provisions with supportive programs, the federal legal frameworks that cultivate domestic AI leadership and those that safeguard the public interest are complementary objectives rather than mutually exclusive pursuits.[4]

However, at the federal level, the United States currently lacks a single comprehensive “AI Act.”[5] Federal efforts instead addressed AI-related issues through targeted legislation and policy initiatives. For example, the National Artificial Intelligence Initiative Act (NAIIA), enacted January 1, 2021, as part of the National Defense Authorization Act of 2020, created the National AI Initiative to coordinate AI research and policy across the federal government and established a National AI Advisory Committee (NAIAC) to advise the President and White House National AI Initiative Office.[6] It did not impose regulations on private-sector AI use, but rather bolstered funding and coordination for AI research, education, and standards development.[7]

AI in Government Act of 2020, on the other hand, promotes AI adoption and expertise within the federal government and requires the Office of Personnel Management (OPM) to identify critical AI skills for the federal workforce and develop new hiring guidelines for AI-related roles.[8] Advancing American AI Act, enacted in 2022 as part of the National Defense Authorization Act for Fiscal Year 2023, delineates principles and guardrails for government use of AI and mandates that federal agencies use AI in ways that protect privacy, civil rights, and civil liberties.[9] The provisions of the Act are included as statutory notes related to section 11301 of Title 40 of the United States Code, which deals with the responsibility of the Director of the Office of Management and Budget (OMB).[10] AI Training Act, enacted in the same year, requires the government to create a training program in AI for federal acquisition and program management personnel to help officials understand the benefits and risks of AI.[11]

Specifically related to semiconductors, the CHIPS and Science Act of 2022 represents a significant legislative intervention in the evolving landscape of U.S. innovation policy.[12] The Act positions the National Institute of Standards and Technology (NIST) as a critical architect of federal efforts to accelerate AI research while preserving public trust.[13] NIST must continue to advance the objectives of the NAIIA and promote competitive research in machine learning and data science across academia, industry, and government laboratories.[14] It is further instructed to deepen its exploration of AI-enabled cybersecurity and conduct rigorous testing regimes aimed at detecting systemic vulnerabilities that traditional threat models may overlook.[15]

Complementing these investigative functions, NIST is to establish open, configurable testbeds where diverse stakeholders can evaluate algorithmic performance, resilience, and ethical robustness under controlled conditions.[16] It is to expand laboratory infrastructure and invest in measurement science tailored to state-of-the-art semiconductor architectures designed for AI workloads.[17] Equally important, it must develop technical standards and related normative instruments that embed safety, reliability, and accountability into AI systems from conception through deployment.[18] The Act calls for a federal framework for identifying, assessing, and mitigating risks associated with algorithmic decision-making, accompanied by the creation and public dissemination of cybersecurity tools, advanced encryption methodologies, and sector-specific best practices in data science.[19]

Overall, the CHIPS and Science Act illustrates the underlying evolution of contemporary American technology governance: rather than imposing prescriptive rules at the outset, Congress empowered an expert agency to develop measurement protocols, convene multi-stakeholder communities, and iteratively refine norms as empirical evidence accumulates.[20] This approach presumes that legitimacy in AI regulation flows from transparent, reproducible testing regimes and from standards that reflect consensus among scientists, industry leaders, and civil-society actors.[21] By embedding such processes into statutes, the lawmaking is to secure both domestic competitiveness and public confidence and to suggest that the path to sustainable AI leadership passes through institutions capable of translating complex technical judgments into widely accepted social outcomes.

[The New Deal for AI Chips]

On March 31, 2025, President Trump signed Executive Order Establishing the United States Investment Accelerator and created a new office within the Commerce Department.[22] This office will assume control of the CHIPS program and aims to speed up corporate investment in domestic semiconductor production.[23] The executive order states that the President views slow, overlapping regulations as a drag on semiconductor investment and job creation in the United States.[24] He asserts that modernizing federal processes is essential for keeping the nation’s economy competitive and for encouraging both domestic and foreign companies to build and expand facilities on U.S. soil.[25]

To achieve this goal, the order directs the Secretary of Commerce working with the Treasury and the White House economic team to create a United States Investment Accelerator within 30 days.[26] Housed in the Commerce Department and led by an executive director, the Accelerator is tasked with guiding projects worth more than one billion through federal requirements, coordinating with states to cut red tape, expanding access to national laboratories and resources, and managing the CHIPS Program Office to secure stronger subsidy agreements for semiconductor manufacturing.[27] The order instructs the Accelerator to look for existing legal mechanisms that can further aid investors without compromising national security.[28] It also clarifies that the directive does not override any statutory powers of federal agencies, must be implemented within the limits of available appropriations, and does not create enforceable legal rights for private parties against the U.S. government.[29]

On May 15, 2025, President Trump used his visit to Abu Dhabi to signal a deeper strategic partnership with the United Arab Emirates (UAE) and unveiled more than two hundred billion in new commercial agreements and pledging closer bilateral cooperation.[30] The two governments also launched a U.S.-UAE AI Acceleration Partnership, highlighted by the ceremonial start of construction on a five-gigawatt AI campus, which is the largest of its kind outside the United States.[31] Under the partnership, Washington will loosen previous restrictions and allow the UAE to import roughly 500,000 of Nvidia’s most advanced AI chips annually, provided the Gulf state invests in U.S. data-center capacity of comparable scale and adopts stronger safeguards to prevent technology diversion.[32]

White House officials framed these steps as mutually reinforcing: the UAE gains cutting-edge semiconductor access, while U.S. firms secure major orders and expanded market share.[33] New projects include a four-billion aluminum smelter in Oklahoma led by Emirates Global Aluminum and a sixty-billion hydrocarbons venture pairing ExxonMobil, Occidental, and EOG with Abu Dhabi National Oil Company; collectively, the deals accelerate a previously announced UAE pledge to channel $1.4 trillion into the U.S. economy over the next decade.[34] The administration, on the contrary, portrayed these signings as evidence that the Gulf region is poised to become a third center of gravity in global AI competition that complements existing powerhouses in the United States and China.[35]

Diplomatically, the trip produced a flurry of announcements: President Trump said Washington was nearing a new nuclear arrangement with Iran, pledged to lift long-standing sanctions on Syria after meeting interim president Ahmed al-Sharaa, and urged Damascus to normalize relations with Israel.[36] Both United States and Emirati leaders expressed confidence that the burgeoning economic and security partnership would continue to deepen in the years ahead.[37] From domestic to international, the new Trump administration started to pursue an integrated AI-chip strategy that couples assertive domestic industrial policy with a realist, balance-of-power approach overseas. Public support measures, streamlined approvals, and targeted research initiatives are deployed to expand domestic production, cultivate a specialized workforce, and embed resilience into critical supply lines. Internationally, it prioritizes a deliberately selective network of partnerships. Cooperation is deepened with a small circle of states judged sufficiently reliable, and access to leading-edge technology is withheld from potential rivals.

On July 23, 2025, the White House further unveiled “Winning the AI Race: America’s AI Action Plan,” which followed President Trump’s January executive order.[38] The Action Plan outlined more than ninety federal actions across three policy pillars: (1) accelerating innovation, (2) building American AI infrastructure, and (3) leading in international diplomacy and security.[39] More importantly, the plan prioritized exporting secure “full-stack” AI packages to allies; accelerating permitting for data centers and chip fabs while growing skilled trades; rolling back federal rules that impede AI development; and revising procurement to favor frontier LLMs that are deemed objective and free of ideological bias.[40] On the same day, President Trump issued Executive Order 14320, titled “Promoting the Export of the American AI Technology Stack.”[41]

To sum up, the emerging framework for AI chips illustrates how the long-running dialogue between liberalism and realism continues to shape U.S. statecraft. By treating advanced AI ships as both a domestic infrastructure of national strength and an external instrument of geopolitical leverage, U.S. policymakers would seek to harvest the efficiencies of selective cooperation without surrendering the strategic advantages of controlled access. However, whether that balancing act will ultimately pay off is still anyone’s guess.

[1] Tatevik Davtyan, The U.S. Approach to AI Regulation: Federal Laws, Policies, and Strategies Explained, 16 Case W. Res. J.L. Tech. & Internet 223, 233–39 (2025).

[2] See NIST, Artificial Intelligence Risk Management Framework (AI RMF 1.0) (2023), https://nvlpubs.nist.gov/nistpubs/ai/NIST.AI.100-1.pdf.

[3] See Gary E. Marchant & Carlos I. Gutierrez, Soft Law 2.0: An Agile and Effective Governance Approach for Artificial Intelligence, 24 Minn. J.L. Sci. & Tech. 375 (2023); Margot E. Kaminski, Regulating the Risks of AI, 103 BU L. Rev. 1347 (2023); Matthew R. Gaske, Regulation Priorities for Artificial Intelligence Foundation Models, 26 Vand. J. Ent. & Tech. L. 1 (2023).

[4] See Davtyan, supra note 1; Laurie Harris, Cong. Rsch. Serv., R48555, Regulating Artificial Intelligence: U.S. and International Approaches and Considerations for Congress (2025).

[5] Cf. Regulation (EU) 2024/1689 of the European Parliament and of the Council of 13 June 2024 Laying Down Harmonised Rules on Artificial Intelligence and Amending Regulations (EC) No 300/2008, (EU) No 167/2013, (EU) No 168/2013, (EU) 2018/858, (EU) 2018/1139 and (EU) 2019/2144 and Directives 2014/90/EU, (EU) 2016/797 and (EU) 2020/1828 (Artificial Intelligence Act) [hereinafter EU AI Act], 2024 O.J. (L 1689) 1, 44–45, art. 1(2), https://eur-lex.europa.eu/eli/reg/2024/1689/oj.

[6] USPTO, National Artificial Intelligence Initiative 2–3 (2022), https://www.uspto.gov/sites/default/files/documents/National-Artificial-Intelligence-Initiative-Overview.pdf; AI Congressional Mandates & Executive Orders, NIST, https://www.nist.gov/artificial-intelligence/ai-congressional-mandates-executive-orders (last visited June 1, 2025).

[7] Id.

[8] Veronica E. Hinton, The Artificial Intelligence Classification Policy and Talent Acquisition Guidance - The AI in Government Act of 2020, Memorandum for Chief Human Capital Officers, U.S. Off. of Personnel Mngt. (Apr. 29, 2024), https://chcoc.gov/sites/default/files/The%20Artificial%20Intelligence%20Classification%20Policy%20and%20Talent%20Acquisition%20Guidance%20-%20The%20AI%20in%20Government%20Act%20of%202020.pdf.

[9] United States of America: Advancing American AI Act (SB 1353), Digit. Pol’y Alert, https://digitalpolicyalert.org/change/4281 (last visited June 1, 2025); 40 U.S.C. § 11301 note, Pub. L. 117–263, div. G, title LXXII, subtitle B, Dec. 23, 2022, 136 Stat. 3668, § 7222(3), https://uscode.house.gov/view.xhtml?req=%28title%3A40+section%3A11301+edition%3Aprelim.

[10] Id.

[11] Natalie Alms, AI Training Bill Becomes Law, Gary Peters (Oct. 18, 2022), https://www.peters.senate.gov/newsroom/in-the-news/ai-training-bill-becomes-law.

[12] NIST, supra note 6.

[13] See id.

[14] Id.

[15] Id.

[16] See also id.

[17] Id.

[18] Id.

[19] Id.

[20] See id.

[21] See id.

[22] Trump Sets Up New Office to Manage CHIPS Act and Speed Up Investments, Reuters (Mar. 31, 2025, at 6:38 PM EDT), https://www.reuters.com/world/us/trump-signs-order-aimed-attracting-corporate-investments-more-than-1-bln-2025-03-31.

[23] Id.

[24] Establishing the United States Investment Accelerator, The White House (Mar. 31, 2025), https://www.whitehouse.gov/presidential-actions/2025/03/establishing-the-united-states-investment-accelerator/.

[25] Id.

[26] Id.

[27] Id.

[28] Id.

[29] Id.

[30] Gram Slattery et al., Trump Announces $200 Billion in Deals During UAE Visit, AI Agreement Signed, Reuters (May 15, 2025, at 11:56 PM EDT), https://www.reuters.com/world/middle-east/trump-heads-uae-it-hopes-advance-ai-ambitions-2025-05-15.

[31] Id.

[32] Trump agrees deal for UAE to build largest AI campus outside US, Guardian (May 16, 2025, at 3.35 AM CEST), https://www.theguardian.com/us-news/2025/may/15/trump-artificial-intelligence-uae.

[33] Indranil Ghosh, How Trump’s Gulf Trip Turned Oil Kingdoms into Tech Superpowers, Rest of World (May 15, 2025), https://restofworld.org/2025/trump-middle-east-trip-ai-chip-tech-deals/.

[34] Slattery et al., supra note 23.

[35] Id.

[36] Id.

[37] Id.

[38] White House Unveils America’s AI Action Plan, The White House (July 23, 2025), https://www.whitehouse.gov/articles/2025/07/white-house-unveils-americas-ai-action-plan/.

[39] Id.

[40] Id.

[41] Promoting the Export of the American AI Technology Stack, The White House (July 23, 2025), https://www.whitehouse.gov/presidential-actions/2025/07/promoting-the-export-of-the-american-ai-technology-stack/.