President-elect Trump will be the first chief executive to take office in the post-ChatGPT era. The second Trump administration will run from January 2025 to January 2029, a period of great importance for advances in artificial intelligence (AI) technology and AI policy in the United States and abroad. the foreigner. Additionally, Republican control of not only the White House but also the Senate will place President Trump in a strong position to advance AI-related legislative priorities.
Changes in AI regulation
What does all this mean for AI? Here are some likely changes:
Relaxation of agency regulations on certain AI-related issues: This was widely reported that President Trump will repeal AI Executive Order (EO) issued by President Biden in October 2023. This EO reflected a heavy with regulations, multi-agency approach to AI. The EO directed the Department of Commerce to adopt new requirements for reporting large AI models and computing clusters to the government.
Also consistent with the EO, the Office of Management and Budget issued binding guidance to agencies. to use And supply of AI. The Trump administration could choose to repeal some or all of these requirements. Or it can make changes by, for example, adopting a less risk-focused policy or reducing consideration of climate impacts.
More broadly, the new administration will likely direct the Commerce Department, Federal Trade Commission (FTC), and other agencies to take a more hands-off approach to AI regulation. The new administration may also take a different position regarding voluntary “soft law” frameworks. This could impact how the Trump White House interprets safety and voluntary testing commitments that the Biden administration secured from leading US AI companies in July 2023.
Increased AI export control: One area where the second Trump administration could maintain or strengthen Biden-era AI regulations is in export controls. Under President Biden, the Department of Commerce’s Office of Industry and Security has strengthened AI export control rules by October 2022, October 2023And April 2024. One of the main goals of these rules is to prevent China from accessing the most advanced AI semiconductor chips.
The second Trump administration will likely further tighten AI-related export restrictions. One of the major challenges that the Biden administration has faced and that the incoming Trump administration will also face is the limited effectiveness of these rules. There is a strong presence abroad black market in banned AI chips. Additionally, cloud computing facilitates remote access to the computing power of AI chips, including across international borders. If the new Trump administration seeks to close these gaps, it will be important to do so in a way that avoids collateral damage to the American AI ecosystem.
An increased focus on China: In addition to strengthening export controls, the new Trump administration will likely focus on competing with China in other ways. While the Biden administration has taken steps to stay ahead of China on AI, it has also open avenues for bilateral engagement on the management of risks linked to AI. If, as expected, President Trump imposes tariffs on a wide range of products from China, continued bilateral engagement on AI will be less likely.
Partly in response to recent reports Given China’s growing AI activities in the military and intelligence domains, the new administration will likely intensify its work on the military and intelligence use of AI, including in increased collaboration with American AI companies. Among other consequences, this will likely involve adopting a less risk-averse approach than that reflected in the Biden administration’s October 2024 document. National Security Memorandum on AI.
Reduced enforcement of antitrust laws related to AI: The new Republican administration is certain of replace FTC Commissioner Lina Khan with someone who has a less interventionist approach. Justice Department antitrust officials under the new administration will also be less likely to pursue antitrust actions in the technology area. For AI, this means that acquisitions by and by AI companies that might have been blocked on antitrust grounds under a Democratic president will be more likely to proceed unhindered.
Additionally, even though Microsoft’s recent decision to hire two co-founders and numerous employees of Inflexion prompted the FTC to open an antitrust investigationIt is less clear whether an analogous deal reached under the incoming Trump administration would lead to similar FTC scrutiny. Furthermore, in the new administration, the threshold for behaviors deemed to constitute exclusionary practices under the Sherman Law may be higher.
Increased federal support for the adoption of autonomous vehicles: Waymo, an autonomous vehicle company subsidiary of Alphabet, the parent company of Google, is already giving 150,000 taxi rides without a driver per week, compared to approximately 50,000 in May. The next four years will see continued growth in the number of driverless cars, a technology in which AI-based driving algorithms play a central role.
The second Trump administration will, so to speak, have a front-row seat to overseeing and updating rules to promote security. integration autonomous vehicles with traditional human-driven vehicles. In formulating the administration’s autonomous vehicle policy, President-elect Trump will likely seek the advice of Elon Musk, who as CEO of Tesla has made major investments in developing self-driving technologies .
Federal preemption of certain state AI regulations: An emerging feature of the US AI ecosystem is the patchwork of state-level AI laws. These laws, including those of Colorado, UtahAnd California…were adopted in many cases in response to a perceived lack of sufficient regulation at the federal level. However, well-intentioned state AI laws, when considered at the national level, can create an overly complex patchwork that increases compliance costs for U.S. AI companies, putting U.S. AI at a disadvantage in the arena. international.
A new Trump administration could help streamline U.S. AI regulation by introducing federal legislation that would preempt the subset of state laws most likely to disadvantage U.S. AI competitiveness. This does not necessarily mean heavier federal regulation: the federal government can also implement negative laws that would explicitly or implicitly prohibit state legislatures from passing particular forms of AI legislation.
The result: a key period for American AI
There is no shortage of early versions of US AI policy under the new administration, notably in TechCrunch, Street R, Fast business, Voice, TransformerAnd Uncontrollable. Together, these elements, along with this article, help to articulate some of the policy opportunities and challenges in AI over the coming years.
But as the release of ChatGPT in late 2022 clearly demonstrated, it’s always difficult to look ahead with technology, and that’s especially true with AI. What is certain is that the next four years will bring unexpected developments in the field of AI and that effectively managing this extraordinary technology will require a set of agile and balanced federal policy responses.