On Wednesday morning, President Donald Trump unveiled the highly anticipated AI Action Plan. The executive order addresses three primary areas to guarantee U.S. leadership in AI and stimulate domestic economic development. As expected, it relaxes regulations on technology firms, described as “onerous” and “bureaucratic red tape,” incorporates provisions for enhancing AI infrastructure within the U.S., and promotes the establishment of U.S. AI technology as the global standard for “like-minded nations” as opposed to adversarial nations such as China. The comprehensive document will be subject to thorough examination in the upcoming days, but several key policies stand out immediately.
Trump advocates for “anti-woke” AI and oversight of state AI regulations. The document directs federal agencies to collaborate with firms developing large language models (LLMs) to ensure their systems remain objective and devoid of ideological bias, tackling AI models perceived as excessively liberal or “woke.” The announcement stipulates that “AI systems must be free from ideological bias and crafted to pursue objective truth rather than social engineering initiatives when users seek factual data or analysis.” To hasten AI development, the executive order also instructs the Office of Management and Budget (OMB) to restrict federal funding for states whose regulations “may impede the effectiveness of that funding or award.”
Data centers could potentially be established on public land. The plan further emphasizes AI infrastructure enhancement and elevating AI manufacturing in the U.S., which includes building data centers, facilitating permit access by deregulating specific environmental standards, and upskilling relevant workforces such as electricians and HVAC technicians. This segment also proposes utilizing federal lands for the construction of data centers.
Trump intends to restrict access to American AI technology. A crucial aspect of the plan centers on advancing American AI technology on a global scale. “The United States must fulfill global demand for AI by exporting its complete AI technology stack—hardware, models, software, applications, and standards—to all countries willing to join America’s AI coalition.” The executive order also directs the Department of Commerce (DOC), the Office of Science and Technology Policy (OSTP), and the National Security Council (NSC) to enhance export controls on critical chips to prevent them from reaching “countries of concern.”
In other areas, the order mandates federal agencies engaged in domestic labor and economic issues to “examine AI’s effects on the labor market” and formulate initiatives for retraining workers impacted by AI job displacement. Previously overlooked in policy discussions, the announcement addresses the sometimes subjective and inadequate assessments of AI models. “Rigorous evaluations can serve as a vital tool for defining and gauging AI reliability and performance in regulated sectors,” the order noted. “Over time, regulators should investigate the application of evaluations in their enforcement of existing laws pertaining to AI systems.”
There’s a great deal to unpack here and even more to consider regarding how these directives will be implemented. Stay tuned for further coverage on what this signifies for the AI sector and its wider implications for our society.