Artificial intelligence is commanding headlines.
In the first week of his second term, President Trump rescinded the Biden Administration’s October 2023 executive order on AI, issued a new less regulatory AI executive order, and announced the $500 billion Stargate Joint Venture on AI infrastructure, involving Oracle, Open AI, and Softbank.
That same week, the Chinese AI firm DeepSeek announced a new AAI large language model comparable to the leading American OpenAI model but developed at a fraction of the cost. This underscored China’s potential as a strong international rival to the U.S. in the quest for AI leadership.
Then on February 11, speaking at a global AI Summit in Paris, Vice President Vance “push[ed]
back on European efforts to tighten AI oversight while advocating for a more open, innovation-driven approach.” He emphasized that the U.S. intends to remain the “gold standard” in AI technology.
In the coming months the Trump Administration may be rolling out new policy prescriptions to support procompetitive innovative growth in AI.
As the Administration weighs alternatives, the time is ripe to revisit the case against AI regulation – and to suggest some specific initiatives the Administration may wish to consider as part of its AI action agenda.
Background – A Regulatory Approach to AI Is Problematic
The case for regulating AI has not been made, as I explained in a 2024 Forbes article.
AI is already subject to all U.S. civil and criminal laws of general application. The full range of potential AI abuses can be addressed by vigorously enforcing statutes that cover civil rights, national security, antitrust, privacy, consumer protection, labor rights, health care, torts, and miscellaneous crimes.
These safeguards should suffice, at least for now, given the net benefits that AI is bestowing on firms and workers. A 2025 World Economic Forum report concluded that AI “transforms companies’ ability to apply intelligence, creating opportunities for innovative business models and new value pools”, and “recent research . . . suggests that these benefits do not come at the expense of jobs.”
The Clinton Administration’s decision not to regulate another high tech sector, the nascent internet, is instructive. It yielded great economic dividends, allowing “permissionless innovation” to thrive.
Further caution is suggested by scholarship showing that regulation has a long history of stifling innovation and economic growth. A 2017 Journal of Regulatory Economics statistical study found that more-regulated industries had fewer new firms and slower employment growth in all firms. A 2023 article in the American Economic Review (the most prestigious economics journal) found that regulation tends to slow innovation. Significantly, innovation is important for higher productivity and economic growth, as explained in a 2017 European Central Bank paper.
The European Union, however, is not convinced. In dealing with high tech industries, it favors the “precautionary principle” of avoiding risk through early adoption of regulation. Perhaps not surprisingly, then, it has had a terrible innovation track record. To take but one example, initial evidence suggests that EU regulation of digital privacy has entrenched incumbent companies, raised barriers to entry, and harmed smaller firms.
The March 2024 AI Act is the latest EU effort to micromanage innovation. It bans AI systems that are “considered to pose an unacceptable risk for the health, safety, and fundamental rights of individuals.” The Act also regulates general purpose AI models that can be applied to a wide variety of tasks, imposing obligations on them according to the level of risk they impose, or are believed to impose, on the public.
Continued heavy-handed European AI regulation would have global, not just regional, repercussions. As Vice President Vance stated in his February 11 remarks, AI presents “the extraordinary prospect of a new Industrial Revolution,” but “it will never come to pass if overregulation deters innovators from taking the risks necessary to advance the ball.”
Perhaps the EU (and maybe some other governments attending the Paris Summit) is starting to get the message. According to one report on the February 11 discussions, “Europe says it will ease regulations on artificial intelligence.”
Nevertheless, this report appears to be in tension with French President Macron’s French plan to “advance[e] international governance of AI” in order to “enable the consolidation of trust, acceleration and innovation in order to set the rules for AI.” This statement smacks of international AI regulation, a position opposed by the new Trump Administration, as Vice President Vance has made plain.
Possible Next Steps on AI Policy
The new January 23 Trump executive order, “Removing Barriers to American Leadership in Artificial Intelligence,” aims to “enhance America’s global AI dominance in order to promote human flourishing, economic competitiveness, and national security.” To that end, it requires senior executive branch officials to submit an “AI Action Plan” to the President within 6 months.
As part of the Action Plan, the Administration may wish to consider announcing various non-regulatory procompetitive initiatives. These might include, for example:
- Directing federal agencies (including independent agencies) to evaluate all regulatory impediments to the development and deployment of AI systems in order to determine whether they are justifiable on cost-benefit grounds.
- Requiring that the Office of Management and Budget study any statutory constraints on AI development that are not necessary to protect the public safety or national security, and report to Congress whether those constraints should be repealed.
- Directing OMB, in consultation with the Federal Trade Commission and the Department of Justice, to report to Congress on possible steps (including new legislation) that could be taken to remove competitive barriers to American AI development.
- Requiring the federal antitrust agencies (the FTC and the DOJ) to provide guidance clarifying their antitrust analysis of AI-related transactions, with an assurance that AI-specific efficiencies will be given full weight in the evaluation of competitive effects.
- Directing that the President’s Council of Economic Advisors, in consultation with OMB, the FTC, and the DOJ, submit recommendations to the President on countering foreign anticompetitive market distortions that harm the American AI sector.
Initiatives such as these could prove helpful in enhancing the competitive vitality of American AI and thereby accelerating American innovation.