The United States is at a historic inflection point. The recent release of advanced artificial intelligence models by Chinese competitors was not just another technological milestone; it was a "wakeup call," as President Donald Trump rightly noted. AI is the most consequential innovation of our lifetime, and the President’s AI Action Plan coming on July 23 should ensure America leads in its development to solve society's most complex problems and to shield our economic and national security.
Despite the urgency, the path to securing America's AI dominance does not require a radical new playbook. The blueprint is embedded in our national DNA: a commitment to permissionless innovation and regulatory humility. This is the formula that built the American-led internet and the world's most dynamic economy. The new AI Action Plan should reflect these principles to maximize success.
"Permissionless innovation" is the simple but powerful idea that experimentation should be the default. Innovators should be free to build, create and test new ideas without first seeking government approval. "Regulatory humility" is the recognition that government cannot predict the future. Rather than trying to preemptively anticipate every potential problem with a new technology and imagine a rule for it — a process that inevitably stifles progress — we should address specific, existing harms with targeted, incremental and sectoral rules if and when they arise.
Unfortunately, recent policies have steered us away from this proven model. In its final days, the Biden administration attached a litany of progressive policy strings to AI infrastructure development, wrapping it in the same red tape that has plagued other federal programs. Its executive order, while paying lip service to American leadership, buried progress under mandates for prevailing wages, "clean energy" quotas and burdensome environmental reporting.
Simultaneously, a radical and expansive approach to antitrust has put our most important AI innovators in its crosshairs. The very companies with the capital, data and talent to lead the global AI race — firms like Google, Amazon, Meta, Microsoft and more — are being bogged down by litigation and investigations. This creates a cloud of uncertainty that chills investment and integration. The Department of Justice's initial attempt to force Google to divest from AI startups is a stark example of this misguided approach.
The assault on innovation doesn't stop there. Changes to the Hart-Scott-Rodino (HSR) merger review process have quadrupled paperwork requirements and imported a European-style model of endless delays and stagnation. This discourages the mergers and acquisitions that are a lifeblood for the tech ecosystem, signaling to venture capitalists that their investments in promising startups have fewer paths to a profitable exit.
To win the AI race, we must reverse course. The path forward is clear.
First, the Trump administration must take immediate action. It should rescind and replace the burdensome Biden executive order restricting AI infrastructure, freeing development from unrelated policy goals. It must seek fair settlements in the antitrust cases against leading U.S. tech firms and roll back the HSR changes to restore a healthy environment for investment and acquisition.
Second, the president should work with Congress to build on key legislative priorities. We need a national data privacy and security law that also preempts states from targeting AI with onerous and superfluous regulations. The current patchwork of state laws on both privacy and AI creates a compliance nightmare that burdens businesses, especially startups, and provides inconsistent protection for Americans. A single, national standard would provide the regulatory certainty needed to fuel data-driven innovation through tools like AI.
Finally, we must reject the false narrative that AI is an unregulated "wild west." A robust legal framework already governs many applications of AI tools. Healthcare AI must comply with HIPAA, financial AI with the Fair Credit Reporting Act, and all AI is subject to anti-discrimination laws, consumer protection statutes and tort law. The right governance is to empower our existing sectoral regulators to oversee AI within their domains of expertise, not to create a new, overarching bureaucracy that would handcuff innovation across the board.
America's legacy of technological leadership was built on a foundation of economic freedom. We have the innovators, the investors and the vision to lead the world in artificial intelligence. But they cannot succeed with one hand tied behind their back. It is time for the government to get out of the way and let them build.
Patrick Hedger is the Director of Policy at NetChoice, a trade association dedicated to protecting free enterprise and free expression online.