Winning the AI Future: Why America Should Double Down on the Freedom to Innovate

[Based on remarks delivered before the Congressional Institute on August 5, 2025]

We stand at the threshold of the next great technological revolution. Artificial intelligence (AI)—supercharged by powerful new computational capabilities and advances in data science—is poised to become the most important general-purpose technology of our age. This foundational technology will touch every part of our economy and transform the way we live.

If that sounds ambitious, so did the internet and current information revolution when it was first being discussed thirty years ago. It’s worth asking: How did that previous information technology revolution happen, and how did the United States win that race to dominate that new frontier? That history offers us important lessons as we ponder public policy for algorithmic, computational, and robotic technologies.

A Walk Back in Time

When I began working on technology policy over three decades ago, the digital future was still up for grabs. The United States and Europe were on roughly equal footing in terms of talent and innovation leaders. At that time, it was very much an open question who would lead the internet revolution.

Today, that race is over, and America won it decisively. Consider these facts:

• Eighteen of the 25 largest digital companies in the world are U.S.-based.
• The U.S. accounts for 75 percent of all unicorns by value in the world while Europe accounts for just 3 percent.
• Of the eight trillion-dollar companies on the planet, seven are American and most are digital tech firms. None are European.
• Finally, in 2022 alone, the U.S. digital economy generated over $4 trillion in output—10 percent of GDP—and created nearly nine million jobs.

These astonishing numbers aren’t the result of chance. They are the byproduct of the smart policy choices our nation made in the 1990s. America embraced a pro-innovation, pro-investment vision for digital technology. Freedom to innovate or “permissionless innovation” was our policy lodestar, and we rejected Europe’s technocratic, top-down regulatory approach.

That policy choice made all the difference and allowed America to dominate the dawn of the Digital Revolution. By taking the other path of preemptive, top-down regulatory micromanagement, the European Union became “the biggest loser” in digital technology with almost no major digital technology players on the global stage to show for their efforts and China now overtaking them as America’s leading competitor on the global stage.

Now America faces a similar challenge—and the same opportunity—with AI. The question is: Can we get policy right again and propel the next great technological revolution forward?

The Stakes: Security, Growth, and Human Flourishing

What are the stakes if we get this wrong? Three things are on the line when we consider what policies will govern our AI future.

1. First, geopolitical competitiveness and national security. America already finds itself in what some scholars call an “AI Cold War” with China. The Chinese Communist Party has declared its intention to dominate AI by 2030 and is investing heavily across the full spectrum of frontier technologies—robotics, quantum computing, biotech, advanced chips, and more. Winning this race is not just about global economic advantage; it is also about the values that will shape the future. China is looking to export its AI stack—and the values of control and censorship embedded in it—across the globe. Lose this race, and we risk ceding not only market share, but the future of free speech, human rights, and open societies.
2. Second, economic growth. AI can help fuel broad-based economic growth and sectoral productivity. Already in 2025, the several largest U.S. AI firms have collectively spent a record $102 billion on data centers and AI infrastructure and they are on their way to spending $400 billion by the end of the year. That is more than 12 times what China spent last year. Without AI capital expenditures, GDP growth this year would be two percentage points lower. This is a massive, private-sector-led stimulus that is propelling our current economy forward. And now almost every major sector is boosting their investment to ramp up their AI investments and applications.
3. Third, human flourishing. AI’s potential to improve individual well-being is unprecedented. From accelerating cures for chronic diseases to democratizing knowledge, AI represents the greatest knowledge-enhancing technology in history with the “capacity to accelerate, mediate, and diffuse human knowledge” faster than ever before. If we get this right, every American—and every human on this planet—will benefit profoundly. Heart disease and cancer accounted for 1.3 million deaths alone last year. AI, machine learning, and advanced data science gives us the greatest chance we have ever had to design cures for those scourges through better drugs and more personalized treatments.

The Policy Choice We Face

Unfortunately, some want to hold back this revolution. We today face the same question we have at the dawn of every other technological revolution: Will algorithmic and robotic technologies be “born free,” or will they be born inside a regulatory cage? The answer remains unclear.

The AI policy debate has exploded in scope over the past few years. In just the first five months of 2025, over 1,000 AI-related bills were introduced across the country. The challenge is that these laws span multiple layers of jurisdiction—state, local, federal, and international—and cover a dizzying array of topics: from model-level regulation to sector-specific rules for healthcare, finance, and transportation, to cross-cutting issues like energy, elections, and jobs. While all these proposals are well-intentioned, this looming tsunami of red tape means compounding compliance burdens for innovators, especially smaller start-ups and open source AI platforms and providers.

At the federal level, the Trump administration has wisely replaced the Biden administration’s fear-based approach with a pro-growth, “try-first” vision in their new “AI Action Plan.” The plan generally embraces permissionless innovation and outlines concrete steps for encouraging AI diffusion and adoption. But success will require congressional support—especially to preempt the worst excesses of state-level overregulation. Without a national framework, a handful of states will effectively set AI policy for the entire country, much as California has done with environmental standards. Just four progressive states—California, New York, Illinois, and Colorado—have introduced a quarter of all AI laws currently pending, and they are looking to regulate quite aggressively.

Federal lawmakers should pass a baseline AI law as well as a privacy bill that both establish guardrails that protect innovation, while also exercising greater oversight over federal agencies that are already regulating AI systems. Congress must not let a patchwork of restrictive state laws or archaic federal regulations smother this technology in its cradle.

How AI Is Already Regulated

Speaking of federal agencies, it is worth remembering that AI is hardly an unregulated Wild West. The United States already has a vast regulatory apparatus—441 federal agencies—plus an extensive body of consumer protection, civil rights, and sector-specific laws. Many leading federal regulatory agencies—FDA, SEC, CFTC, EEOC, FAA, NHTSA, and others—already oversee AI applications in their fields. In some sectors, like commercial drones, overregulation has actually stifled innovation over the past decade.

If needed, we can take steps to ensure agencies have the resources and training needed to address novel AI-related issues, especially cybersecurity matters. But the bottom line is that the tools to address AI risks already exist. The real danger is layering on redundant or premature rules from multiple agencies and jurisdictions that slow AI deployment and undermine the many AI benefits I have outlined.

Conclusion: Betting on Freedom Again

It is true that there is great uncertainty about AI’s trajectory—in policy, in business, in society. The answer to uncertainty, however, is not to freeze, pause, or smother innovation. It is to embrace experimentation, learn quickly, and to adapt continuously as we go along. To beat China and continue to lead the world, we must reject fear and embrace a pro-freedom AI opportunity agenda. Beyond crafting a national policy framework, we need our policymakers to:

• Embrace open source AI innovation and let it blossom globally;
• Ensure diverse, competitive energy markets for AI advancement;
• Win the talent war by attracting the world’s best and brightest data scientists and computer engineers;
• Ensure balanced copyright and data privacy policies;
• Defend freedom of speech and the pursuit of knowledge in the algorithmic age.

This is a technology race we cannot afford to lose. America’s great advantage has always been its innovation culture—our willingness to let entrepreneurs try, fail, and try again without first asking permission from a regulator. That spirit won the internet revolution. It can help the United States win the AI revolution, too.

The AI policy path forward for America is clear: double down on the freedom to innovate. The alternative is to let others—perhaps authoritarian rivals—write the script for the future. I know which world I’d rather live in.