Congress needs the discipline to distinguish between a guardrail and a roadblock — and understand who benefits from each.
Congress needs the discipline to distinguish between a guardrail and a roadblock — and understand who benefits from each.
T here is a race underway that will help determine who leads the world for the next generation. It isn’t happening on a track. There won’t be a finish line or a medal ceremony. But make no mistake: The United States and China are racing for dominance in artificial intelligence. Beijing would be perfectly happy to watch Washington scare America into slowing itself down. In fact, China isn’t just cheering for American regulation to stifle innovation. It is actively working to exploit Americans’ fears about emerging technology.
I spent a year serving on the bipartisan House Task Force on Artificial Intelligence. We talked to researchers, national-security officials, startups, academics, and companies actually building this technology. Our task force consulted more than 100 experts and ultimately produced 85 recommendations for Congress. I came away convinced of two things: AI presents real risks, and losing the AI race presents much bigger ones.
That distinction matters because Washington has a bad habit of confusing caution with action. Something new emerges. Someone predicts catastrophe. Cable news discovers a new thing to fear because, candidly, fear sells. Congress holds a hearing, members demand that “something be done,” and pretty soon, Washington is writing rules for a technology most of the room has never used.
That is not strategy — certainly not an America First strategy. That is panic on U.S. government letterhead. And in the competition with China, panic has consequences. The country that leads in artificial intelligence won’t simply have better chatbots. AI will increasingly shape military planning, intelligence, cyber defense, logistics, medicine, manufacturing, agriculture, energy, finance, and scientific discovery.
We need to stop pretending our only choices are regulating AI into submission or letting Silicon Valley run wild. They aren’t. We can protect Americans and still build. We can punish fraud and bad actors without punishing innovation. We can protect children from predatory uses of AI without making it impossible for the next American startup to compete. In other words, we can establish rules of the road without paving the road directly to Beijing.
China does not need some technological breakthrough that makes American innovation disappear. It benefits every time America makes itself slower, more expensive, and more afraid to experiment. And fear is something Beijing understands very well. Chinese influence operations don’t always need to convince Americans that China is wonderful. Sometimes the more effective strategy is making Americans distrust one another, distrust our institutions, and fear our own future. That is the point of cognitive warfare: You don’t always have to defeat your opponent if you can convince your opponent to defeat himself.
Artificial intelligence makes that battlefield even more complicated. Deepfakes can impersonate public officials. Voice cloning can target seniors. Generative AI can manufacture propaganda at extraordinary scale. Bots can make a fringe idea look like a national movement. False images can travel around the world before the truth has finished putting its shoes on.
Those threats are real. I’ve sat through the briefings. I’ve seen what these systems can do. I don’t dismiss them. But the answer to China’s use of emerging technology cannot be for America to become afraid of emerging technology. Imagine applying that logic to every previous technological revolution. Our adversaries have airplanes, so we stop building airplanes? They have computers, so we regulate computers until nobody wants to build them here? They use the internet for propaganda, so America retreats from the internet? We would recognize the absurdity immediately.
Yet Washington is flirting with exactly that mentality on AI.
The answer to legitimate concerns is not retreat or the heavy hand of government. It is American strength. Build better models. Build better chips. Build more energy. Harden our infrastructure. Protect American data. Develop tools to identify synthetic content and foreign influence operations. Give law enforcement the tools to combat AI-enabled fraud. Invest in the researchers and entrepreneurs who will build the next generation of technology here — not somewhere else.
That is the direction the Trump administration has pursued: accelerating innovation, expanding AI infrastructure and energy capacity, and strengthening American technological leadership.
Congress also needs to understand the technology it is trying to govern.
As a millennial, I was born in an analog era but grew up digital. I use AI every day. I’ve built AI agents myself. That experience matters when lawmakers are considering rules governing a technology moving this quickly. Yet some of Washington’s loudest calls for regulation come from people with remarkably little firsthand experience using the technology.
Then there is another strange dynamic: Some of the biggest companies in AI are themselves asking Washington for regulation. Congress should ask why. Large incumbents can absorb compliance costs that startups cannot. A regulatory regime requiring enormous legal teams, technical infrastructure, reporting systems, or capital expenditures may be manageable for the biggest companies in the world while becoming a barrier to entry for everyone else.
That doesn’t mean every regulation is bad. It means Congress needs the discipline to distinguish between a guardrail and a roadblock — and understand who benefits from each. Draw the chalk lines. Establish the boundaries. Protect Americans from fraud, exploitation, and abuse. Hold bad actors accountable. Then let America do what she does best: compete and win.