The Beautiful Baby and the Faustian Bargain: Unpacking Trump's AI Action Plan
There's a moment that captures everything you need to know about where America stands on artificial intelligence (AI). Picture this: a Washington auditorium last week. Tech luminaries filling gilded seats, and there at the podium stands President Donald Trump calling AI "a beautiful baby that's born".
The metaphor is certainly odd, but is also accidentally profound. Babies are innocent, full of potential, requiring careful nurturing. They're also unpredictable and capable of fundamentally changing everything about your life in ways you never anticipated. Trump's AI Action Plan doesn't ask whether we should raise this baby (that ship has sailed). It asks whether we're prepared for what kind of adult it becomes when Silicon Valley writes the parenting manual and baby books.
I've watched a lot of political theater, but this felt different. Something both seductive and unsettling about watching the president flanked by tech titans, all nodding along to the idea that the only thing standing between America and AI dominance is bureaucratic red tape and what they call "woke" guardrails.
You could almost smell the disruption in the air. Or maybe that was just the scent of regulatory capture.
Silicon Valley's Victory Lap
Trump's AI Action Plan represents six months of the most sophisticated lobbying campaign in recent memory. Over 10,000 public comments, they say, though one suspects most came from people who can afford K Street addresses, and THC Group, of course. The plan identifies over 90 Federal policy actions across three pillars: Accelerating Innovation, Building American AI Infrastructure, and Leading in International Diplomacy and Security.
Translation: let tech companies do whatever they want, build whatever they need, and export it everywhere they can.
President Biden's approach, by contrast, focused on safety reports and bias prevention. Trump scrapped that on day one, calling it innovation-killing red tape. Fair enough. But what replaced it reads like a Silicon Valley wish list with a presidential seal slapped on top.
“We will be adding at least as much electric capacity as China. Every company will be given the right to build their own power plant.”
Pillar 1: Deregulation with Extra Steps
The first pillar promises to accelerate innovation by removing barriers. Sounds reasonable until you realize those "barriers" include things like environmental reviews and state-level consumer protections.
The plan recommends the federal government "consider a state's AI regulatory climate" when distributing funding. In Washington speak, this means: nice federal dollars you've got there, shame if something happened to them because you tried to regulate our AI companies.
Trump was blunt about it: "we have to have a single federal standard, not 50 different states regulating this industry." Because nothing says small government quite like threatening to defund states that dare to think differently about things like consumer protection, fraud prevention, and education.
The crown jewel here is eliminating "woke AI." Federal contracts will only go to companies whose AI systems are "objective and free from ideological bias." Which raises the obvious question: who exactly gets to define objective? Spoiler alert: it won't be the philosophy or ethics departments.
Pillar 2: Drill, Data Center, Repeat
"We will be adding at least as much electric capacity as China," Trump declared. "Every company will be given the right to build their own power plant."
The audacity is breathtaking. We're talking about doubling America's industrial energy infrastructure to feed the computational hunger of artificial intelligence. Environmental regulations? Those are "radical climate dogma" that need to go.
It's a Faustian bargain with a silicon chip: unlimited power for unlimited processing, and if the planet gets a little warmer in the process, well, that's the price of American AI supremacy.
Pillar 3: Digital Empire
The third pillar sounds almost quaint: export American AI technology to friends and allies. But read between the lines and you're looking at something more ambitious. Full-stack AI packages including hardware, models, software, applications, and standards.
We're not just competing with China anymore. We're trying to make American AI the global standard the way the dollar became the world's reserve currency. Except this time, instead of Bretton Woods, we've got Silicon Valley calling the shots.
The Standing Ovation
Tech industry reaction has been rapturous. Nvidia's Jensen Huang called Trump "America's unique advantage that no country could possibly have." When CEOs start talking about presidents like they're competitive assets, you know the capture is complete.
The applause makes sense. This administration has delivered everything tech wanted: deregulation, infrastructure spending, export promotion, and a promise to get those pesky states to stop asking uncomfortable questions about algorithmic bias in hiring and lending.
Even Sam Altman, who spent years calling for AI regulation, now holds joint press conferences with the president. Nothing changes minds quite like market access.
But there's a trap here. The plan's requirement that AI systems be "ideologically neutral" sounds reasonable until you remember that neutrality is itself an ideology. And once governments start defining what counts as unbiased AI, every other government will want the same power.
The Resistance
More than 80 groups signed a counter-proposal they're calling the "People's AI Action Plan." Labor unions, environmental groups, civil rights organizations, all basically saying the same thing: maybe the people building AI shouldn't be the only ones deciding how it gets regulated.
Congress already rejected Trump's approach once, with a 99-1 Senate vote against preempting state AI laws (we previously wrote about that proposal). Seventeen Republican governors opposed it. But what Congress won't do, executive action apparently will.
"The AI industry is deeply concerned," said Neil Sahota, who advises the UN on AI. Which is a polite way of saying even the tech companies are starting to realize they might have asked for more than they bargained for.
Three Approaches to Global AI Power
The world's major powers have chosen radically different paths to AI supremacy.
America is going full Silicon Valley: deregulation, corporate freedom, and market-driven innovation. The bet is that unleashing tech companies will naturally produce both profits and patriotism. Priority one is speed. Priority two is beating China. Everything else is negotiable.
Europe took the lawyer's approach: comprehensive regulation, consumer protection, and individual rights. They want to win by setting the rules everyone else has to follow. Think of it as regulatory imperialism with a human rights face.
China chose the authoritarian playbook: massive state investment, centralized control, and AI in service of political power. Innovation happens, but only when it serves the party's vision of technological dominance.
The result? Europe wants to be the world's AI referee. China wants to be the world's AI overlord. America wants to win by letting corporations do whatever they want and hoping they remember which flag to salute.
The Upside
To be fair, the plan does get some things right. AI development is a race, whether we like it or not. China is investing hundreds of billions in state-directed AI research. If we're going to compete, we need infrastructure, speed, and scale.
America does have genuine advantages: the best chips, the best talent, the best cloud services. The plan leverages these strengths in ways that could cement long-term dominance.
And yes, some regulations are genuinely stupid. If permitting delays cost us years while China builds thousands of data centers, that's a strategic disaster.
The Downside
But every strength here contains the seeds of disaster. "Move fast and break things" works great for app development. Less great when the thing you're breaking is environmental protection or democratic oversight.
The climate implications alone are staggering. AI data centers already consume enormous amounts of energy. This plan essentially promises to build as many as we want, environmental consequences be damned.
The "anti-woke" provisions could backfire spectacularly. What happens when other countries decide American AI is too "capitalist" or "imperialist" for their procurement? Are we ready for a world where every government gets to define ideological purity for AI systems?
Racing Toward What, Exactly?
Here's the question nobody wants to answer: what does winning actually look like?
The plan talks endlessly about dominance, supremacy, beating China. But dominance in service of what? A world where every human interaction is mediated by AI systems designed in Silicon Valley? A future where computational power matters more than human judgment?
"What are we racing toward when it comes to AI?" one expert asked. "The White House's AI Action Plan hasn't resolved this question."
Maybe that's intentional. It's easier to get consensus on beating China than on what kind of society we want AI to create.
The Gilded Age Playbook
We've been here before. In the 1880s, railroad barons convinced Washington they needed massive subsidies, land grants, and regulatory exemptions to build the infrastructure America needed to compete globally.
They weren't wrong about the infrastructure part. The transcontinental railroad transformed America. But it also created massive corruption, environmental destruction, and wealth concentration that took decades to unwind.
The AI Action Plan feels like the same playbook with better graphics. Promise national greatness through corporate liberation, and worry about the consequences later.
What Comes Next
We need a different conversation. Not reflexive opposition to everything tech wants, but not reflexive embrace either. The question isn't whether we should compete in AI (we should). The question is whether we can compete without abandoning the values that make competition worthwhile.
That means real public input, not 10,000 comments from lobbyists and policy wonks (like us). It means transparency about how these systems work and accountability when they don't. It means remembering that our democratic values could be our competitive advantage, not our handicap.
Most of all, it means humility. AI is powerful precisely because it can do things we don't fully understand. That should make us more careful, not more reckless.
“We have maybe five years to get this right. The decisions we make now about AI governance will echo for decades. We can choose thoughtful stewardship or reckless speed. We can choose democratic oversight or corporate capture.”
Growing Up Fast
Trump called AI "a beautiful baby that's born." Fair enough. But babies grow up to reflect the values of the people who raise them.
Right now, we're letting Silicon Valley write the parenting guide while Washington writes the checks. The plan assumes this will produce an AI ecosystem that serves American interests, but there's little evidence that corporate interests and public interests naturally align.
The beautiful baby is indeed growing up fast. The question is whether we're raising a responsible adult or a powerful sociopath.
We have maybe five years to get this right. The decisions we make now about AI governance will echo for decades. We can choose thoughtful stewardship or reckless speed. We can choose democratic oversight or corporate capture.
What we can't choose is to opt out. The baby is born, it's growing, and it's going to reshape everything whether we guide it thoughtfully or not.
The least we can do is make sure we're raising it to be something we can live with.
Sources, Resources, and Suggested Reading:
Sources
“Winning the AI Race: America’s AI Action Plan,” White House (Fact Sheet and Full Plan), July 2025
“Executive Order 14179: Removing Barriers and Accelerating American Leadership in Artificial Intelligence,” White House, January 2025
“Executive Order 14110: Safe, Secure, and Trustworthy Artificial Intelligence,” White House, October 2023
“European Union Artificial Intelligence Act,” European Commission, 2024
“China’s Next-Generation Artificial Intelligence Development Plan,” State Council, PRC, updated 2025
“AI Policy - United States,” Future of Life Institute
“Department of Commerce: Export Controls on AI Technology,” U.S. Department of Commerce, April 2024
“NBC News/SurveyMonkey Poll on AI Policy,” June 2025
“Morning Consult AI & Tech Trend Tracker,” July 2025
“RAND: AI Proliferation Risk Assessment,” RAND Corporation, 2024
“Safe, Secure, and Trustworthy AI: OECD Recommendations,” OECD, 2023
“White House Fact Sheet: President Trump’s Plan to Secure American AI Leadership,” July 2025
“State Media Coverage of U.S. AI Export Plans,” China Daily, July 2025
“ACLU Statement on AI Regulatory Rollback,” July 2025
“AFL-CIO: AI and Labor Concerns Press Release,” July 2025
Suggested Reading
“The Coming Wave: Technology, Power, and the 21st Century’s Greatest Dilemma,” Mustafa Suleyman, 2023
“Weapons of Math Destruction: How Big Data Increases Inequality and Threatens Democracy,” Cathy O’Neil, 2016
“Atlas of AI: Power, Politics, and the Planetary Costs of Artificial Intelligence,” Kate Crawford, 2021
“Artificial Intelligence: A Guide for Thinking Humans,” Melanie Mitchell, 2019
“AI 2041: Ten Visions for Our Future,” Kai-Fu Lee and Chen Qiufan, 2021
“Making AI Work: Policy, Practice, and the Future of Artificial Intelligence,” Brookings Institution Panel Discussion, March 2025
“Artificial Intelligence and Life in 2030,” One Hundred Year Study on Artificial Intelligence (Stanford University), 2016
“AI Superpowers: China, Silicon Valley, and the New World Order,” Kai-Fu Lee, 2018
“Governing Artificial Intelligence: Upholding Human Rights & Dignity,” United Nations, 2024