The AI Race We're Running
What the White House AI Action Plan reveals about America, China, and power.
On 23rd July 2025, the White House released a twenty-five-page document with the sort of title that telegraphs its own urgency: “America’s AI Action Plan: Winning the Race.”
The subtitle alone should have been a warning enough. When governments start talking about races, particularly races they insist they must win, and we learn to pay attention not just to what they say, but to what they seem unable (or unwilling) to say.
The document opens with a quote from President Donald Trump, set apart in italics like a biblical epigraph:
“Today, a new frontier of scientific discovery lies before us, defined by transformative technologies such as artificial intelligence… it is a national security imperative for the United States to achieve unquestioned and unchallenged global technological dominance.”
Unquestioned and unchallenged. The words sit on the page with the blunt force of doctrine.
What we have here, dressed in the measured language of policy, is a manifesto. Not for AI as such as but for AI as an instrument of American power. The document’s three pillars: accelerate innovation, build infrastructure, lead internationally, read less like a roadmap for technological development and much more like a battle plan for a digital empire.
Consider what the authors choose to emphasise. Under the first pillar, they promise to “remove red tape and onerous regulation.” They pledge to revise AI standards to “eliminate references to misinformation, Diversity, Equity and Inclusion, and climate change.” They insist that government AI systems must be “objective and free from top-down ideological bias.”
The irony would be almost comic if it weren’t so revealing. A document that positions itself as the antidote to ideological bias opens by declaring America’s intention to achieve “unquestioned and unchallenged” dominance. A plan that claims to pursue objective truth begins by eliminating references to climate change, as if the planet’s warming were a matter of political opinion rather than a measurable fact.
But perhaps this isn’t irony at all. Perhaps this is simply honesty of a particular American sort, i.e. the honesty that mistakes power for principle, that confuses the ability to impose one’s will with the possession of truth.
The document’s treatment of workers offers another window into its authors’ thinking.
“AI will improve the lives of Americans by completing their work — not replacing it”, they write, the em dash adding a note of breathless reassurance.
They promise that “American workers are central to the Trump Administration’s AI policy”, that the
“breakthroughs in medicine, manufacturing, and many other fields that AI will make possible will increase the standard of living for all Americans.”
These are the sort of promises politicians make when they know the opposite is likely true. When they speak of workers being central to AI policy, they mean that workers will be central to AI policy in the way that coal is central to coal policy, like a resource to be managed, optimised and potentially depleted.
The plan’s second pillar concerns infrastructure, and here the language grows more concrete, more urgent.
America must “build vastly greater energy generation than we have today.” The government will “reject radical climate dogma and bureaucratic red tape.” The rallying cry: “Build, Baby, Build!”
This is where the document’s priorities become clearest. Not in its promises about workers or its rhetoric about values, but in its frank acknowledgement that artificial intelligence requires power (literal electrical power) on a scale that will demand fundamental changes to American energy production.
The authors understand, in a way that much AI commentary does not, that we aren’t simply talking about software. We’re talking about industrial transformation that will reshape the physical landscape.
The third pillar, international leadership, makes explicit what has been implicit throughout:
This is about China.
“The United States is in a race to achieve global dominance in artificial intelligence”, the authors write. “Whoever has the largest AI ecosystem will set global AI standards and reap broad economic and military benefits.”
The plan calls for “countering Chinese influence in international governance bodies”, “strengthening AI compute export control enforcement”, and ensuring that America’s allies “are building on American technology.”
This isn’t diplomacy; this is containment.
What strikes me the most forcefully about this framing is its curious mixture of confidence and anxiety. The authors are certain that America can win this race, yet the entire document vibrates with the fear that America might not.
They speak of “unquestioned dominance” whilst simultaneously acknowledging that this dominance is very much in question. They promise that AI will usher in “a new golden age of human flourishing” while devoting most of their energy to ensuring that this golden age remains exclusively American.
There are things this document doesn’t say, absences that feel almost deliberate.
For all its talk of “American values”, it offers no sustained discussion of what those values might mean in practice. For all its promises about objective truth, it provides no framework for distinguishing truth from propaganda. For all its concern with AI safety, it focuses almost entirely on the safety of American interests rather than the safety of people.
The document mentions “AI interpretability” and “robustness breakthroughs”, but these appear as technical problems to be solved rather than as fundamental questions about the nature of intelligence and control.
There is no discussion of algorithmic bias except as something to be avoided in government procurement. There is no mention of the environmental costs of the massive data centres that the plan envisions, only the promise to “reject radical climate dogma.”
Most telling, there is no acknowledgement that artificial intelligence might pose risks that can’t be managed through American technological superiority. The authors seem to believe that if America can build the most powerful AI systems, it can thereby control the consequences of those systems.
This is the sort of thinking that builds empires and loses them.
In the end, what we have in “America’s AI Action Plan” is a document that reveals more than it intends.
It tells us that the US government sees AI primarily as a weapon in a global competition for technological supremacy.
It tells us that this government is willing to reshape energy policy, trade policy and international relations in service of that competition.
It tells us that concepts like “objective truth” and “American values” have become rhetorical tools rather than governing principles.
What it doesn’t tell us is whether this race is worth winning. Or whether the act of running it might change us in ways we haven’t yet begun to imagine. But perhaps that is a question for another day, and another kind of document entirely.
This one has a race to win.
(The full document is available here.)
Hi, I'm Miriam - an AI ethics writer, analyst and strategic SEO consultant.
I break down the big topics in AI ethics, so we can all understand what's at stake. And as a consultant, I help businesses build resilient, human-first SEO strategies for the age of AI.
If you enjoy these articles, the best way to support my work is with a free or paid subscription. It keeps this publication independent and ad-free, and your support means the world. 🖤