America faces one of those rare, historic moments when government, business, schools and families could be working together to meet a truly generational challenge: winning the AI race.

So far, we're blowing it.

Starting next year, AI companies in the U.S. will spend roughly the same in one year as the Manhattan Project, the Apollo moon landings, the Interstate Highway System and the Human Genome Project combined -- about $1 trillion.

That's just the money. The stakes are even bigger: who controls space and warfare, dominates the world economy and shapes the first technology with super-intelligence surpassing our species.

It's easy to blame tech companies or politics for our blurry vision. But arguably the fault lies more in our collective imagination and our shared inability to think of societywide issues and efforts. Society has slowly surrendered all that as it has retreated into tribes, bubbles and screens. American businesses, universities and families have a part to play in the solution.

The awesome technological advances of world-leading American engineers at Anthropic, OpenAI and elsewhere, could be zipping the country further ahead of China. Instead, it's locked in a panicked and confused paralysis across government, business, campuses and the workforce. China, meanwhile, has a state-directed, countrywide plan to put AI into action and lock down the supply chain for future dominance. This threat is present, real and intensifying.

Here's what separates artificial intelligence from every prior technological disruption in American history: We know what's coming before it arrives. This was true two years ago. It's true today.

I run a media company that covers and uses AI aggressively. I speak often with the architects of AI, the elected officials trying to govern it, the university president reacting to it and families and workers struggling to understand it. I am neither an AI cheerleader, nor doomer. I'm a brutal realist: Bad things happen if this gets botched.

Here are seven society-level ideas to shape AI to benefit all.

Step 1: Frame it as a national project, not a technological one.

AI is treated primarily as a business or tech story, a race between rich companies with sci-fi ambitions. That is all real, but incomplete.

A better frame is the one America has used in its finest hours: the moonshot. Think of our mobilization after the Great Depression and in World War II. Or the Marshall Plan to rebuild after. Or those fleeting months post Sept. 11. This could lead to a shared national project with a clearly defined goal (win), a clear competitor to beat (China) and a clear benefit to the country if we win and spread the benefit broadly (shared prosperity).

That framing would give politicians a reason to engage constructively and workers a reason to lean in rather than fear what's coming. It would also give businesses reason to move fast, explain and train better, and help everyone understand the stakes and shifts. And it would give the American public, who increasingly distrust the technology and the companies behind it, a sense of ownership and wonderment rather than exclusion and worry.

Step 2: Build the coordinating infrastructure.

Bring together the best and brightest drawn from the federal government, from the leading AI companies, from business and labor, from economics, from public health, from ethics to form a working group with actual authority. People who are paid to anticipate, not to react haphazardly.

This group would have three primary functions: to map the potential problems and upsides before they hit, build the playbooks before they're needed and level with the American people along the way.

We just witnessed two great examples of what the absence of prior planning produces. We knew in advance AI would code better than humans and then enable highly sophisticated cyberattacks. In both cases, we scrambled improvisationally, after the fact, stoking fear and confusion.

Step 3: Anticipate the displacement problem before it becomes a crisis.

The debate about AI and jobs has become exhausting and largely unproductive, oscillating between "AI will take all jobs" and "AI will create more jobs than it destroys." Neither conversation is terribly useful.

What would be useful is a staged response plan. Right now, AI companies are in an investment phase, one you might not want to slow down since it's what's driving the U.S. lead over China. There's also little evidence of substantial job loss wholly attributable to AI.

But if unemployment takes a sudden turn, prebuilt, pre-debated, pre-legislated solutions would be ready in advance. If unemployment hits 6%, for example, there would be triggers to pull instead of improvised panic when the politics are already on fire. This could include job training programs or temporary mechanisms to slow layoffs.

Dario Amodei warns job loss could be much worse. Elon Musk claims everyone could choose not to work. OK, let's have a plan for that, too.

Some of that work can be new forms of communal, human-centered work. A nurse corps. An eldercare corps. A tutor corps. A community rebuilding corps. Service structures that bring people together, that address real social deficits: the loneliness epidemic, the collapse of local institutions, the teacher shortage, an aging nation. This could be funded by the enormous wealth that the technology would generate, if it generates what its proponents believe.

Step 4: Create a real-time, user-friendly app for jobs

Recently I was talking to business leaders in Phoenix and their lament was not job loss but the inability to find the right workers. Finding talent could be easier and one of the most concrete and achievable ideas would be a national AI labor app.

The concept is straightforward: Pull real-time data from the AI companies on where they need data centers, energy infrastructure, chips, engineers and technicians, and where the data shows current jobs might be threatened or opening up. Spot and adapt to job shifts in near real-time.

Feed that into a public-facing tool that matches workers with jobs and appropriate training: here is where the jobs are, here is what skills are needed and here is who will pay for your retraining.

Meta just funded a small version of the training piece. Google then did, too. Big tech companies will likely be the primary beneficiaries of this transition so they could help build the on-ramp.

Step 5: Prepare business, colleges and families

AI companies know in real time what people are using AI for at work and home. Anthropic even has a public database that captures how people are incorporating AI into their lives.

Businesses, schools and families could benefit tremendously from continuously updated teaching and tools for using AI for their job or personal needs smartly and responsibly.

Imagine, for starters, every college and high school offering basic AI skills, fed by constant updates as the tech evolves. Teachers should definitely retain the right to limit or even outlaw AI in the classroom. But do we really want to send students into the wild and workforce blind to the most essential skills they need?

Step 6: Take the future dangers seriously

The dangers are as knowable as the opportunities. And we could be more vigilant about those too.

Start with what already happened. Anthropic recently built a model so capable it decided not to release it publicly over safety concerns. That moment -- a leading American AI lab concluding its own technology was too dangerous to deploy -- arrived with essentially zero government preparation and no public discussion of what it means or what comes next. So the government used an executive order and export controls to slow it. Cyber is but the first of many threats to come. Scary? Yes, but true.

Biosecurity risk follows the same logic. As models become more capable, the barrier to engineering dangerous pathogens drops. A serious effort would convene the relevant experts to map the scenarios and build response infrastructure before the capability hits.

We might be months away from true recursive self improvement $(RSI)$. Get familiar with this term, because it basically means AI teaches itself and then learns and works on its own. This is when we lurch into AI that can potentially do superhuman things. And go rogue.

In each case, the risk is known now. Preparation is optional. We're choosing not to prepare.

Step 7: Use American leverage to build a global coalition.

China is a real threat in the AI race. Beijing has already designated AI a national strategic priority, aligned its top universities, state companies and military around a single development road map, and is spending at a scale akin to a wartime mobilization.

To stay competitive and dominant we have one choice: Build a global AI framework built on American rules. The Europeans pushed for something like this at the recent G-7.

Our ask -- and offer -- could be clear and enticing: Join a U.S.-led coalition built around American technology, models, rules, and help establish a worldwide AI supply chain and safety protocol. In exchange, member countries would be part of the most powerful economic and technological alliance in the world, more dynamic, more forward-looking and more consequential than the United Nations or NATO.

It would make the global AI ecosystem dependent on American leadership rather than vulnerable to Chinese alternatives. It would entice developing nations to align with the U.S. rather than accept Chinese infrastructure deals. And it would give America a diplomatic instrument that matched the moment.

What's still possible

The window to do this hasn't closed. It has narrowed.

The problems that could have been anticipated two years ago are now arriving in real time: data center backlash, energy strain, some workforce dislocation, cybersecurity vulnerabilities, the early contours of biosecurity risk. Dealing with them reactively is harder, more expensive, and more politically toxic than dealing with them proactively would have been and will be.

But most of the playbook described above is still executable.

China can impose that kind of coordination by decree. They can align government, industry and society toward a single technological objective through instruments that are unavailable and undesirable in a democracy. That's not an option in the U.S. We would need to make a different choice.

Shame on us if we don't.

Jim VandeHei is co-founder and CEO of Axios.

(END) Dow Jones Newswires

July 09, 2026 14:47 ET (18:47 GMT)

Copyright (c) 2026 Dow Jones & Company, Inc.