Back in January, the Congressional Budget Office published a report called "The Demographic Outlook: 2026 to 2056." It deserved front-page treatment and got almost none. The data inside is a bombshell: by 2030, more Americans will die each year than are born. That's four years from now.

There are three things that can fill that gap. More babies. More immigrants. Or more productivity from every worker we've got. The first isn't happening. The second is the most politically toxic debate in the country. And the third involves something few want right now: AI.

What the numbers actually say

American women are having about 1.6 children on average. You need roughly 2.1 just to keep the population steady. That gap hasn't closed since 2007. The CBO doesn't expect it to close in the next 30 years.

Today there are 349 million Americans. By 2056, the CBO projects 364 million. Then the number starts going backward. That's 15 million people added over three decades. For context, the U.S. added more than that between 2010 and 2020 alone.

Take away immigration and the population starts shrinking in 2030. That's less than four years away. And every time the CBO updates its projections, the picture gets worse. Each recent revision has pointed the same direction: smaller, slower, and older.

You can already feel it

This isn't hypothetical. Walk into any hospital, call a contractor, or talk to someone who manages a trucking fleet. The worker shortage is already here.

There are 6.9 million unfilled jobs in America right now. Not because people are lazy. Because the workers don't exist yet. The median registered nurse in this country is 46 years old. A million nurses plan to retire by 2030. Trucking is short 80,000 drivers, and the average driver is pushing 50. Construction needs nearly half a million new workers this year, and 41 percent of the current workforce will hit retirement age by 2031.

That's what a shrinking labor pool looks like before the shrinking officially starts.

Fewer workers means employers bid up wages. Higher wages mean higher prices. And those prices don't come back down when the labor pool keeps getting smaller. The Federal Reserve can raise or lower interest rates all day long. It can't create workers. That's the kind of inflation that monetary policy wasn't built to fix.

So what does fix it? If every year brings fewer hands, every remaining pair of hands has to produce more. That's the unsexy case for artificial intelligence. Not the Silicon Valley pitch about changing the world. The boring, demographic math: fewer workers means each worker has to do more, or the economy contracts. AI reading medical scans at a hospital short on radiologists. AI dispatching freight for a trucking company that can't find drivers. We're staring a future that will need AI to save the future.

Japan has been living this for 50 years. Their birth rate fell below replacement in 1974, their economy has averaged half a percent of growth per year for the last decade, and their national debt sits at 240 percent of GDP. They've tried everything from cash bonuses per baby to free childcare. Their birth rate hit a record low of 1.15 last year anyway.

The immigration math

From 2030 forward, every single person added to the American population will come from immigration. Natural growth turns negative and stays that way for the next 30 years. By 2056, the country will record 1.2 million more deaths than births every year. That's just simple CBO arithmetic.

Enforcement brought net immigration down from 3.3 million in 2023 to about 410,000 in 2025. The border needed order, and it got it. Now the question is what comes next. Enforcement solves one problem. The demographic math is a different one.

Some people look at the birth rate and say the answer is family policy. Fair enough. But France runs the most generous family support system in Europe — the government covers up to 85 percent of childcare costs, offers years of subsidized parental leave, and sends monthly cash payments per child. France's birth rate? 1.62. Exactly the same as ours. The best pronatalist program in the Western world hasn't cracked replacement.

That doesn't mean family policy is pointless. France might be in far worse shape without it. But no single lever solves this alone. You need family policy, legal immigration, and productivity growth working at once.

The strongest objection

Immigration doesn't fix the age problem by itself. Immigrants get old too. A country that depends on immigration to grow still faces a rising number of retirees drawing on Social Security and Medicare, funded by a tax base that has to keep expanding just to stay even. Anyone who waves that away isn't being serious.

But the alternative — no immigration growth, no productivity gains, and a birth rate that hasn't recovered in 20 years — is mathematically worse. The CBO's projections assume no new legislation. Legal immigration reform paired with investment in AI and automation could change the curve. The uncertainty is real. But every revision has gone in the same direction. The trend hasn't reversed in nearly two decades.

The tools exist

America has advantages that most countries facing this problem don't. People still want to come here. We lead the world in artificial intelligence. Our birth rate, while below replacement, is better than Japan's, South Korea's, China's, Germany's, Italy's, and Spain's. We have time to act, but that window is the gap between now and 2030, not now and 2056.

The CBO gave us the math. Three levers: babies, immigrants, productivity. Turn one, and the decline slows. Turn two, and it stops. Turn all three, and America keeps growing. Japan waited 50 years and got stagnation. France spent billions and barely held the line. America still has the advantage of time, talent, and a country people want to come to. But we need to act now, before the math of 2030 forces us down worse paths.