The economy has apparently found a promising new support beam, and it looks suspiciously like an expensive computer chip. Wonderful. Before we hang the national chandelier from it, perhaps somebody should check the bolts.

On October 13, the IMF raised its 2026 global growth forecast to 3.1%, from July’s 3.0%, AP reports. It projects 2.3% U.S. growth in both 2026 and 2027. These are forecasts, not results already achieved.

Reuters reports that AI investment is supporting growth while the IMF warns about debt, war, and a possible investment reversal. Its adverse AI scenario puts global output 1.1% below the baseline by 2028. That is a conditional scenario, not a prediction of a 1.1% economic contraction.

A slightly brighter headline with a substantial warning label is still a warning label. Politicians who celebrate only the first number are reading the report like somebody buying a used boat: enthusiasm up front, inspection details conveniently underwater.

Growth for whom?

Here is the liberal question I want attached to every speech about technological prosperity: who gets to live inside the good news?

Aggregate growth matters. A productive economy can support higher wages, better services, and more choices. But a national percentage cannot tell you whether a particular family can afford rent, whether a worker has negotiating power, or whether a technological gain becomes a cheaper service rather than a larger executive bonus.

If all you show people is a rising line, do not act bewildered when they ask whether the line accepts payment at the grocery store.

AI can be useful. Better tools can reduce drudgery, improve research, and help small organizations do work they previously could not afford. Treating every advance as a scam would be as lazy as treating every new product announcement as the arrival of economic salvation.

The sensible position is to test the benefits, measure the costs, and keep the sales pitch out of the driver's seat. A forecast should not become an excuse to let investors write the rules for everyone else.

Build a floor, not just a launchpad

My policy preference is straightforward: pair technological investment with durable public protections. Give workers a meaningful voice when jobs change. Fund education and transitions in ways people can actually use. Keep competition strong enough that customers share in productivity improvements.

Also ask who pays for the infrastructure that private projects need. If a venture wants public support, citizens deserve clear terms, enforceable obligations, and a public benefit worth the bill. A ribbon-cutting photo is a poor substitute for a contract.

The IMF’s risk discussion is a reason to take that work seriously before a setback. It is not a reason to declare that a crash is guaranteed. Uncertainty should encourage resilience, not another competition to see who can sound most certain on television.

The same applies to budget choices. Protecting public finances should involve honest scrutiny of tax breaks and powerful interests alongside spending. A government that calls every ordinary household benefit unaffordable should have to explain why wealthy beneficiaries keep finding an available checkout lane.

Democrats should resist offering their own version of the technological miracle tour. Promising that growth will automatically fix distribution is not an economic program. It is a motivational poster with a server room behind it.

We can welcome innovation and still insist that prosperity be broadly usable. Indeed, a technology that improves life should survive questions about wages, costs, reliability, and who bears the risk.

The chip may help hold up the economy. It still cannot eat the grocery bill. Somebody should tell the people writing the victory speech.