For over a year now, sober-minded observers have warned that the AI bubble can’t last for much longer. Unfortunately for everybody, it’s managed to outperform even the most cynical doomsday forecasts, to the degree that the US economy is now in even worse shape than it was right before an infamous downturn in the late 1920s.

That’s according to the Telegraph‘s economics columnist Russ Mould, who notes that the overvaluation of US stocks has passed the level that brought the stock market to its knees to kick off the Great Depression.

As Mould observes, US stocks trading on the S&P 500 are now priced 41 times their average earnings over the last decade, a measure called the Shiller CAPE ratio, named after laureate economist Robert Shiller. Said another way, investors are paying $41 for every $1 of average annual profits the S&P 500 has generated over the past decade.

It’s more or less the financial equivalent to the tide receding before a freak tsunami. While it can’t tell us exactly when the wave will come, or how big it will be when it does, a Shiller ratio this high indicates danger; the fact that at some point in the not-so-distant future, things are going to get rough. To put today’s ratio of 41 into context, the historical average stands at about 17.3, according to Investing.com. On Black Tuesday, the day that kicked off the worst financial catastrophe in modern history, the ratio was 32.5 — 8.5 points lower than it stands today.

The justification for those unbelievable prices relies almost entirely on AI. Nearly the entire US financial apparatus has swallowed the tale that AI is about to unlock a permanent, global revolution in productivity and profit, even as the technology turns out to be worthless in the vast majority of workplaces.

Of course, similar stories were told about the internet in the late 1990s, and about electrification in the late 1920s. In the end, we incorporated those technologies — but not before the hype around those technologies collapsed into financial disaster, dragging the economy kicking and screaming into a hard-reset.

A contradiction of this size and scope can only be resolved in one of two ways: either earnings actually catch up to the pie-in-the-sky AI fantasies, or the market closes the gap between financial valuation and productivity the hard way. Given that the former scenario looks less likely with each passing day, the next decade could give us a first-hand look at the state of the world circa 1930.

More on AI economics: Employees Are Using Their Jobs’ Super-Expensive AI Tokens for the Most Hilariously Pointless Tasks Imaginable