Layoffs and AI were supposed to cut costs. For many companies, they did the opposite

Oh, how wonderfully predictable. A fresh KPMG survey has revealed that senior executives are experiencing "invoice shock" from rising AI costs — because, of course, the technology that was supposed to make everything cheaper has somehow managed to do the opposite. The heavy computing demands and usage-based pricing that were meant to be a minor footnote in the business case are now eating into the very savings that justified laying off thousands of workers in the first place. One can almost hear the collective sigh of CFOs everywhere.

The logic was flawless on paper: replace expensive humans with cheaper algorithms, and watch the profits roll in. What the spreadsheets omitted, however, was that running those algorithms requires enough electricity to power a small city and enough GPUs to make Nvidia blush. The consultants at KPMG — those same fine folks who probably helped companies design their "efficiency through automation" strategies — now get to charge for the follow-up work of explaining why the numbers didn't work. It's the consultancy circle of life, and it's beautiful in its naked cynicism.

So here we are: companies have culled their workforces, signed massive cloud contracts, and are now discovering that the promised savings have evaporated into a haze of API calls and compute cycles. The "invoice shock" won't just be a footnote in the boardroom; it will likely trigger another round of layoffs to recoup the unexpected costs of the first round. The KPMG survey is not a warning shot — it's a post-mortem dressed as a trend report. But by all means, keep telling us that AI is here to liberate us from drudgery. The invoice disagrees.