Suits won't quit AI spending, even if they can't prove it's working

Leaders double down

It has been reported that KPMG's global survey of 2,110 business leaders found executives intend to keep AI at the top of their spending lists even when measurable returns are thin on the ground. Sixty-five percent say they will maintain investment whether they see immediate ROI or not, 70 percent of UK leaders expect AI spend to stay high through a downturn, and 94 percent plan to deploy AI agents. Call it what you like — KPMG's Leanne Allen describes the shift as seeing AI "as a strategic enabler for enterprise‑wide transformation." Bold words. Big bets.

The messy math

Companies can measure wins in pockets — productivity (76 percent), quality and performance (71 percent), speed and accuracy of decision‑making (67 percent) and profitability (64 percent) — yet only 14 percent are confident about proving the business value of analytics feeding the C‑suite. It has been reported that other studies complicate the rosier picture: a nearly 6,000‑exec survey found more than 80 percent detect no clear impact on employment or productivity from AI, and Gartner recently said only 28 percent of infrastructure AI use cases fully succeed with ROI. Meanwhile, vendors are fronting much of the spend: Gartner forecasts AI investment to reach $2.52 trillion in 2026. Big money. Big expectations.

Mixed messages and real risks

The emotional heart of this story is the tension — tech teams getting mixed signals from the board. Heads want transformation; engineers want proof. Pressure is real: a Harris Poll commissioned by Dataiku reportedly found 98 percent of tech leaders feel board pressure to demonstrate ROI, and 71 percent of CIOs expect their AI budgets could be cut if targets aren’t met. So what happens when the spreadsheet speaks louder than the strategy memo? Prune the garden, or risk an overgrown jungle. Either way, enterprises are choosing faith over faithfulness to the numbers — for now.

Sources: The Register