New: the shifts that come back across the latest research. See what to focus on

Horizon: now

AI spending must now prove its value

AI is moving from cheap experiments to ongoing usage bills, making spending ownership and verified business results essential.

In the report's words: “AI is moving from subsidized experimentation to usage-based costs that must earn measurable business returns (p6–7).”

The number that matters

72%p16

Organizations receiving an unexpected AI bill last year

Where to focus

  1. 1Assign AI spending to clear owners and teamsp43
  2. 2Reduce repeated information and wasteful AI task loopsp14
  3. 3Link AI spending to verified business resultsp54

The details

Who it affects

CFOs and technology leaders at enterprises; employee size unspecifiedp43–46

What holds them back

  • CFOs struggle to forecast AI spending because usage costs lack clear attribution and ownershipp43
  • Technology leaders struggle to control AI bills because repeated context and agent loops multiply consumptionp14
  • CEOs struggle to justify AI investment because spending is not tied to verified business outcomesp54

The numbers

  • 72% of surveyed organizations received an unexpected AI bill in the past yearp16
  • Combined engineering improvements delivered 40–60% lower token spending at broadly equivalent qualityp20

What people search for

  • why is my AI bill so high
  • track AI spending by team
  • set AI spending limits
  • reduce AI API costs
  • cheaper AI without losing quality
  • measure if AI saves money