Procurement teams are increasingly being pulled into a problem that used to sit mainly with IT and finance: how to keep artificial intelligence spending visible, controllable and tied to business value.
In a recent post, Aashima Gupta argued that the issue is no longer simply how much organisations are paying for AI, but how little many of them can actually see. As AI tools spread through direct API contracts, cloud services and features embedded inside software-as-a-service pr...
Continue Reading This Article
Enjoy this article as well as all of our content, including reports, news, tips and more.
By registering or signing into your SRM Today account, you agree to SRM Today's Terms of Use and consent to the processing of your personal information as described in our Privacy Policy.
oducts, the spend is often scattered across departments and budgets, making oversight difficult.
That fragmentation is now showing up as a broader governance challenge. McKinsey has said many organisations still lack a consolidated view of AI costs, with spending spread across copilots, model contracts, experimentation environments and business-unit purchases. The consultancy estimated that a meaningful share of AI spend is often unaccounted for, which means overruns are frequently discovered only after consumption has already happened.
The problem is not limited to visibility. Recent reporting from ITPro described a wave of heavy AI usage by employees, sometimes dubbed “tokenmaxxing”, which is pushing bills higher in much the same way early cloud adoption once did. The publication said AI now accounts for a sizeable slice of cloud expenditure, with a portion of that spending wasted, and noted that more than half of companies still lack clear accountability for AI costs.
Flexera’s 2026 State of ITAM Report paints a similar picture. It found that only 31% of organisations have accurate visibility into AI software spending, even as nearly half are tracking AI as part of their software budgets. The same report said 59% of respondents had seen wasted AI spend rise year on year.
For procurement leaders, the emerging answer is not blunt cost-cutting. It is a more disciplined operating model. Gupta set out five practical priorities: identifying where AI spend enters the business, improving transparency when contracts come up for renewal, assigning ownership for consumption, building a consistent taxonomy for AI spend and giving executives regular reporting.
That approach aligns with a wider push towards FinOps-style discipline. ITPro said cross-functional financial management principles, already used in cloud cost control, are increasingly relevant to AI. TechRadar, meanwhile, has argued that companies often focus too narrowly on short-term savings and neglect the full lifecycle cost of AI, including energy, infrastructure and compliance demands.
The stakes are high because many AI programmes never progress far enough to justify their spend. TechRadar reported that more than half of AI projects are abandoned after proof of concept, often because organisations fail to account for the full cost of deployment and maintenance.
Taken together, the message from consultants, software asset managers and industry commentators is clear: as AI becomes a normal line item in enterprise technology budgets, procurement is moving from negotiation to governance. The organisations most likely to benefit from AI may not be those that spend the most, but those that can trace every pound, assign ownership and decide, with confidence, which uses are worth keeping.
Source: Noah Wire Services
