Like most of you, I have learned to be suspicious whenever the market reaches near-unanimous agreement on anything. Except Gangnam Style (and its 6B+ YouTube views).
Since starting in the industry 25+ years ago, Procurement has had its share of technology moments over that time. Some changed the function dramatically (but over years and decades, not days and months), while others produced a great deal of noise and attention, ending with a whimper. AI is different in both its potential and its pace, but one thing about this cycle feels very familiar: what the market believes is happening and what our research shows is actually happening are not exactly the same.
Spend five minutes at any procurement conference today and you might conclude that AI has already taken over the function. Everyone has a strategy, every provider is launching agents, and every CPO roadmap leads down the road to procurement intelligence. The CPO’s interest in AI (“extremely high”) and perspective of it (“transformational”) appears almost universal, and the adoption numbers bear this out.
Ardent Partners’ latest research, based on responses from 300+ global CPOs and senior procurement leaders, finds that roughly 90% of procurement organizations are now exploring, piloting, or deploying AI in some capacity. Three years ago, a number like that would have seemed implausible.
So while there is huge momentum, what I find more interesting is that fewer than one in four organizations have moved beyond pilots and trials and are now using AI in live operations. That data point is where we really are as an industry. For example, almost no CPO today says that AI is fully-embedded into their procurement function. Procurement has embraced AI remarkably quickly. (Let’s put aside, for now, what fully embedded means to you and whether or not that is the desired future state). Like any technology deployment, the hardest work with AI isn’t the RFPs and piloting, it is figuring out how to leverage it to create a better operation.
Procurement’s Tactical Burden Is Today’s Baseline
Amidst all the AI questions, this study asked CPOs a more basic question – How does your procurement team spend its time. We asked CPOs to calculate the amount of tactical work their team does overall and 87% told us that tactical demands consume a meaningful share of the work.
Inherently, anyone working at a large enterprise understands that tactical work comes with the territory. Whether it’s a salesperson entering contact info into a new CRM, an AP staffer validating account numbers from an OCR intake or the procurement specialists who spend hours copying requisition data from emails or spreadsheets into ERPs, generating POs, and chasing internal approvals for low-dollar items. That 87% number gives us new context bc it describes the actual environment where AI is being introduced.
For years at the annual CPO Rising Summit (should we bring it back?), CPOs talked about shifting their teams toward more strategic work, and the great progress that had been made. There has been progress and procurement has become more influential, more data-driven, and more directly aligned with executive priorities, but the constraint of tactical work still weighs on most.
At the same time, CPOs are placing AI on top of that operation and expecting something very different to emerge. I think this is where much of the current AI discussion gets ahead of itself. AI can automate individual activities inside an inefficient process, but the larger opportunity is when organizations begin redesigning their work around what these systems can do. That, of course, requires better data, clearer processes, defined governance, etc. and the organizational capacity to manage all of it.
What’s different with AI vs the first launch of procurement technologies is that which segment is ahead has flipped. In 2000, ProcureTech was far behind the actual needs and requirements of procurement organizations (they closed the gap, but not quickly). Today, AI technology is advancing at extraordinary speed while procurement operations are moving more slowly. What I called the “execution gap” in the report sits between the two. And this becomes increasingly important because procurement leaders are not lacking in ambition.
Sixty-one percent of the CPOs in 2026 see AI primarily as a productivity and scale enabler, while nearly three quarters consider AI capabilities important or critical when evaluating procurement technology. Their challenge is building an organization capable of capturing the value that they believe is already there.
AI Ambition Is Ahead of Readiness
This gap became much clearer when we mapped the research across the two dimensions at the center of our AI-First Procurement framework: AI ambition and operational readiness. Ambition captures the extent to which procurement organizations are embracing AI, investing in it, and planning to expand its role. Readiness looks at whether the underlying operation, including its data, processes, governance, technology environment, and people, is actually prepared to support that ambition.
The results were pretty striking. Roughly two-thirds of procurement organizations landed in what we call the Overreaching quadrant. These are organizations with meaningful AI ambition that are already moving forward with pilots, deployments, and, increasingly, plans for agentic systems, but whose operating foundations have not advanced at the same pace. By comparison, only about 23% of organizations currently sit in the AI-First quadrant, where ambition and readiness are much more closely aligned.
I do not view “Overreaching” as a pejorative. In fact, given the speed of this market, I would have been surprised if the results looked much different. When a technology as consequential as AI emerges, business leaders should move quickly, test deliberately, but aggressively, and figure some things out as they go. Procurement teams are and should be doing that, but what matters from my perspective is if CPOs understand where the current gaps are and how to build on the lessons learned from their pilots.
Each round of pilots will help procurement teams learn more about AI, but it will not fix the underly problems linked to data, processes, teams and governance. And those issues are what will determine who quickly AI capabilities can be scaled.
The questions I think CPOs should focus on now are challenging and consequential.
- Where should AI be allowed to act autonomously?
- What information should agents be able to access?
- Which decisions still require human judgment?
- Where do controls sit when work begins moving across multiple systems and agents?
- And perhaps most importantly, what should the procurement organization itself look like when technology begins successfully tackles the tactical workload that has shaped the function for decades, and still defines it in 2026?
The market has plenty of “AI ambition,” but it’s the answers to these questions that will define the next phase of procurement AI.
This article is the first in a multi-part series based on Ardent Partners’ report The Path to AI-First Procurement: Closing the Gap Between AI Ambition and Execution, one of the most comprehensive studies conducted to date on how AI is reshaping procurement. The full report, produced by Ardent Partners in partnership with Ivalua, is available from Ivalua.
