Procurement Expects AI Gains, but Only One in Four Has Redesigned the Jobs

Procurement leaders have made up their minds about AI. They believe it will make their function markedly better within three years. What most of them have not done is change the jobs of the people who are supposed to deliver that improvement, and Gartner now argues that this omission, not the technology, is the thing most likely to decide whether the payoff arrives.

According to the Gartner findings published on October 8, 63% of senior procurement leaders expect AI to significantly improve procurement performance over the next three years. Yet only 25% of procurement leaders have undertaken job and role redesign as a direct result of AI implementation. The data comes from two surveys concluded in April 2026: the Gartner Chief Procurement Officer Survey of 213 senior procurement leaders, and the Gartner C-Suite AI Survey of 1,303 respondents across business functions, including 134 procurement leaders.

Why Gartner Calls It Job Chaos

Gartner’s warning is specific. AI is hollowing out standardized, middle-skilled work while reshaping responsibilities, workflows, and required skills across the function. When those changes happen without a deliberate redesign, the organization drifts into what Gartner calls job chaos: people whose job descriptions no longer match what they do, talent mismatches, lower morale, and weaker competitiveness.

“Successful AI implementation at the functional level is fundamentally a role redesign challenge,” said Fareen Mehrzai, Senior Director Analyst in Gartner’s Supply Chain practice. Her point is that chief procurement officers who wait leave employees without a clear path forward, and leave the organization without the talent it needs to capture AI’s value. Procurement is a revealing test case because so much of its day is structured, repeatable work, such as purchase order processing, supplier onboarding, and contract checks, which is exactly the layer AI absorbs first.

Gartner expects demand to split in two directions. Lower-skilled roles augmented by AI on one side, and higher-skilled roles built around creativity and complex problem-solving on the other. The middle, where many buyers and analysts sit today, is where the pressure concentrates.

The Same Gap Shows Up Across the C-Suite

Procurement is not unusual. Grant Thornton’s 2026 AI Impact Survey of 950 business leaders, fielded in February and March, found that only 22% of operations leaders are working with a fully developed and implemented AI strategy. Training is the most underfunded area of AI investment, with 34% of finance leaders saying it is not getting enough. And only 6% of executives identify change leadership and workforce enablement as a top essential skill, even though respondents pointed to frontline employees (37%) and middle managers (30%) as the groups that need the most support to implement AI.

The people on the receiving end feel it. Glean’s Work AI Index 2026, a survey of 6,000 full-time digital workers in the US, UK, and Australia, found that AI automates 27% of workers’ output today and that they expect it to reach 35% within 12 months. More than half, 51%, say AI has already automated meaningful work they would have preferred to keep. That is what role drift looks like from the inside: tasks disappear, and nobody redefines what the job is now.

Managers make a measurable difference. Among workers with good managers, 62% have had a recent AI reskilling conversation, compared with 30% of workers with other managers, and 57% say their organization rewards AI skills, against 25%. Glean also found that 77% of AI users juggle multiple tools each week and 33% use four or more, a “toggle tax” that grows when work is automated piece by piece rather than redesigned as a whole.

What Gartner Tells CPOs to Do Next

Gartner’s recommendations are practical. Audit roles to decide which tasks stay with people and which move to AI. Assess both technical skills and enduring human skills such as creativity, influence, agility, and ethical judgment, and target development accordingly. And create a “workforce sandbox” where employees can test AI tools, where failed experiments count as learning, and where leaders are transparent about what AI is meant to achieve. Mehrzai added that role redesign is not a one-time exercise and needs continuous adjustment as AI advances and employees gain practical experience.

The Role Redesign Playbook

• Start with a task inventory, not an org chart: List what each role actually does week to week, then mark which tasks AI already handles, which it will handle soon, and which need human judgment.
• Protect the middle on purpose: Middle-skilled roles face the most pressure. Decide now whether each one moves toward AI-augmented execution or toward higher-skill analysis, and tell the people in them.
• Rewrite job descriptions and career paths together: A redesigned role without a visible next step leaves employees exactly where Gartner warns: without a clear path.
• Fund training as part of the AI budget: If training is the most underfunded AI line item, the rollout is underfunded. Tie enablement spending to each deployment rather than treating it as optional.
• Equip managers to have the reskilling conversation: Workers with good managers are twice as likely to have discussed AI reskilling. Make that conversation an expectation for every people leader.
• Redesign the workflow, not just the task: Automating steps one tool at a time multiplies handoffs. Map the end-to-end process before deciding where AI fits so the remaining human work holds together.

The procurement numbers make the risk easy to see. Leaders have priced AI’s benefits into their three-year plans, but three in four have not changed the jobs that have to produce them. The functions that close that gap first will be the ones that treat role design as part of the AI program itself, not as cleanup after it.