Share with your CHRO
Most enterprise AI programs are stalling not on the technology side but on the people side, according to Glory Nelson, Country Head of India Delivery Centre at Xebia. Her argument, laid out in a People Matters interview, is that capability gaps, workflow fragmentation, and cultural resistance are consistently outpacing what AI infrastructure can deliver. Xebia’s own data puts a number on one piece: 46% of organizations cite the absence of a data-driven culture as a primary blocker to AI success.
What this means for your business
The uncomfortable read for any CHRO sitting on an AI upskilling budget is that Nelson’s diagnosis makes HR the critical path, not a support function. If your organization is running pilots that aren’t converting to embedded workflows, the deficit almost certainly lives in how work is structured and how confident employees feel operating inside AI-assisted processes, not in which model you bought. Whether this story is about your organization depends on one question: does your AI learning program feed real workflows, or does it feed a completion rate dashboard?
Nelson draws a distinction worth taking seriously, between organizations that are digitally capable and those that are AI-native. A digitally capable workforce uses tools effectively. An AI-native workforce treats AI as the substrate of how decisions get made and value gets created, where employees validate model outputs, apply business context, and operate inside governance guardrails as a matter of routine. The gap between those two states isn’t closed by training courses. It’s closed by redesigning jobs, which is squarely a CHRO problem. Nelson’s framing, shaped by Xebia’s commercial interest in workforce transformation services, tilts toward organizational complexity as the bottleneck, but the underlying pattern holds regardless of who’s selling the fix.
The leading indicator to watch is whether your organization’s AI governance sits inside a single business unit or spans functions with shared standards. Fragmented governance is the structural tell that pilots won’t scale, because each unit ends up optimizing for its own use case with no connective tissue. CHROs who’ve been handed AI upskilling as a project but not AI workflow redesign as a mandate are holding the wrong brief. The budget defense that matters now isn’t the training line item; it’s whether workforce transformation has executive sponsorship equal to the infrastructure spend sitting in the CTO’s budget.
Based on reporting from Why AI pilots struggle to scale inside enterprises: Glory Nelson, originally published 2026-07-23 01:38:00.
