There’s a question I ask whenever I’m walked through a modernization plan, tending to land awkwardly: once this goes live, who’s going to run it, and are they ready?
It isn’t a gotcha. It’s how I was trained. I spent the early part of my career in the Air Force maintaining B-52s, tracking every skill I needed in a little green maintainer’s notebook to reach the next proficiency level. When I cross-trained into education and training management, I saw a bigger picture. The point was never checking training boxes. It was making sure Airmen were qualified, with the right skills, at the right time, to accomplish the mission. That taught me to look past what’s documented to what’s actually operational, and to ask whether the people carrying out a plan are equipped to succeed when it counts.
Across much of the public sector, those questions don’t have a natural owner. They sit in a gray space between workforce planning, learning, talent management, and operational leadership. That gap used to be manageable. With AI, fast-changing skill requirements, and compressed timelines, it isn’t.
Modernization Is Outpacing Readiness
Modernization that once took years is now happening in months, and some of the most sophisticated work anywhere is happening in government. The Department of War’s consolidation model is one example, and the most interesting part of it is the sequencing, not the technology. Instead of solving talent management, workforce analytics, skills intelligence, recruiting, performance, and learning as separate projects, the organization built a common digital foundation and expanded outward. New capabilities get layered onto it rather than requiring their own platforms, governance, and integration work. On the learning side, multiple organizations are moving into a shared ecosystem with common learner records and enterprise reporting, at real pace.
Adoption still takes time, given the integrations, policy considerations, organizational change, and workforce dependencies involved. And implementation often begins before three questions are answered: How will people be prepared to use the technology? Which use cases come first? What mission outcome is the investment meant to produce?
Those are readiness questions, sitting at the intersection of technology, learning, workforce development, and mission execution. Because they don’t belong to any single function, they get addressed after the technology decision rather than before. Modernization has an owner. Readiness often doesn’t. As timelines compress, that matters more.
You Can’t Plan Against a Workforce You Can’t See
The ownership gap usually gets misdiagnosed as a technology problem. When a modernization effort underdelivers, the platform is rarely the cause. It works as designed. What went wrong is that planning rested on assumptions that were never validated: that people knew how to use the tool, understood where it applied to their work, and could turn it into mission outcomes. Readiness is what we’ve confirmed people can do, not what we hope they can do.
Skillsoft’s Workforce Readiness Report found that only 11 percent of individual contributors and 13 percent of managers say their organizations use formal skills assessments. Most rely on manager judgment or self-reported ratings, and fewer than one in four employees receive training before a new tool arrives.
The public-sector version of this is more complicated, for two structural reasons. The first is guidance. Many agencies issue AI policies that set guardrails and leave the operational details to individual organizations: training, governance, use cases, expected outcomes. Flexibility is often necessary, but guidance that defines the boundaries and not the destination leaves employees clear on what they’re allowed to do and far less clear on how AI helps them do their jobs better. That’s one reason so many stay at the experimentation stage, using AI for simple prompts rather than mission work.
The second is capacity. Even agencies with dedicated learning teams are absorbing more change than their structures were built to handle, and leadership transitions, reorganizations, and competing priorities make it hard to sustain a vision long enough for readiness work to take hold.
Claimed Capability and Earned Capability
A claimed skill is when a person reports, completes, or lists on a resume. An earned skill has been demonstrated under real conditions, where the work itself provides the evidence. Most workforce records hold far more of the first than the second, and every readiness plan built on them inherits that limitation.
For technology leaders, this should feel familiar. You wouldn’t trust an asset inventory based only on what people said they’d installed. You’d verify it. Workforce capability deserves the same rigor, assessed against the work being performed rather than inferred from a job title or a training record.
That verification is the heart of readiness. Start with the mission outcome, identify the human and AI capabilities required to reach it, close the gaps that matter most, and confirm that capability improved performance. Most agencies have pieces of this, and strong programs absolutely exist. Many operate locally, though, succeeding within a bureau, command, or agency without a framework that survives leadership transitions and shifting priorities. Progress gets repeated rather than compounded.
Three Things Worth Owning
None of this requires a new mandate. It requires deciding that readiness is somebody’s job.
Inventory capability before the next system goes live, not after. Put readiness on the acquisition timeline alongside integration, testing, and deployment, because anything unscheduled loses to whatever is most urgent that week. Assign accountability for workforce capability the same way you assign accountability for the migration itself.
I’d add a fourth underneath all of them. Don’t confuse access with capability. Giving people new technology doesn’t mean they’re prepared to use it well. Capability comes from guided practice, relevant use cases, and measurable application to the mission.
When customers ask where to start, I describe it as building a house. You don’t move the furniture in before the foundation is set. Capability doesn’t appear at go-live.
Agencies aren’t sitting out AI. The mandates are here; the investments are being made, and implementation is underway. The workforce side deserves the same planning, rigor, and executive attention as the systems themselves.
The author, Nita Wade, is Market Solutions Leader, Public Sector, at Skillsoft.