Published on LinkedIn July 13, 2026
This is the third issue in a series building on Digital Applied’s “Change Management for AI Adoption: A 2026 Playbook,” published June 14, 2026. Their research framework is the foundation. This newsletter adds the practitioner layer: what these patterns actually look like inside organizations, where the standard response falls short, and what to do first.
38% of AI implementation difficulty traces back to human proficiency. Technical issues account for 16%. Those numbers come from Prosci’s study of 1,107 professionals, and they have not received nearly enough attention from the leaders approving AI budgets.
Organizations are spending heavily on the 16% problem. The 38% problem is largely getting a communication cascade and a one-day training session.
The misdiagnosis that keeps repeating
When an AI rollout stalls, the instinct is to look at the tooling. The model is too slow. The integration is brittle. The interface is not intuitive enough. Sometimes that diagnosis is accurate. The Prosci data suggests it is accurate about 16% of the time.
The far more common stall is quieter and harder to see on a project dashboard. People do not know how to use the tool well enough to trust it. They are not prompting effectively. They are not sure when to accept the output and when to question it. They are completing tasks the old way, not because they are resistant, but because the new way does not yet feel faster or safer than what they already know.
Most organizations respond to this with more of what they already tried. Another training session. Another all-hands. Another email from the executive sponsor naming the strategic importance of the initiative. The behaviour does not change because the problem was never a communications problem.
What makes AI different from every other technology rollout
Standard change management is built on a premise that does not hold for AI. The premise is that there is a defined end state. A system goes live, people are trained on it, the project closes, and the change team moves on. That model works reasonably well for an ERP implementation or a new CRM. It does not work for AI.
The tools change monthly. Use cases keep expanding. The capabilities available to employees in December are not the same ones they were trained on in January. One change practitioner described it in a Prosci workshop with a precision that most project plans have not caught up to yet: “AI changes so fast — what are we chasing? It’s a never-ending Phase 2.”
That framing is the central operating reality most AI change programs are not designed for. The rollout is not the destination. It is the starting line. Organizations that treat AI adoption as a project with a go-live date will watch usage spike at launch, drift down once the rollout team disbands, and never fully understand why.
The training gap is not a detail
SurveyMonkey’s 2026 report found that only 13% of U.S. workers received any AI training from their employer. DataCamp reported that the share of organizations offering formal AI upskilling actually fell to 26% in 2026, down from 35% the prior year. Those numbers are moving in the wrong direction while AI investment is moving in the opposite one.
The cost of that gap is visible in the adoption data. Bright Horizons research found that reported adoption jumps to 76% with employer support, compared to 25% without it. That is roughly a 3x lift from training alone. Organizations that are not investing in it are not just leaving capability on the table. They are actively building a workforce that cannot use the tools they are paying for.
The training gap is also a trust gap. People do not trust tools they do not know how to use. When the output looks wrong, and they cannot tell whether it actually is, the safe move is to revert to what they know. Organizations that measure AI adoption only at the deployment level never see this happening. The licenses are provisioned. The tool is live. The workforce is routing around it.
What the practical response actually looks like
The reframe is not subtle. AI adoption is not a project management problem. It is a change-strategy problem and requires a fundamentally different design.
A one-time training event is not training for a moving target. Role-specific enablement, prompt libraries built around the actual tasks each role performs, and ongoing office hours that keep pace with tool changes are what close the proficiency gap. These are not dramatic interventions. They are steady, unglamorous investments that most organizations are not making because they do not fit neatly into a project timeline or a go-live celebration.
The measurement shift matters as much as the training shift. Organizations that track licenses deployed are measuring what was purchased, not what changed. The relevant metrics are active usage rates, depth of use, and whether the behaviours the rollout was designed to produce are actually showing up in how work gets done. Without that baseline, there is no way to know whether the adoption gap is closing or simply invisible.
The “never-ending Phase 2” framing is not a problem to be solved. It is the operating condition to be designed for. Organizations that accept that premise early will build change programs that compound over time. The ones still treating AI adoption as a project will keep reliving the same stalled rollout with slightly different tooling.
Sources
38% human proficiency / 16% technical difficulty: Prosci, “8 Ways AI-Driven Change is Different” (1,107 professionals) Building on: Digital Applied, “Change Management for AI Adoption: A 2026 Playbook,” June 14, 2026 https://www.digitalapplied.com/blog/change-management-ai-adoption-2026-overcoming-resistance-playbook
“Never-ending Phase 2” quote: Change practitioner, Prosci AI Adoption Workshop, North America, 2025 Cited in Digital Applied, June 14, 2026 (above)
13% employer AI training figure: SurveyMonkey, 2026 AI Workforce Survey Cited in Digital Applied, June 14, 2026 (above)
26% organizations offering formal AI upskilling (down from 35%): DataCamp, 2026 Cited in Digital Applied, June 14, 2026 (above)
76% vs 25% adoption with and without support (3x lift): Bright Horizons, 2025-2026 research Cited in Digital Applied, June 14, 2026 (above)
HBR November 2025; organizations fail to capture AI value due to people, processes, politics: Harvard Business Review, November 2025 https://hbr.org (search: AI adoption organizational barriers 2025)
Strategic Signals publishes monthly on strategy systems and execution clarity for senior leaders. Subscribe to receive each issue directly. Subscribe
If this piece resonated, ALIGN: The Leadership Framework for Clarity in Complex Organizations is available on Amazon. Click Here
