As companies rush to deploy new artificial intelligence tools, one recurring lesson has emerged. Technology alone does not guarantee value. The real bottleneck lies in people, especially in the ranks of middle management. Fortune interviewed Feon Ang, LinkedIn’s Asia Pacific managing director, who argues that investment should prioritize the human side of AI adoption so that middle managers can lead change effectively. This article unpacks what that means for organizations, practical steps for manager enablement, and how boards and executives should refocus resourcing to get tangible returns from AI investments.

Why Middle Managers AI Is the Critical Lever

Middle managers sit at the intersection between senior leadership strategy and day-to-day execution. They translate vision into priorities, coach teams, and decide which processes to change. When AI adoption stalls, the problem often surfaces here. Senior leaders may buy and deploy AI tools, but middle managers determine whether those tools embed in daily workflows and deliver productivity gains. If middle managers are undertrained, stretched thin, or incentivized only by short-term metrics, they will deprioritize AI use. Conversely, when managers are empowered, AI adoption can scale much faster across teams.

Practical evidence supports the idea that managers lead in AI uptake. Surveys and workplace studies show managers adopt AI tools at higher rates than individual contributors, often because they face more pressure to improve efficiency and report outcomes. That pattern creates both opportunity and risk. Opportunity arises because managers can amplify AI benefits across many direct reports. Risk arises because uneven manager buy-in will produce pockets of adoption and leave broader organizational change incomplete.

People First: What Manager Enablement Actually Looks Like

Adopting a people first approach means resourcing manager enablement deliberately. Manager enablement is more than a training slide deck. It includes role redesign, realigned incentives, accessible coaching, tactical playbooks, and measurement. Here are practical elements:

Training and hands-on practice. Managers need cohort-based programs that pair learning with immediate application to real team tasks. Short, practical labs work better than long theoretical courses.

Clear use cases and playbooks. Provide managers with concrete scenarios explaining how AI fits into specific workflows, how to evaluate outputs, and when to escalate issues.

Time and role redesign. If managers are expected to adopt AI, they must have time allocated to lead adoption. That might mean temporary allocation of a project budget or reducing other nonessential responsibilities.

Incentives and performance metrics. Reward adoption behaviors and outcomes, not just output volume. Tie promotion and performance discussions to how a manager led team transformation.

Support and communities. Peer forums and coaching help managers learn from the frontline experience of other managers.

These investments turn AI from an IT project into an organizational capability driven by managers who can coach, iterate, and sustain new behaviors.

Common Barriers and How to Overcome Them

Several persistent barriers slow middle managers AI adoption. First, anxiety and lack of confidence. Managers may fear being replaced or feel unequipped to judge AI outputs. Addressing this requires psychological safety and clear role boundaries. Second, time pressure. Managers who are overloaded deprioritize experimentation. Organizations must carve out space for managers to learn. Third, measurement mismatch. If KPIs do not account for transformation activities, managers focus on short-term deliverables and ignore adoption work. Refocusing incentives and measuring adoption progress are essential.

To overcome these barriers leaders can run sponsored pilots that include manager time and accountability, create safe failure spaces where experiments are encouraged, and build measurement frameworks that capture leading indicators such as tool usage, process time saved, and quality improvements.

Case Examples and Early Wins

When organizations succeed, they often follow a similar pattern. A sponsored pilot led by a group of managers shows measurable gains, managers share learnings via internal communities, and HR adjusts promotion paths to reward digital leadership. For example, companies that invested in manager enablement reported faster diffusion of AI tools and better alignment between the tools’ capabilities and real work problems. These kinds of success stories are powerful because managers trust peers more than top-down mandates.

How Executives Should Allocate Investment Differently

Senior leadership often allocates most AI budgets to technology and platform build. The people first argument asks leaders to rebalance spending. Investments should explicitly fund manager enablement programs, coaching, and role redesign as part of any AI project budget. Procurement decisions should include adoption metrics, not only performance benchmarks. Boards and C-level executives must ask how an AI procurement will affect managerial work and what resources managers will get to implement it. Without this lens AI initiatives risk high technical fidelity and low organizational impact.

Measuring Success: Metrics That Matter

Measure both leading and lagging indicators. Leading indicators include manager training completion rates, active usage by managers, and number of playbooks adopted. Lagging indicators include process cycle time reduction, employee satisfaction, and business outcomes linked to AI-driven processes. Importantly, track variance across teams to find adoption gaps and target further enablement where it is most needed. Successful programs iterate on these metrics to refine training and support.

If companies want AI adoption to deliver real returns, they must treat middle managers AI as an investment priority. Technology will not transform organizations by itself. Manager enablement, people first strategies, and realistic measurement frameworks are the levers that convert tools into sustained performance improvement. Leaders who reallocate budget to people, design clear adoption playbooks, and align incentives will find that middle managers become the accelerant that turns pilot projects into scale. The future of AI at work depends less on the next model release and more on whether managers are prepared to lead the change.