As organizations race to deploy artificial intelligence (AI), many are discovering an uncomfortable truth: Technology alone does not create transformation. In fact, when AI is layered onto ineffective or poorly understood processes, it often accelerates the wrong outcomes. As the saying goes, if you take AI and put it onto a bad process, you don’t fix the problem. You simply make it more efficiently bad.

This reality should force a reevaluation of how organizations prepare managers for the next phase of managers’ work. Traditional manager training has emphasized coaching, communication and performance management. While these skills remain essential, they are no longer sufficient. Over the next few years, the managers who add the most value will be those who can translate AI potential into operational reality. That requires focusing on three new capabilities in manager training: process expertise, systems thinking and accelerated change management.

1. Process Expertise: Understanding Work at a Granular Level

AI does not replace the need for managerial judgment; it actually increases the demand for it. To determine where AI can be deployed effectively, managers must deeply understand the processes they own. Process expertise goes beyond knowing work outcomes or key performance indicators (KPIs). It requires the ability to break work down into its component tasks, understand handoffs between roles and team members, identify bottlenecks in a process and recognize where variation or judgment truly matters in the process. This kind of task-level analysis becomes the foundation for deciding what work can be automated using AI, what should remain human-driven and what parts of a process need to be redesigned altogether.

Managers must be able to ask critical questions of a process:

• Why does this process exist?
• What value does each step create?
• Which activities are essential, and which are legacy artifacts?
• Where are people spending time on work that could be simplified, eliminated or augmented?

Without this clarity, organizations risk automating inefficiency. With it, managers and leaders can identify opportunities for AI to reduce friction, improve quality or free up capacity for higher-value work.

Training managers in process thinking, which includes workflow mapping, task decomposition and outcome-based analysis, is no longer optional. It is a prerequisite for successful AI deployment.

2. Systems Thinking: Seeing the Organization as an Interconnected Whole

Even the best-designed AI initiatives fail when they are implemented in isolation. The deployment of AI that improves efficiency in one function may create downstream issues elsewhere, shifting work rather than eliminating it. This is why systems thinking is becoming a critical managerial capability.

Systems thinking enables managers to see how processes, technologies, incentives and behaviors interact across the organization. It helps them anticipate second- and third-order effects of AI implementation: how changes in one area ripple across teams, customers and decision-making structures.

Managers trained in systems thinking are better equipped to:

• Identify dependencies across functions.
• Avoid local optimization that harms overall enterprise performance.
• Align AI initiatives with broader organizational objectives.
• Contribute to a cohesive vision for AI rather than a patchwork of disconnected pilots.

As organizations experiment with AI, many fall into the trap of launching isolated use cases that never scale. Managers with strong systems thinking capabilities can help prevent this by ensuring that AI investments reinforce, rather than fragment, how value is created.

This is especially important as companies revisit fundamental questions about differentiation: What is truly unique about how the organization creates value? Does AI change the basis of competition? Does it enable new levels of personalization, speed, or insight? Managers who can connect these strategic questions to operational reality become essential translators between vision and execution.

3. Managing Through Sustained Ambiguity: Leading Teams Through Continuous Transformation

Even the most elegant AI solutions fail without adoption. Research shows that successful outcomes depend most on an organization’s ability to secure use. Organizational readiness and user acceptance are repeatedly identified as central determinants of whether AI delivers value. Managers sit at the center of this challenge. They are the ones who must explain why change is happening, how work and processes will change, and what it means for individuals’ roles.

AI introduces an ongoing cycle of adjustment for employees: processes will evolve, tasks will shift, skills will need to be refreshed continuously. Managers must be equipped to lead through this sustained ambiguity.

Effective management through sustained ambiguity in an AI context includes:

• Communicating clearly about what is changing and what is not.
• Addressing employee concerns about job impact and continued skill relevance.
• Reinforcing continuous learning and experimentation.
• Helping teams adapt workflows in real time as AI tools improve.

Managers must also model adaptability themselves. Teams take cues from how leaders respond to uncertainty. Research shows that adoption accelerates when managers demonstrate curiosity, openness, and confidence in navigating change. When they resist or disengage, progress stalls.

Training Managers to Maximize the Value of AI

AI will not replace most managers, but it will expose the differences between those who understand how work really happens at the task and process levels and those who do not.

Organizations that want to maximize the value of AI must rethink manager development now. This means moving beyond generic leadership training and investing more in three capabilities that enable managers to redesign work, think systemically and guide teams through sustained change.

How Manager Training Must Change to Build These Capabilities

To prepare managers to maximize the value of AI, organizations need to shift manager training so it builds operational capability.

• Train managers to map and diagnose their own workflows. Build process expertise by having managers document end-to-end workflows for their teams, including task sequences, handoffs, decision points and exceptions. Workshop-style training should emphasize identifying where work adds value versus where it exists by habit, legacy or compliance inertia.

• Teach task-level analysis, not just outcome management. Managers should learn in whiteboarding workshop training sessions to break roles into discrete tasks and categorize them into one of five categories: eliminate, simplify, automate, augment or retain as human-led. This creates a practical foundation for deciding where AI can, or should not, be deployed.
• Use real operational use cases instead of generic AI examples. Replace abstract AI demonstrations with exercises based on the organization’s actual processes. Have managers evaluate where AI could reduce friction, improve quality or increase speed in their own teams, while also assessing the downstream organizational impacts.
• Build systems thinking through cross-functional scenario work. Incorporate training exercises that show how changes in one function affect others. Managers should work through scenarios where AI improves efficiency locally but creates unintended consequences elsewhere so they develop the ability to anticipate ripple effects before implementation.
• Teach managers to connect AI initiatives to enterprise value creation. Training should help managers revisit fundamental questions: Why does this process exist? What differentiates us? Does AI change how we compete or personalize offerings? This training approach ensures AI efforts align with strategy rather than becoming isolated pilots.
• Equip managers with change leadership tools they can use immediately. Provide concrete guidance on how to communicate change, address fear and resistance, and explain how roles evolve over time. Role-playing conversations about job impact and workflow shifts should be standard, not optional.
• Normalize experimentation and iteration as part of management. Managers should be trained to run small experiments, gather feedback and adjust workflows continuously. This helps teams view AI as a continuous learning journey.
• Coach managers to model curiosity and confidence during uncertainty. Training should reinforce that how managers behave matters as much as what they say. Managers who demonstrate curiosity, openness and calm confidence create psychological safety, which directly accelerates adoption.

The managers who develop process expertise, systems thinking and sustained change leadership skills will become force multipliers for AI in their organizations. They will help their organizations avoid expensive missteps, scale what works, and ensure that technology investments translate into real performance gains.

In the coming years, the question will not be whether organizations adopt AI. That ship has sailed. The question will be whether their managers are trained in ways that enable them to make that adoption matter.