University at Buffalo

Knowing When, Why and How to Use AI

AI changes the environment in which engineering managers make decisions. It gives them faster access to analysis, alternatives and recommendations, but that greater capability does not guarantee better judgment. Engineering Management education should therefore prepare professionals not just to use AI, but to decide when AI is appropriate, when it is not, how to design appropriate guardrails for its use and how to avoid using it as a shortcut that bypasses problem framing, root cause analysis or critical thinking.

I often connect this to a lesson from process improvement. Automating a poorly designed process does not necessarily improve it; it may simply allow the organization to produce waste, defects or undesirable outcomes faster. AI presents a similar risk. If the underlying problem is poorly framed, the process is not well understood or the data are weak, greater computational capability can amplify those weaknesses rather than correct them. 

For Engineering Management education, this means shifting attention from tool proficiency alone toward technology-enabled judgment. Working professionals therefore need to assess data and process readiness, validate AI-generated outputs, understand limitations and risks, and more importantly, keep human judgment and accountability at the center of the decision process. The goal is not to prepare professionals to use AI everywhere, but to prepare them to make better decisions about where, why, and how technology should be used.  

Where Theory Meets the Reality of Implementation

Experiential learning is essential because technology decisions rarely occur in the clean environment suggested by an analysis or AI-generated output. Organizations have incomplete information, competing priorities, legacy processes, stakeholder concerns and resource constraints that may not become visible until someone attempts to implement change.

In my work with Engineering Management capstone students, I have repeatedly seen professionals arrive with a solution idea before fully understanding the problematic situation. Oversimplification without deep understanding is risky, and AI can amplify that risk. Professionals need opportunities to test assumptions, plan interventions, implement them, observe what happens and evaluate whether the intended improvement was actually achieved. Implementation becomes a form of validation because it tests whether an AI-supported recommendation still holds when confronted with actual people, processes and operating conditions. Experiential learning provides something theory alone cannot: the opportunity to discover what happens when an idea confronts a real system.

"AI can expand what engineering managers are capable of doing, but better decisions still depend on the systems we design around it and the human judgment we bring to it."

Seeing AI Decisions through a Systems Thinking Lens

AI does not enter an organization in isolation. It enters a system of people, processes, information, technologies, incentives and business objectives. A decision that appears technically sound may create unintended consequences elsewhere in that system.

Systems thinking helps engineering managers look beyond immediate efficiency gains. The relevant question is not simply, “Can AI make this task faster?” but, “How will this change the system, who will be affected, what new capabilities or risks will emerge and does the change advance the organization’s larger purpose?” My background in Industrial and Systems Engineering and Lean Six Sigma strongly shapes this view: meaningful improvement requires understanding relationships among people, processes and technology rather than optimizing one component in isolation. 

Where Technical Literacy Meets Leadership Judgment

I would resist reducing this to a list of technical AI skills. Technical literacy matters, but effective leaders need a broader combination of analytical, managerial and human capabilities. They need strong problem-framing skills, systems thinking and enough data and AI literacy to understand capabilities, limitations, data readiness, model outputs and validation requirements.

They also need economic and organizational judgment. An AI-enabled solution still has to create value, fit the operating environment and be adopted by the people who will use it. That requires communication, change management, stakeholder engagement and the ability to translate between technical and managerial perspectives. Just as important are intellectual humility and learning agility. AI is changing too quickly for education to rely on a fixed set of tools; professionals need the capacity to question assumptions, learn from evidence, adapt and know when additional expertise is needed. 

Creating a Learning Ecosystem for the AI Age

Engineering Management education will increasingly be defined by the development of judgment. Knowledge remains essential, but access to information and analytical capability is expanding rapidly. The differentiating value of education will lie in helping professionals apply knowledge, challenge it, adapt it and integrate new capabilities into uncertain and context-dependent situations.

This evolution cannot focus only on students. Faculty need opportunities to experiment with AI within their disciplines and learn from one another, while universities and industry need stronger feedback loops so that changes in professional practice inform curriculum. The opportunity is to build a learning ecosystem in which students, working professionals, faculty industry partners learn together as technology evolves. Engineering Management is well positioned to contribute because it already operates at the intersection of technology, organizations, people and information. Its role in the age of AI should be to ensure that increased technological capability is always accompanied by better systems, better decisions, and stronger human judgment.