From Memorization to Experiential Learning: Reconfiguring Classroom Pedagogy in Management Education through Generative AI

Synthesis: Shi, Dai, and Zhang (2026) argue that traditional management education stays centered on memorization and therefore fails to build the decision-making capability that complex business environments demand. They diagnose three interrelated problems: a disconnection between theory instruction and business practice, a mismatch between textbook content and contemporary needs, and a divide between classroom simulations and real-world contexts. Working from a strategic management course as the example, the paper proposes three generative AI mechanisms for reconfiguring pedagogy: a theory-practice dialogue mechanism, an emerging-knowledge tracking mechanism, and a dynamic business-simulation mechanism. All three move instruction away from validating settled conclusions toward learning through experimentation, failure, and unexpected developments, positioning generative AI as infrastructure for redesigning Pedagogies and Teaching Strategies rather than as an efficiency tool.

Key Findings

• Three design problems, not a deficit in student effort. The authors trace the gap to pedagogy: theory-first cases with known outcomes, textbook updates that lag practice, and simulations bounded by rules fixed in advance.
• Student adoption makes the lag visible. Chegg's 2025 Global Student Survey found that more than 67% of university students across the 15 countries surveyed used generative AI to support their learning, with a global average of 80%.
• Theory-practice dialogue. AI decomposes classic theories into core assumptions, key variables, and boundary conditions, then assembles financial statements, brokerage reports, and media coverage into a database approximating real business information.
• Emerging-knowledge tracking. AI compiles new cases of strategic transformation and mines leading journals for recently introduced concepts, so instructors can juxtapose established and emerging explanations of one phenomenon.
• Dynamic business simulation. Given a role that randomly generates challenging events, AI calculates the consequences of a team's decision, then adds cascading effects such as the target firm's R&D team resigning.
• One shared goal: judgment under ambiguity. All three mechanisms redirect instruction from validating known conclusions toward the capacity to exercise judgment when information is incomplete and signals conflict.

Three challenges in traditional management education

The first is that theory instruction is separated from business practice. A distilled framework is introduced, then validated through a case whose outcome is already known, which turns learning into matching. The authors' example is Nokia's decline, usually reduced to the innovator's dilemma, omitting what made the decision hard: feature phones still generated billions of euros in quarterly profits.

The second is that textbook content lags contemporary needs, a structural risk amplified by uneven Prior Knowledge in digital literacy. Frameworks built for stable macroenvironments struggle to explain cross-sector convergence, and students who meet emerging practice through internships and industry news may conclude that course content has fallen behind, after which classroom participation and learning engagement tend to decline.

The third is the divide between classroom simulations and real-world contexts. Simulations resemble board games with rules fixed in advance and portray the firm as a purely rational actor, so implementation proceeds smoothly where real implementation meets competing internal interests and resistance to change.

Three AI-driven mechanisms for reconfiguration

The theory-practice dialogue mechanism runs in two stages. Before class the instructor uses AI long-document analysis to break classic theories into assumptions, variables, and boundary conditions, and web-search tools to gather financial statements, brokerage research, and media coverage for an industry. During class students work through instructor-designed prompt sequences on that database, asking whether Porter's assumptions about supplier bargaining power still hold.

The emerging-knowledge tracking mechanism keeps that material current. AI compiles worldwide cases of corporate strategic transformation, reviews leading journals, and compares recent practice with emerging theory. In class, one group applies textbook theory to a compiled case of a digital platform firm disrupting a traditional industry while another argues from the emerging materials, and the groups debate competing interpretations.

The dynamic business-simulation mechanism changes the environment itself. The instructor supplies parameters such as initial capital, market growth rates, macroeconomic volatility in the target country, and latent labor-management tensions, and assigns AI a role that lets it generate disruptive events. After a team submits an acquisition decision, AI computes the direct consequences, then issues cascading updates that force the team to reassess resource allocation under incomplete information.

From matching answers to exercising judgment

The mechanisms share a change in what AI is for: used as infrastructure, it does more than return answers faster. The emphasis shifts from memorizing established knowledge to engaging with situations that have yet to occur, and the instructor's work moves upstream into designing the prompts and parameters that make a class productively uncertain.

The capability at stake is a different kind of Problem Solving: not risk-free calculation that matches a standard answer, but acting on incomplete and conflicting information. Their concern is Transfer of Learning: decision habits formed in a protected classroom can become a cognitive liability outside it, because simulated environments filter out uncertainty and reinforce a rule-following mindset.

What this means for practice

• Instructors. Build the pre-class database yourself: decompose each theory into its assumptions, key variables, and boundary conditions, and gather real filings and coverage for the industry you teach.
• Instructors. Replace the closed case with an instructor-designed prompt sequence that asks students whether a classic theory's assumptions still hold against current data, ending the case in an open question.
• Instructors. Give the simulation a role that can generate disruptive events, then require teams to revise their plan in class on incomplete information rather than submitting one decision.
• Curriculum designers. Build the tracking mechanism into the course calendar so new cases and concepts re-enter the syllabus between textbook editions.

Limitations

• This is a conceptual proposal: three mechanisms and worked examples from one strategic management course, with no implementation, participants, or outcome measure, so the claimed capability gains are untested.
• The authors name unresolved risks themselves: variation in instructors' AI literacy, concerns about the accuracy of generated content, and possible student dependence on AI assistance.
• The examples come from one subject area and one level, strategic management in higher education, with no trial in another course or institution.

Connected Concepts

• Generative AI
• Experiential Learning
• Situated Learning
• Simulation
• Business Education
• Transfer of Learning

Connected Articles

• Generating a Student-Informed Teaching and Learning Conceptual Framework for GenAI in Business Schools: A Case Study — a student-informed GenAI framework for business schools
• From Experimentation to Integration: Embedding GenAI in Business Higher Education through the Lens of Constructive — constructive alignment for GenAI in business education
• Innovative assessment and grading practices in higher education: A critical exploration for management educators — innovative assessment and grading in management education
• Designing Conversational Agents for Adaptive Instructional Support in Business Simulation Gaming — conversational agents inside a business simulation game
• AI Personas: Can LLMs Replace Fieldwork in Experiential Learning? — AI personas in experiential fieldwork

Citation

Shi, X., Dai, T., & Zhang, G. (2026). From memorization to experiential learning: Reconfiguring classroom pedagogy in management education through generative AI. Journal of Teaching Innovation and Practice, 2(1), 23-32.