Implications for pedagogy , assessment, and policy—plus a r o admap for action in Indonesia and beyond
Patih M Pratama
Executive Summary
-
GenAI compresses cognitive load (e.g., case-prep summaries), improving efficiency but risking shallow learning. Leading schools now permit AI for preparation but restrict it in the room, while redesigning pedagogy to keep the “struggle” that builds judgment
-
Across sectors, performance signals are shifting from recall and first-draft writing toward creativity, moral reasoning, and structured thinking—domains where current AI remains a tool, not a substitute. OECD/UNESCO guidance urges human-centred, teacher-led integration.
-
Integrity tech is surging (online proctoring market CAGR ~16–26%), yet AI-detection is unreliable, prompting institutions to pivot to assessment redesign over pure policing.
-
Indonesia is moving: public consultation on a national AI roadmap and plans to embed AI across schooling and highereducation—an opening to set assessment and ethics standards nationally.
Introduction
Generative AI (GenAI) has emerged as a transformative force in nearly every sector, but its implications for education are especially profound. From business schools to elementary classrooms, the way we teach, learn, and assess is being reshaped. What once seemed like the stable cornerstones of pedagogy—memorization, writing, case study analysis—are being disrupted by tools that can instantly summarize, draft, and even simulate reasoning.
This whitepaper explores how GenAI is altering educational practices, with three focal points. First, we examine the specific case of business schools and their historic reliance on the case study method. Second, we broaden the lens to look at how the entire education sector is adapting, from assessment integrity to the commoditization of knowledge. Finally, we propose solutions—spanning pedagogy, assessment redesign, and policy—that can prepare humanity for a future where AI is not a competitor but a collaborator.
Part I: Business Schools and the Case Study Dilemma
For over a century, business schools have relied on the case study method to train leaders. By immersing students in complex scenarios, the case method forces them to digest large volumes of information and then place themselves in the shoes of decision-makers. The value lies in the struggle: wrestling with ambiguity, parsing incomplete data, and developing judgment under uncertainty.
Generative AI changes the equation. A student faced with a thirty-page case on corporate strategy can now prompt an AI system to produce a crisp, two-paragraph summary or even a draft solution. Studies have shown that large language models can perform at the level of a “B” student on MBA-style exams, including operations management and strategy. This means that AI is not merely a note-taking aid; it can already short-circuit the intellectual labor once required to master the material.
Some schools have responded with restrictions. Harvard Business School, for example, allows the use of AI for case preparation but prohibits it during examinations. Instructors are granted discretion to set classroom policies. This approach acknowledges both the inevitability of AI and the risks of eroding the intellectual rigor of the case method.
Yet bans and restrictions only go so far. The more pressing challenge is redesigning the pedagogy to preserve its value. Innovative instructors are experimenting with dynamic cases, where new information is released mid-discussion, making static summaries insufficient. Others are requiring students to submit AI logs—records of prompts and outputs—and then defend their reasoning orally. By shifting the focus from the final memo to the process of thinking, schools can ensure that students still build the judgment muscles required for leadership.
The message is clear: the case method is not obsolete, but it must evolve to remain effective in an age of instant synthesis.
Part II: The Transformation of Education at Large
The shifts visible in business schools mirror a broader transformation across all levels of education. For centuries, formal education has measured performance by testing recall, written expression, and structured reasoning. But GenAI compresses these tasks. A student can now generate a well-structured essay or solution set in seconds, raising fundamental questions about what it means to “know” and how learning should be measured.
Competence is moving up the stack. Organizations such as UNESCO and the OECD have argued that future education systems must pivot from testing knowledge recall to cultivating creativity, moral reasoning, and structured thinking. These are domains where AI remains a tool but not a replacement for human capability. The emphasis is shifting from what students know to how students think and act in complex, value-laden contexts.
Integrity is under strain. The rise of online testing has coincided with AI’s emergence, creating a perfect storm for academic dishonesty. Remote proctoring technologies are proliferating, with global market estimates ranging from USD 836 million in 2024 to over USD 2 billion by 2030. Yet AI-detection software remains unreliable, producing both false positives and false negatives. Universities from the United States to Europe have cautioned against punitive reliance on such detectors. Instead, the trend is toward assessment redesign—developing authentic tasks that AI alone cannot complete credibly.
Personalization is on the rise. Beyond integrity, AI is enabling a shift toward adaptive learning and personalized assessments. Research shows that adaptive testing improves measurement accuracy while also enhancing student engagement. Instead of “one size fits all,” exams can now be tailored to a student’s knowledge path, making learning more relevant and precise.
For Indonesia, these shifts are particularly timely. The Ministry of Education, Culture, Research, and Technology has begun integrating AI into curricula, from pilot programs in schools to consultations on a national AI roadmap. The government’s focuson ethics and responsible AI use suggests an opportunity to build standards for assessment, pedagogy, and integrity from the ground up—positioning Indonesia not just as a consumer of EdTech but as a shaper of global best practice.
Part III: Building an Education System for the AI Era
The purpose of education has always been to prepare humanity for the future. In the GenAI era, that means prioritizing skills that define human distinctiveness: moral reasoning, creative synthesis, structured thinking, and the ability to collaborate effectively with machines. Achieving this requires coordinated change at multiple levels.
Redesigning Pedagogy
Educators should move beyond banning AI to embracing it as part of the learning process. Assignments can require students to document their AI use, critique its outputs, and reconcile them with their own analysis. Oral defenses and in-room micro-assessments can validate authenticity. Dynamic simulations and role-plays can introduce real-time complexity that AI cannot anticipate. Portfolios of work, combining written, oral, and reflective components, can capture a richer picture of student learning than static exams.
Rethinking Assessment and Integrity
Assessment must move from “policing” to “proof.” Instead of relying on unreliable AI detectors, institutions can adopt a tiered approach:
- Closed assessments without AI, to anchor baseline competence.
- Assisted assessments where AI use is allowed but must be documented and critiqued.
- Collaborative assessments where teams use AI transparently, demonstrating how they manage risks and trade-offs.
Remote proctoring will remain relevant for high-stakes exams, but authenticity should increasingly come from task design: assignments rooted in local data, personal reflection, or ethical dilemmas that resist generic AI answers.
Policy and Institutional Levers
At the national level, governments can set standards for AI use in education, fund teacher training, and create regulatory sandboxes for EdTech innovation. At the institutional level, universities and schools can launch “Assessment 2.0” taskforces to redesign evaluations, create AI studios for faculty, and require AI-use statements in all submissions. At the family level, parents can be supported with guides on constructive AI use, helping children balance empowerment with responsibility.
Indonesia, in particular, has a window of opportunity. With an AI roadmap and ethics framework under consultation, the government can establish national standards for assessment integrity and AI literacy. By embedding these priorities early, Indonesia can avoid the pitfalls of reactive policy and instead lead in designing an education system fit for the AI age.
Conclusion
Generative AI is not the end of education but a call to rethink its fundamentals. For business schools, it threatens to hollow out the case method unless adapted. For the broader sector, it challenges the relevance of memorization, writing, and traditional testing. For policymakers, it demands new standards of integrity and equity.
Yet within these challenges lies opportunity. By redesigning pedagogy, rethinking assessment, and reframing learning outcomes, education can prepare humanity not just to live with AI but to thrive alongside it. The future of education will be judged not by how well it resists AI, but by how effectively it integrates AI into the pursuit of creativity, morality, and human wisdom.
Appendix
- Policy: HBS permits AI for prep, bans it in exams; Harvard schools empower course-level policy setting. Poets&Quants+1
- Performance: LLMs can pass parts of MBA exams (B/B-), underscoring the need to grade judgment, not recall. Mack Institute for Innovation Management
- Integrity: Detectors show non-trivial false-positive risk; guidance from universities warns against punitive reliance. teaching.pitt.edu+1
- Market: Remote/online proctoring projected to ~US$2B by 2029–2031 (with higher estimates to 2033–2034), reflecting persistent demand for secure assessment delivery. GlobeNewswire+2The Insight Partners+2
- Indonesia: Active AI roadmap & ethics consultation (Aug 2025) and AI learning in schools/higher-ed initiatives—timelymoment to set national assessment standards. BABL AI+2GovInsider+2
