For more than a century, business schools have adapted to the needs of the workplace. We have taught future accountants to reconcile ledgers, marketers to craft campaigns and managers to lead teams. Yet over the past decade, undergraduate business education has shifted toward producing tangible artifacts: spreadsheets, dashboards, reports, even lines of code. These practical exercises have been valuable in that they give students confidence and employers a portfolio to review.
But a quiet revolution has been unfolding. Artificial intelligence, in just the past two years, has become capable of producing many of these same artifacts on demand. What once required hours of student work – building a financial model, drafting ad copy or writing database queries – can now be generated in seconds by a well-crafted prompt.
This development changes the calculus of what business schools should teach. The core question is no longer how to make the spreadsheet, but why one model is better than another, what its assumptions are and where it might fail. In other words, theory and judgment, long considered the “softer” side of business education, are suddenly the most valuable skills we can impart.
The End of Entry-Level as We Know It
Recent research from Stanford shows that the earliest signs of AI’s labor market impact are appearing among the youngest white-collar workers. Entry-level roles in accounting, software and customer support, which were jobs that traditionally served as training grounds, are thinning out. Fewer openings exist, not because firms no longer need ideas or oversight, but because AI can handle the procedural, routine parts of the job (Brynjolfsson, Li & Raymond, 2025).
Goldman Sachs predicts that while AI may cause short-term disruption, it will ultimately be woven into nearly every business role (Goldman Sachs Research, 2023). Employers are already signaling this shift: More than half of U.S. job postings that mention AI skills are outside the tech sector (Lightcast, 2024). In other words, AI is no longer a niche tool, instead it is becoming the default.
For business schools, this means our graduates cannot differentiate themselves by producing first drafts of memos or marketing plans. Machines can do that now. What remains uniquely human is the ability to question, persuade and decide.
Why Learning Theory and Principles in Business Education is Not Optional
Learning scientists have long known that conceptual knowledge supports adaptability. A student who understands why accrual accounting matters, or why the CAPM (capital asset pricing model) assumes frictionless markets, can apply those principles even as tools evolve. By contrast, a student trained only to follow step-by-step procedures will be lost when those steps are automated (National Academies of Sciences, Engineering, and Medicine, 2018).
AI’s brilliance is also its vulnerability. It generates polished outputs, but it does not “know” whether those outputs are appropriate, ethical or even correct. A finance model with spurious assumptions, a marketing campaign tone-deaf to culture, an accounting entry that violates standards are all within AI’s range. Only a human with strong theoretical grounding can catch such errors, explain why they matter and steer the organization accordingly.
So the need of the times is that the smarter the tools become, the more we must lean into the intellectual roots of each business discipline.
Reimagining the Teaching of the Business Disciplines
Below are some examples to help illustrate how business education may need to change.
Accounting: Instead of drills in debits and credits, we should emphasize recognition principles, materiality and internal control frameworks. Students can use AI to generate journal entries and then learn to evaluate them against GAAP or IFRS, identifying where the system misclassifies or oversimplifies.
Finance: Spreadsheets can already calculate CAPM, Fama-French factors or discounted cash flows. The real lesson is in debating the assumptions: When is CAPM sufficient and when do we need a multifactor model? What risks are invisible in the dataset? Students should use AI to run the numbers, and then defend their theoretical choice.
Marketing: AI can churn out dozens of ad variations in minutes. But should the campaign focus on segmentation or positioning? What does consumer behavior research tell us about framing? Here, the learning lies in evaluating strategies, not in writing copy.
Operations and Supply Chain: Optimization software, often AI-assisted, already solves linear programs or simulates queues. What matters is teaching students to interpret variability, to understand when “optimal” solutions collapse under real-world uncertainty and to weigh robustness against efficiency.
Information Systems: Coding remains essential. Every graduate should be able to read and write code, if only to understand how information systems are constructed. But the goal is not to grade students on whether they can build a working artifact line by line. Instead, we must teach them the principles of coding and system design: how to structure programs logically, how to document their work, how to maintain and refactor code and how to safeguard data integrity. Coding will remain important, but the more enduring value lies in database design, systems design and governance principles. Students should experiment with AI-generated information systems, and then critique their adequacy for integrity, security and alignment with business goals.
Management: Perhaps most of all, AI underscores the importance of human interaction. Negotiation, persuasion and cross-cultural management are not “nice to haves.” They are the skills that determine whether organizations act wisely on AI outputs. Research shows that cultural intelligence predicts workplace effectiveness (Schlaegel, Richter, & Taras, 2021) and that negotiation training measurably improves outcomes (Deming, 2017). These capabilities should no longer be electives; they belong at the core of the curriculum.
Human Skills in an AI World
Two areas, in my view, deserve special emphasis.
Persuasive narrative: Business is built on argument and story: convincing a board to invest, a community to support or a customer to believe. As AI generates more of the raw material, the human differentiator will be how well we frame decisions, explain trade-offs and inspire trust. Courses in persuasive business communication are not just about PowerPoint polish; they are about cultivating clarity, logic and integrity.
Working across cultures: As our graduates join global teams, they will encounter human colleagues who think, communicate and decide differently. AI tools may translate language, but they cannot translate values or norms. Cultural intelligence, which is the ability to adapt and collaborate across boundaries has been shown to predict everything from job satisfaction to performance (Schlaegel et al., 2021). We owe it to our students to make this a centerpiece of their education.
What This Means for the Future
It is tempting to see AI as a threat to business education. But I believe it is a call to return to our roots. When I began my career in academia more than two decades ago, we emphasized theory because the tools were expensive and slow. In recent years, we shifted toward hands-on projects because employers wanted “job-ready” graduates. Now, with AI automating the entry-level tasks, the pendulum swings back; theory and judgment are not luxuries, instead they are survival skills.
Our graduates will enter workplaces where AI produces the first draft, the first forecast, the first option set. Their value will lie in shaping those outputs into wise decisions. They must be able to say not only, “Here is what the model produced,” but also, “Here is why this model is appropriate, and here are its blind spots.”
That is the vision for business schools in the age of AI: not less practical, but more profoundly human. We will still ask students to work with data, to write reports, to model scenarios. But the assessment will focus on their ability to reason, persuade and decide.
And perhaps, in an ironic twist, AI will free us to do what a university education was always meant to do: cultivate judgment, nurture curiosity and prepare leaders not just for their first job, but for a lifetime of wise decision-making.
References
Brynjolfsson, E., Li, D., & Raymond, L. R. (2025). Canaries in the coal mine? Six facts about the recent employment effects of AI. Stanford Digital Economy Lab.
Deming, D. J. (2017). The growing importance of social skills in the labor market. The Quarterly Journal of Economics, 132(4), 1593–1640.
Goldman Sachs Research. (2023). The potentially large effects of artificial intelligence on economic growth. Goldman Sachs Global Investment Research.
Lightcast. (2024). Global AI skills outlook. Lightcast Labor Market Analytics.
National Academies of Sciences, Engineering, and Medicine. (2018). How people learn II: Learners, contexts, and cultures. Washington, DC: National Academies Press.
Schlaegel, C., Richter, N. F., & Taras, V. (2021). Cultural intelligence and work-related outcomes: A meta-analysis. Journal of World Business, 56(2), 101–209.