Generative artificial intelligence (Gen AI) is changing how business education is delivered and how students engage in the learning process. Drawing on a review of 14 AI-enabled learning tools, we examine where these tools add value, where they risk displacing the effort that learning requires, and what this means for business schools, faculty, and students. Our central argument is that AI is most useful when it preserves or creates the productive friction on which durable learning depends. Our findings are captured in the report, From Chalkboard to Chatbot: A review of AI-Enabled Learning Tools for Business Schools.
Since ChatGPT launched in November 2022, business schools have faced a different kind of technological disruption. Earlier innovations – laptops, online learning, mobile devices – changed how business education was delivered without fundamentally changing how students learned. Gen AI, however, can intervene across the learning process, produce answers rather than simply provide access to information, and do so almost instantaneously.
Students have adopted it at remarkable speed. A 2026 U.K. study found 95 per cent of students using Gen AI in at least one way, and business students are especially likely to use it to produce reports, presentations, and other assessed work. Many also use it to bypass parts of the learning process – for example, by summarizing cases before class or drafting graded assignments.
Faculty are moving quickly as well. Some teach about AI; others teach with it, developing chatbots, simulations, and coaching tools. At the same time, a growing ecosystem of companies is bringing AI-enabled learning tools to market.
Threat and opportunity
Gen AI can enhance productivity and understanding when it supports students’ own efforts. The danger arises when students delegate that effort to AI. The immediate result may be a better output (e.g., a well-written report), but the reasoning, practice, or judgment that produces lasting learning is often lost. Researchers describe this tendency as metacognitive laziness: short-term efficiency gained at the expense of long-term learning.
Durable learning requires students to grapple with unfamiliar concepts, practice skills repeatedly, and exercise judgment under ambiguity or uncertainty. These activities create productive friction between what the task demands and what the student can already do. Gen AI has made that friction optional by allowing students to delegate the effort. As educators, it is our job to preserve friction where learning depends on it – for example, using AI to enable new forms of practice such as repeated role-plays, which were previously unaffordable – and to use AI's data trails to examine the process of learning, not only its product.
What we found
In summer 2026, under our supervision, a team of Ivey MBA graduates reviewed 14 AI-enabled learning tools. Seven were publicly available tools from edtech companies; the other seven were pilot projects created by Ivey faculty and staff. The tools ranged from conversational tutors and AI coaching systems to simulations and interactive cases in which students had to question AI-generated personas, decide what information to pursue, and make a recommendation.
The team assessed each tool along two dimensions: where it was used in the learning journey (preparation, in-class, reinforcement, assessment) and what it was intended to develop (conceptual knowledge, skills, or judgment). We saw four recurring patterns.
· The strongest tools have a specific purpose. Tools designed to cover the entire learning journey rarely perform every part well. The most useful ones target a clearly defined learning activity.
· Tools demand different levels of effort and cognitive engagement from students. Some deliberately withhold answers and require students to reason; others apply challenges more selectively.
· Instructors need to see how students are thinking. Tools that reveal students’ reasoning give instructors insights into the process of learning, not just the final deliverable.
· A specialized tool must offer something that generic AI cannot. If a student can reproduce the same experience by pasting a case or assignment into ChatGPT, the tool adds little. Interactive cases without a static document and specialized simulations are more difficult to replicate.
Where this is heading
How will these AI-enabled learning tools evolve in the years ahead? We expect these tools to improve considerably, while still facing limits on what they can do. We make two predictions.
First, no single “killer app” is likely to dominate. Barriers to entry are low, network effects are weak, and these tools are most valuable when they complement in-person learning rather than attempt to replace it. We therefore expect specialized tools to proliferate, even as more generic AI functions are absorbed into learning management systems like Canvas and Moodle.
Second, AI tools will not displace in-person teaching. None of the tools we reviewed replicated a well-run discussion, where students must articulate their reasoning, respond to disagreement, and learn from one another in real time. These are precisely the elements that are difficult to delegate to AI. As AI makes individual outputs easier to produce, the case method may therefore become more, rather than less, valuable.
Practical advice
So how is the advent of AI-enabled learning going to affect you? Here is our advice.
For business schools. Understanding how AI can be used to enhance student learning is important. We distinguish between using AI as a traffic cop to sequence learning activities and as a sparring partner to challenge students’ thinking. A well-designed course likely makes use of both approaches.
Be willing to declare some settings AI-free – particularly assessment and discussions where social skills are the point. Above all, schools should ensure that individual experiments accumulate into institutional learning. Agree on common measures in advance so that tools can be compared substantively, and record what was tried, with whom, and what happened –including the failures.
For faculty. Start by identifying the specific learning activity in your course where students may delegate effort to AI. Then be clear about the intervention – are you restoring productive friction that AI has removed (closed-book assessments, device-free discussion), or using AI to create new forms of productive friction (e.g., an interactive simulation that requires students to determine what information they need)? Finally, decide what evidence would tell you the AI tool is working toward durable learning. Adoption and student satisfaction are weak signals – a tool students love may be one that enables shortcuts. A better indicator is whether students are engaging with the friction or finding ways around it, something that data trails from most modern tools allow you to see.
For students. Experiment with the tools you're given and build your own where useful. But take responsibility for your own learning. Effective learning involves struggle and effortful engagement. Delegating that effort to AI is a false economy. Use AI to extend your effort, not replace it. Here is the rule of thumb: learning works when you answer the AI's questions, not when the AI answers yours.
Finally – keep some of your learning deliberately AI-free: show up in person, meet colleagues face-to-face, attend the seminars. Learning when to set AI aside is part of learning how to use it well.
Conclusion
Business education is in flux. Assessment has had to adapt quickly; case teaching, less so. The imperative across the board is to understand what Gen AI makes possible, run disciplined experiments, and learn systematically from them. These are challenging times – but for schools willing to engage seriously with productive friction, they are also unusually exciting ones.
For a deeper look at how generative AI is reshaping business education and what it means for the future of learning, see the white paper by Julian Birkinshaw, Mazi Raz and Markus Walters, From Chalkboard to Chatbot: AI-Enabled Learning in Business Schools.
