Most students are using artificial intelligence (AI) every day. At the same time, many faculty are working to keep up, and colleges are still sorting through what counts as learning in this new environment. At the foundation of all the talk about AI is a fundamental question: If AI can do so much of the work we assign as faculty, what are we actually asking students to learn?
This question led me to reach out to Alex Chan, a business analytics professor at Molloy University and an expert in AI use in classrooms and beyond. For Chan, the central role of higher education is “empowerment.” But he is also clear that empowerment is not something that simply happens because students complete coursework. As he explained to me, using a metaphor, if you are training a boxer, “Getting them to run and hit a punching bag on a daily basis is an important component of the empowerment process. After a certain period, we can check how long it would take for them to run - say 5k - and how powerful their punches are - maybe measured in Newtons. That would be equivalent to measuring the effectiveness of the empowerment process through coursework. Of course, that would not be quite enough.” He added, “If we want to make them a good boxer, building up their confidence in fighting potential real-life opponents is also a critical factor. We should organize matches for them to fight in. How do we do that in the classroom?"
Chan’s question is what many of us in faculty roles are thinking about today. AI can handle a lot of “training work,” such as drafting, summarizing, and even generating ideas. If the goal is merely to complete assignments, students will lean into AI. However, if the goal is empowerment, then our expectations have to shift in new directions.
Chan is already making changes in his courses. He is teaching his students to build with AI. As he relayed to me, “The empowerment comes from the hands-on implementation of a business AI application that has already proven to be working in real life. Students will build up their confidence as they are able to create Minimal Viable Prototypes (MVPs) of elaborate real-life AI applications on their own computers.”
He also wants students to use what they are learning once they leave campus, noting, “All the demo apps that they have built throughout the course are published in open-source forums. Those would form a portfolio in which they could show their future employers as evidence of what they are capable of doing.”
His teaching strategies also align with what employers are telling him. As he stated, employers share that graduates are often “certified geniuses on paper but still can’t perform without months of handholding.” Chan does not blame AI for this disconnect that employers are seeing. He stated, "This has been happening over the past 20 years or so and, therefore, is not just in the wake of AI. The employers are not necessarily looking for fully formed professionals. They are after proactive problem solvers, which are, alas, currently in short supply among university graduates. The narrative from a business school should be that the candidates we are sending are good at helping identify problems, coming up with innovative solutions, and communicating with all the relevant stakeholders to ensure the smooth roll-out of those solutions.”
Sharing his concerns about the future of learning with me, Chan commented, “What I am more concerned about is that most students are not able to take ownership of the deliverables. Students should be able to present, defend, and sell those deliverables to different stakeholders. This is how they add their value.”
Chan also mentioned his classroom expectations. In many ways, the type of learning and rigor that Chan is pointing to is different than what happens in most classrooms. Chan is less interested in whether the work is “technically” correct and more interested in whether students understand a topic, can stand behind it, and can make it meaningful to others. He shared, “If I ask a student to do a presentation on a particular topic, I will not care so much if the research was done with AI and the PPT slides were generated by AI. I would be overjoyed if the student could talk through the materials in the presentation in their own words (showing understanding), defend against any criticisms (showing that they are aware of the shortcomings and know of potential remedies), and pitch it powerfully (showing that they understand the value proposition and how it relates to the relevant stakeholders).”
Overall, what stood out about my conversation with Chan is the realization that AI is forcing higher education to confront what it may have been avoiding for decades. If we truly want to empower students to be engaged and creative members of the nation, we have to be clearer about what that empowerment looks like in practice and be more willing to create learning experiences that foster it.
AI can help students produce work much more quickly, but as faculty, we have to make sure that students know that this is not enough. Chan reminds us that a student might use AI to help produce something, but they still have to understand it, defend it, and make decisions about it.