Strategies for Integrating AI Tools into Business Education
Bringing AI to Class Without Letting It Do All the Work
This post is adapted from my recent presentation to faculty and PhD candidates at the University of Texas at Austin McCombs School of Business, "Strategies for Integrating AI into McCombs Classrooms - Part 1", available here: https://utexas.box.com/s/9ys1hy3aqk3kmgb5sypu6uhhb3qavxco
AI in Business Education: A New Frontier
The adoption of AI tools among students has exploded over the last few semesters. A recent survey of my business school students revealed that 70% are regularly using AI tools like ChatGPT and Microsoft Copilot, although few are opting for paid versions of these platforms. This widespread usage aligns with what many of us in academia are experiencing: AI is no longer an optional tool, but a core part of students’ educational workflow.
While an estimated 50-60% of McCombs professors are regularly using AI tools, there’s still a significant portion who are not. This presents both a challenge and an opportunity. If we do not adapt our course delivery and engage with AI tools we risk falling behind both in student learning and in preparing students for a future AI-driven workplace.
My AI journey started after November 30, 2022 when OpenAI released ChatGPT 3.5. I spent my 2022 winter break exploring the potential of this fascinating new tool. It didn’t take long to see how revolutionary AI Large Language Models (LLM) could be. I spent hours testing it and pondering its implications for marketing education. December 22, 2022 I shared my first impressions in an email to all McCombs marketing faculty, outlining what I saw as the transformative potential of AI tools like ChatGPT:
Playing with these technologies over the past weeks it is clear the current GPT 3.5 (ChatGPT) technology will have a tremendous impact on our civilization. The next generation GPT-4 launches in mid-2023 and is anticipated to be 10x better. Our AI future is unfolding rapidly.
We should expect that our students will use ChatGPT and similar AI tech in completing their coursework. As faculty we can review assignments, quizzes, and exams and anticipate how students will use this tech.
ChatGPT still has a long way to go. The AI was instructed using a data set through 2021, so it isn't knowledgeable about current events. The OpenAI Initiative cautions, "ChatGPT sometimes writes plausible-sounding but incorrect or nonsensical answers."
I recently used ChatGPT to generate answers to a 10-question open book, open internet quiz I used in an Executive MBA course this semester. Using only ChatGPT to answer the questions would likely have resulted in a score of 3 or 4 points out of a possible of 10 due to specialized concepts and vocabulary, but all the ChatGPT answers were impressive. I expect my students will use ChatGPT as a starting point from which to further edit and enhance their answers.To me, it was was clear: AI was going to reshape how we teach, how students learn, and how both faculty and students would need to adapt.
In that email, I also highlighted a critical takeaway from those early experiments: while AI could help streamline tasks like writing and summarization, its limitations—particularly in logic and reasoning—meant we needed to be cautious in its application. The past 20 months have only reinforced that perspective. Today, the challenge we face is integrating AI tools into our teaching while maintaining academic rigor and critical thinking.
Improving Course Design with AI
An immediate benefit of AI tools in academia is the ability to enhance course design. AI tools like Perplexity, Microsoft Copilot, ChatGPT, and Claude have proven to be incredibly effective in helping faculty develop more robust and dynamic course materials. As faculty members, we are constantly seeking ways to ensure our courses are up-to-date and relevant to today’s business environment. AI helps us do this more efficiently than ever before.
AI can assist in developing assignments that better reflect real-world business scenarios. By simply feeding the tool a general idea—say, a scenario in which a company needs to forecast market demand using AI—I can generate multiple variations of the same exercise, each with unique angles and complexities that challenge students to think critically. I regularly use AI to help craft new questions and cases, tailoring them to the specific learning objectives of my course.
Beyond assignments, AI can also assist with research. With tools like Perplexity’s real-time web search capabilities, I can gather the industry insights and academic papers, integrating them into the materials for our class discussions.
In the presentation PowerPoint linked at the top of this post, are sample prompts including one I have cultivated for summarizing academic papers. Academic papers are lengthy and frequently dense. In the past, I reviewed an academic paper every week or so. With a consistent detailed AI summary I find in the same time I can review 5-10 papers a week, significantly expanding my learning and the content I can consider for integration into my courses.
Integrating AI Into Course Delivery
AI can also enhance course delivery. My own experience in the classroom has shown that AI can be used to engage students in ways that traditional teaching methods simply can’t match. Whether it’s through personalized feedback, thematic analysis of student responses, or AI-powered tutoring.
I’ve started using AI tools like Claude and Microsoft Copilot to analyze large sets of student submissions. A sample prompt for thematic analysis of student responses in a CSV file is also in the presentation PowerPoint linked at the top of this post. By categorizing and summarizing responses, these tools allow me to quickly identify trends in student understanding and adjust my lectures accordingly. If I notice that a majority of students are struggling with a particular concept, I can tailor the next class session to address those gaps.
Similarly, AI can be a powerful tool for providing feedback. In large classes, giving personalized feedback to each student can be daunting, but AI tools can help. While I don’t recommend AI for grading, AI can certainly assist in providing formative feedback throughout the semester. For instance, students can use AI tools to draft responses, and I can offer feedback on both their use of AI and their critical thinking. This allows students to see the limitations of AI-generated content and encourages them to develop their own voices.
Moreover, AI can enhance classroom discussions by providing real-time analysis of student questions and answers. In one of my marketing electives, I encourage students to use AI tools like ChatGPT during a case discussion. They use AI to explore alternative perspectives or to gather additional data points, which they can then bring into the conversation. This not only makes for richer discussions but also mirrors how they will likely use AI tools in their future careers.
Project-based leverages students' creativity and problem-solving skills, areas where AI is currently less proficient. This approach not only addresses the immediate challenge of students relying excessively on AI, but also enhances the overall learning experience by making it more engaging and relevant to real-world problems. In a recent project for a B2B Marketing elective, students used AI tools to analyze market trends and competitor pricing for a new product launch. While the AI helped them gather information, real learning occurred when they had to verify the accuracy of the AI-generated data and apply it to their specific case.
AI Limitations and Best Practices
While the potential for AI in course design and delivery is vast, it’s important to remain cautious. AI tools are far from perfect. One of the biggest challenges is the inconsistency in their outputs. With AI even seemingly simple tasks, like counting the number of “R”s in the word “strawberry,” can trip up AI models. OpenAI’s newest model, OpenAI o1, is part of new technology code named “strawberry” to signal progress in moving AI tools towards handling basic logic.
Students come to learn that AI tools often “hallucinate,” generating content that seems accurate but is entirely fabricated. This is problematic in academic settings where accuracy and reliability are important. In my own classroom, I’ve had students turn in assignments with AI-generated citations that, upon closer inspection, didn’t exist. To mitigate this, I instruct my students that when it comes to AI, “trust but verify”—always double-check AI-generated data, especially when using it to support research or analysis.
As educators, need to be mindful of the ethical considerations that come with AI use. If we rely too heavily on AI for tasks like grading, we risk dehumanizing the learning process. With the current state of AI tools I don’t recommend grading with AI. AI is unreliable at math and the stochastic element intrinsic to the design of LLMs means that scoring will be inconsistent. Our own grading will be more consistent and helps us stay in tune with students advancement in learning.
For students, the challenge is ensuring they use AI responsibly. AI tools can generate seemingly authoritative citations, but these citations are often fabricated. I teach my students to “trust but verify” any AI-generated content, particularly when it comes to research. By doing so, they learn both the utility and the limitations of these tools.
Looking Ahead: The Future of AI in Education
AI’s role in business education will only continue to grow. However, as educators, we must remain vigilant. While AI can streamline many aspects of teaching and learning, it can also introduce risks that, if unchecked, could undermine the educational process. The key is not to resist AI but to integrate it thoughtfully and ethically into our courses, ensuring that students gain both the technical skills and the critical thinking abilities they need to thrive in a world increasingly shaped by artificial intelligence.
By integrating AI into our classrooms while remaining mindful of its limitations, we can better prepare our students for the challenges of tomorrow. As we continue to experiment with these tools, our goal should always be to enhance, not replace, the learning process.
no