AI from a Learning Design Perspective

With OpenAI’s conference that occurred this week, the integration of AI into our lives is becoming more convenient and prevalent. What implications are present for learning designers and instructors?

The rise and prevalent use of Generative AI (GenAI) has made waves in all fields including education. As GenAI tools continue to develop and evolve, users are not only becoming more familiar with them but are also creating distinctive use patterns that influence how these tools are applied in both learning and teaching environments. However, with its growth, ethical concerns have surfaced around the authenticity of work produced with AI, as well as questions about dependency, biases in AI responses, and the role of critical thinking in a GenAI-supported educational experiences [4]. Despite these concerns, GenAI still holds enormous potential to personalize learning and foster deeper educational experiences when it is thoughtfully and effectively integrated into learning designs.

How can AI be integrated into a learning experience to support learning?

To leverage GenAI effectively in education, instructors can design learning experiences that go beyond surface-level integration and enable students to use AI as a metacognitive tool. AI can support learning in numerous ways, including providing instant feedback, generating customized content, helping students access a wider range of resources, and scaffolding complex learning tasks. For example, AI can be used to simulate one-on-one tutoring, adapt to a learner's progress and preferences, and support inquiry-based learning by enabling students to ask questions and explore complex topics interactively [1,6].

One of the main advantages of GenAI in learning is its adaptability. Unlike static resources, GenAI tools like ChatGPT can adjust their responses based on each student’s needs and prior knowledge [3]. In a well-structured learning environment, this adaptability allows AI to help students construct knowledge actively rather than passively receiving information.

A case study of preliminary data in a study may shed some light on ways to think about integrating GenAI as part of a course.

Chatgpt vs. Other AI

Before getting into the case study, we must first clarify how GenAI differs from previous AI in education. Prior to ChatGPT's existence, AI has long been used and researched for educational purposes. Historically, AI in education was largely confined to systems like Intelligent Tutoring Systems (ITS) that could assist students in specific, well-defined subjects. ITS are often rule-based and programmed to provide feedback or guidance within narrow parameters, relying on pre-determined content rather than generating new material in real-time [2]. But what makes ChatGPT's existence so revolutionizing is its ability to actually generate content. This is mostly thanks to breakthroughs that created the generative pre-trained transformer (GPT) architecture. This architecture allows the computer to generate humanlike text, making it useful for tasks like answering questions, engaging in conversations, and providing explanations. Such abilities provide more interaction possibilities and thus make it a novel and effective tool to integrate into learning design.

Case Study: GenAI-Supported Assignments for Deep Learning

A learning experience was intentionally designed to incorporate GenAI use into class assignments. Undergraduate students at a public R1 university during a summer education course were asked to use ChatGPT to create an outline of a lesson plan as part of their weekly assignment.

Each student was asked to select a theoretical framework, such as constructivism or social learning theory, before interacting with the AI. Their assignment was to prompt ChatGPT to generate sections of a lesson plan that aligned with their chosen theory, requiring students to actively link lesson components to theoretical concepts. For some students, this was their first experience with ChatGPT, while others were already familiar with its use. Over the course of several weeks, students gradually found ChatGPT to be a valuable aid, as it freed them from the mechanical aspects of the assignment and allowed them to dedicate more attention to making meaningful connections between lesson activities and learning theories [5].

Instead of providing examples for the students, ChatGPT offers an opportunity for students to “create” relevant case studies for themselves to further examine topics in education.

What stood out from this study was how quickly students understood and determined how they wanted to use ChatGPT. One student stated,

"If I were unfamiliar with creating a lesson plan for a specific topic, using ChatGPT as a guide or reference could be beneficial. However, it would be necessary to make adjustments and incorporate important concepts related to social cognitive learning theory…"

This student recognized what ChatGPT was good at and what it was not so good at. This allowed them to see when human intervention is necessary. Similar sentiments were echoed in other student reflections from the course. This brings an interesting point of consideration for the AI literacy conversation. In order to gain AI literacy, we must allow for guided experimentation that lets students use the AI tools. Through continued use, students will iterate upon their understanding of how they can properly use AI to serve their needs. As instructors and learning designers, we need to design tasks that are worthy of serious consideration by students and ensure we voice our expectations for the quality of work submitted.

We need to start thinking about integrating AI into education in a way where there is a synergy between course content, learners, and AI, and merely in parallel.

References

1. Brown, T., Mann, B., Ryder, N., Subbiah, M., Kaplan, J. D., Dhariwal, P., ... & Amodei, D. (2020). Language models are few-shot learners. Advances in neural information processing systems, 33, 1877-1901.

2. Colace, F., De Santo, M., Lombardi, M., Pascale, F., Pietrosanto, A., & Lemma, S. (2018). Chatbot for e-learning: A case of study. International Journal of Mechanical Engineering and Robotics Research, 7(5), 528-533.

3. Feuerriegel, S., Hartmann, J., Janiesch, C., & Zschech, P. (2024). Generative ai. Business & Information Systems Engineering, 66(1), 111-126.

4. Halaweh, M. (2023). ChatGPT in education: Strategies for responsible implementation.

5. Limna, P., Kraiwanit, T., Jangjarat, K., Klayklung, P., & Chocksathaporn, P. (2023). The use of ChatGPT in the digital era: Perspectives on chatbot implementation. Journal of Applied Learning and Teaching, 6(1), 64-74.

6. Xu, W., & Ouyang, F. (2022). The application of AI technologies in STEM education: a systematic review from 2011 to 2021. International Journal of STEM Education, 9(1), 59.