Integrating Large Language Models in Online Course Design

AI and the Learning Design Center at USD

The integration of artificial intelligence (AI) into education has rapidly transformed instructional design, and at the University of San Diego’s Learning Design Center (LDC), we are exploring how Large Language Models (LLMs), such as ChatGPT and Microsoft Copilot, can enhance online learning. Our work in instructional design involves developing effective, engaging, and pedagogically sound courses for online learners. By leveraging AI tools, we have redefined how we create engaging learning materials, assess student learning, and encourage meaningful engagement with these exponentially developing technologies.

Revolutionizing Course Design with LLMs

One of the most significant contributions of AI in our instructional design workflow is its ability to generate structured course content. Using tools like ChatGPT and Copilot, we streamline the creation of course maps, learning outcome alignments, assignments, discussion prompts, presentations, and even branching scenarios. These tools enable us to create a more efficient workflow by reducing the time spent on content development while maintaining instructional integrity.

For instance, when designing course maps, AI assists in organizing learning materials systematically. We first use inputs from subject matter experts to provide the draft design of the module. Then, AI-generated prompts may be used to create possible revisions that align module learning outcomes (MLOs) with course learning outcomes (CLOs), ensuring assessments align with instructional goals. This has proven particularly useful when working with subject matter experts (SMEs) who may need additional guidance in structuring content for online delivery.

Enhancing Assessments with AI

Assessment design is a critical component of instructional design, and AI has introduced innovative ways to craft and refine these evaluations. The SMEs create assignments using Bloom's verbs to specify student outcomes, and we optionally use AI to develop rubrics from those specific assignment criteria that clearly define performance expectations for assignments.

If requested by the ID or SME, AI can enhance discussion prompts by ensuring they provoke critical thinking. Instead of generic questions, AI assists in crafting nuanced prompts that challenge students to analyze, synthesize, and evaluate course material.

By using AI-generated scenarios, we can simulate real-world applications of theoretical concepts, making learning more interactive and meaningful.

AI-Powered Customization of Learning Materials

One of the most exciting aspects of integrating LLMs in instructional design is the personalization of learning materials. AI can tailor content to meet the diverse needs of students, from generating synchronous elements to course designs. For example, the designer can create prompts that are either part of semi-automated CustomGPTs, which have preloaded prompts designed so that, when students click on a link, they are presented with turn-based activities in which they take on a role and respond to situations that require the application of their learning to solve issues as they are presented to them in real-time. For instance, in a course on Cybersecurity, students are challenged to "Find the Insider Threat." Their job is to find an internal hacker in the company by asking pertinent questions and following the clues that the AI provides that lead to the discovery of the perpetrator. Not all questions will solve the mystery, however, so diligence and critical thinking are required.
AI has been instrumental in crafting engaging branching scenarios that challenge students to navigate complex, real-world dilemmas, fostering deeper critical thinking and applied learning. These tools assist in refining dilemmas, creating storyboards, generating images that are personalized to the content instead of relying on generic stock images, and synthesizing AI-generated voices that enhance learner immersion and engagement.

Addressing Ethical Considerations in AI Use

As with any technological advancement, the use of AI in education raises important ethical considerations. At the LDC, we actively engage in discussions about AI’s role in instructional design, ensuring that we use these tools responsibly. Our goal is to design courses where students learn to work with AI effectively while maintaining their critical thinking skills.

One major ethical concern is the potential for bias in AI-produced content. Addressing ethical issues requires ongoing evaluation and a commitment to transparency in AI integration. To reduce these risks, we emphasize AI literacy among instructional designers and faculty through the LDC and USD’s Center for Educational Excellence (CEE).  Understanding how AI models generate responses allows us to critically evaluate their accuracy and relevance before incorporating them into course materials. Additionally, we advocate for transparency in AI use, ensuring students are aware of how these tools are being integrated into their learning experiences.

 

 

 

AI’s Impact on the Future of Instructional Design in LDC

The role of AI in instructional design is evolving rapidly, with trends developing and converging at exponential rates. At the same time, we process our world linearly which makes planning for exponentially changing technologies extremely challenging. On the one hand, Instructional Designers actively engage with the current, practical use of these tools for course design, teaching and learning. In addition, we continue to iterate our philosophy to ensure a balanced and mindful approach to incorporating these tools into the learning process. At the same time, we are tracking these exponential, and at times, converging trends. For example, our proximity to information is shrinking—from the distance of library card catalogs, to desktop computers to smartphones and soon to AI-driven Smart glasses equipped with AR/XR capabilities.

At the same time, personalized learning is being enhanced through just-in-time information delivery, increasingly integrated with Learning Management Systems. The convergence of these accelerating technologies is driving a revolution in personalized, real-time, on-demand educational experiences, including dynamic simulations tailored to individual learners' needs. Due to the exponential nature of these trends, this future is nearer than we think and LDC continues to partner with Canvas/Instructure on their AI enhancements to their LMS.

With tools like ChatGPT and Copilot, we can move beyond traditional course design approaches, embracing more dynamic and flexible methods for online learning. At the LDC, we are actively developing a knowledge base that catalogs our most effective AI integrations into course content and assessments. As this repository expands, it will empower us to rapidly prototype and test diverse instructional strategies, refining courses based on real-time student feedback and engagement patterns.

Furthermore, AI is reshaping the skillset required for instructional designers. While content expertise remains essential, effectively integrating AI into course content to enhance student engagement and learning—without fostering over-reliance or diminishing critical thinking—requires careful balance. At the LDC, we are actively upskilling our team to harness AI’s current and future potential in creating high-quality, interactive learning experiences.

Conclusion

Integrating LLMs into instructional design at the University of San Diego’s Learning Design Center has been a transformative journey. From streamlining course development to enhancing assessments and fostering personalized learning, AI is proving to be an invaluable asset. However, as we continue to leverage these tools, we remain committed to ethical considerations and best practices to ensure AI enhances, rather than replaces, human-centered instructional design.

As AI technologies continue to evolve, so too will our strategies for integrating them into education. By embracing innovation while maintaining integrity, we can create more engaging, effective, and equitable online learning experiences for students.