Welcome to the Teaching and Learning Guide on using Supporting Course Design with AI! 

This guide is designed to help you: 

• Apply best practices when using LLMs for teaching-related tasks

• Write effective AI prompts for course design

• Apply tailoring and revision techniques to improve AI-created course content

Backward design is a useful approach to designing a course. The framework for backward design encourages instructors to first summarize their goals for the course and write measurable course- and module-level objectives. After considering objectives, instructors will then determine what constitutes evidence of met learning objectives in order to create effective learning materials and assessments. 

We’ve included a helpful worksheet in our Designing Effective Courses guide. Yet, designing a course, especially from scratch, can be daunting. Generative AI, like ChatGPT, Claude, or Gemini, can be a useful tool to assist with course design tasks, especially when used thoughtfully. 

Ensuring Objectives are Measurable

Good course design begins with designing effective objectives. Measurable objectives connect directly to assessments and assignments. Effective objectives allow students to demonstrate their mastery of the content. 

Objectives must be measurable, according to Bloom’s Taxonomy, in order to be considered effective. In other words, you should be able to say definitively whether a student has met an objective or not, given the evidence they provided. Objectives can be course-level or module-level. 

Some examples of measurable learning objectives would be: 

• Create a grant proposal for a local organization 

• Explain the impact of the 14th Amendment in the United States

• Communicate mathematical concepts and ideas in writing 

• Identify errors in mathematical problems 

• Defend a persuasive argument in a speech 

An example of a non-measurable learning objective would be: 

• Understand the fundamental principles of art and design 

  • This is not measurable because this objective uses the verb “understand,” and it is not easy to see whether a student understands something. The objective should be specific about how you want to measure student understanding. 

  • You might change “understand” to “identify” if you’re asking students to pick the fundamental principles out of a list (such as in a multiple-choice quiz).

  • You might change “understand” to “describe” if you want students to tell you about the fundamental principles in their own words, like in an assignment such as a discussion board. 

The table and image below provide more information about using Bloom’s Taxonomy to create stems for learning objectives. The table lists objective categories in order from highest to lowest-order thinking.

The above image is licensed under the Creative Commons Attribution 2.0 Generic license. The image was created by Vanderbilt Center for Teaching and has not been modified here. 

Using AI for teaching tasks and material creation can help lighten your workload as an instructor and also help you generate tailored content that is relevant to your students. It’s important to remember that AI should not be used to plan a whole course or to replace the role and expertise of the instructor. Rather, AI can be a helpful assistant to brainstorm and hone your ideas. Considering the high energy needs of AI, it’s also important to consider when using it can be most impactful. 

Generative AI can support you from large-scale course planning to creating course materials. Here are some ideas for types of content you can ask AI to help create: 

• A first draft of measurable objectives

• A course outline and topic ideas

• Graphs and inclusive images using image generators

• Personalized and engaging examples for the entire class, groups of students, or a single student to explain a complex topic

• Activities and games you can assign to students to apply or practice concepts

• Assignments that ask students to create renewable products

• Outlines, worksheets, or lecture note templates

• Slides or other presentation materials for lectures

• Study guides and other study materials, like flashcards 

• Quiz questions that meet particular objectives 

• Alternative quiz questions based on existing questions

One of the most important things to remember when exploring AI is that AI should not replace our critical thinking skills. As an instructor, be sure to bring your own knowledge and expertise of your course subject to your AI use. This will ensure AI is a tool that helps us create the best content for our students.

Prompt Engineering

Prompt engineering is one of the most important skills to learn when it comes to generative AI use. When you open a Large Language Model (LLM) such as ChatGPT, you start by asking the AI for something. For example, you might ask for help developing module-level objectives.

Be sure to always think about privacy concerns before you begin using AI. Never share personal student or university information with AI. Before you share information with AI, reread it carefully to make sure it does not contain private information. 

Here’s a prompt engineering framework you can start with: 

• Persona

• Context

• Objective

• Guardrails

• Be Nice 

Here’s an example prompt: 

I am a university professor teaching an introductory-level course on general biology. I am working on developing a module introducing my students to molecular biology. I care a lot about my students’ ability to apply biology concepts to new contexts. Can you help me identify three measurable learning outcomes that would help guide my teaching of the most important things about molecular biology for my students to know? Thank you for your help!

Let’s break down the prompt using our framework: 

Persona: “I am a university professor”

Context: “[...] teaching an introductory-level course on general biology.” “I am working on developing a module introducing my students to molecular biology. I care a lot about my students’ ability to apply biology concepts to new contexts.”

Objective: “Can you help me identify learning outcomes”

Guardrails: [identify] three measurable [learning outcomes] that can help guide my teaching of the most important things about molecular biology for my students to know?”

Be Nice: “Thank you for your help!”

When plugged into an AI tool, the AI will create three learning objectives plus their application. The first content AI generates may not come out exactly as you envisioned. Most of the time, using AI means honing our language until we get the right output. This is especially true when generating images. When using AI, it is important to rely on your knowledge and critical thinking to review AI content and adapt what we ask AI to do until we get the desired results. This technique is called tailoring.

Tailoring

Our approach to introducing AI to course design practices focuses on expanding, rather than replacing, our thinking. Specifically, we recommend a human-in-the-loop (HITL) approach to using AI. In this context, HITL is a collaborative approach that encourages AI users to interact with the LLM to form a continuous feedback loop. In simpler terms, if you are using AI to help generate ideas, you can talk back to the tool to tailor the content to your specific needs. 

In the example above, the AI will give you three learning objectives about molecular biology. Maybe two of those objectives work well for your course, but the third isn’t quite right. You can ask the AI to create another third objective on a particular concept or topic by giving it new guardrails or context. Having a conversation with AI can help you tailor your content. You would then revise all three objectives to make sure they suit your specific needs and align with your course goals.

Tailoring for Effective Objectives

When asking AI for help with objectives, you might find that you aren’t getting quality suggestions or that they aren’t quite relevant for your needs. In that case, you may need to provide additional details, including: 

• Course context, such as: 

  • Grade level

  • Credit hours

  • Type of course (such as thesis, lecture, introductory)

  • If you are asking for module-level objectives, provide the course objectives for reference

• Output, such as:

  • The number of objectives

  • How they should be formatted. For example, “each objective should start with a measurable action verb.” Or “Please number each objective”

• Details about measurability, for example:

  • Indicate that the AI should use Bloom’s Taxonomy

  • Whether you need objectives to target lower-level or higher-order thinking skills from Bloom’s Taxonomy

  • If relevant, which categories of Bloom’s Taxonomy the objectives should come from

  • Example of a measurable objective

Once you have developed course-level and module-level objectives, you can begin creating a course map that includes instructional materials, learning activities, and assessments that align with your objectives. In other words, your instructional materials, learning activities, and assessments should all work together to support your learning objectives. 

AI can help brainstorm summative assessments, discussion prompts, lectures or presentations, interactive online or in-person activities, and scaffolded assignments. When prompting AI for course materials, be sure to ask for content that aligns with your learning objectives. Other details to share might include: 

• The grade level 

• Time restraints for activities

• Number of assessment questions

• Types of assessment questions

• Instructional materials you shared with students for context

• Context about course timeline

  • For example, you could ask for help with a presentation about a certain topic and include the topics you talked about before and plan to talk about after.

•  Any information about where students often struggle

Here’s an example prompt for a 10-question quiz assessment: 

I’m a university professor teaching a junior-level anatomy and physiology course. We just finished our unit on the skeletal system. Please help me create 10 questions for a quiz about the skeletal system based on the attached PowerPoint lecture. The questions should include a mix of fill-in-the blank, matching, or short-answer questions. I don’t want any multiple-choice questions. This quiz should align with the following module learning objectives. Thank you for your help.

Here’s an example prompt for a 3-part scaffolded assignment: 

I’m a university professor teaching a freshman-level U.S. history course. Please help me create a 3-part scaffolded project. Each of the three parts should have its own guidelines, instructions, and scoring rubric. This assignment will assess students’ understanding of the following topics: the Gilded Age, the American Civil War, the Great Depression, and the Civil Rights Movement. I want students to be able to pick the type of assignment they submit, so please include some options. This assignment should align with the following course learning outcomes. Thank you for your help. 

Each of these prompts will require more tailoring based on the specific needs of your course and your students. The idea with prompting AI is to give you a place to start, rather than to get a fully-formed assignment or quiz directly from the AI tool. 

Be sure to plug any AI-created instructional materials into your course map and ensure that they align with your objectives. 

For more help using AI for course design, contact your Instructional Designer.

• Bowen, Ryan S., (2017). Understanding by Design. Vanderbilt University Center for Teaching. https://cft.vanderbilt.edu/understanding-by-design

• Martin, F. (2011). Instructional Design and the Importance of Instructional Alignment. Community College Journal of Research and Practice, 35(12), 955–972. https://doi.org/10.1080/10668920802466483

• Nilson, L. B., & Goodson, Ludwika A., author. (2018). Online teaching at its best : merging instructional design with teaching and learning research (1st ed). 

• Quality Matters (2023). Bridge to Quality: A QM Online Course Design Guide. https://www.qualitymatters.org/higher-ed-bridge-guide-basic

• Vai, M., & Sosulski, K. (2015). Essentials of Online Course Design: A Standards-Based Guide. New York: Routledge Taylor & Francis.