Carnegie Mellon University

Eberly Center

Teaching Excellence & Educational Innovation

CMU’s Generative AI + Education Modules

As part of CMU’s broader strategy regarding Generative AI and education, the Office of the Vice Provost for Teaching and Learning Innovation is developing a series of instructional modules on generative AI.

The first set of modules is designed as a primer for CMU students – to support them in their role as learners. It targets several fundamental learning objectives related to learners’ effective and responsible use of generative AI tools.

By the end of these modules, learners should be able to…

  1. Describe the basic mechanisms behind how generative AI tools are built and how they work.
  2. Analyze the ethical implications and other concerns related to these tools – in order to be a responsible user and/or creator.
  3. Explain why students’ decisions about and applications of generative AI tools will differ across individuals, contexts, tasks, and goals. 
  4. Identify and apply strategies to appropriately and responsibly use generative AI for a given educational task. 

In addition to developing students’ knowledge and skills in the above areas, these modules are designed to promote students’ self-efficacy for appropriately using generative AI tools.


Studying the modules’ impact in action!

An exciting project associated with these modules has involved a collaboration with the Eberly Center, more than 30 instructors, and more than 2,000 students in which we investigated the modules’ impacts on learning and self-efficacy via a randomized, controlled trial. This large-scale study was the first GAITAR@Scale project, conducted in September, 2024. The data are currently being analyzed (so check back here for future updates!). Results will be used not only to gauge the effectiveness of the v1 modules but also to guide iterative improvements that will be incorporated into v2! 

Timeline of modules design, testing, and dissemination

Spring-Summer 2024 Design team (Learning Engineers, Subject-Matter Experts, and Student Co-Creators) develops v1 modules on Generative AI for Learners and Learning
September 2024 GAITAR@Scale project, studying the modules’ impacts on learning and self-efficacy
Early October 2024 Initial refinements to the modules based on early feedback & results
Late October 2024 Dissemination of modules to the broader CMU community of students and educators
November 2024-?? Ongoing data-informed, iterative improvements to the modules
Spring 2025 Design and development of the next set of generative AI modules in this series

Key Contributors

  • Hoda Heidari, Assistant Professor, MLD, S3D, HCII
    Subject-Matter Expert and module co-author
  • Nicky Agate, Principal Librarian, University Libraries
    Subject-Matter Expert and module co-author
  • Elaine Gombos, Computer Science
    Student contributor and module co-creator
  • Chuong Truong, Philosophy
    Student contributor and module co-creator
  • Harrison Leon, Machine Learning Department
    Student contributor and module co-creator
  • Zach Mineroff, Senior Learning Engineer, Eberly Center
    Lead Learning Engineer coordinating and guiding the modules’ design
  • Avi Chawla, Senior Learning Engineer, Eberly Center
    Learning Engineer assisting in the modules’ design and implementation
  • Judy BrooksDirector of Design, Technology-Enhanced Learning & Online Programs, Eberly Center

Primer Outline

  1. Module 1 Overview
  2. Generative AI Primer
    1. ask students to sign into Copilot and enter prompts, then reflect on usefulness
  3. How LLMs Work
    1. MCQ on LLM mechanisms
  4. The Central Role of Training Data
    1. MCQ on training data sources
  5. Comparison with  Traditional Web Search
    1. categorize features as being associated with generative AI vs trad. search engine
  6. Risks of Hallucinations
    1. Identify generative AI outputs as hallucinations vs not
  7. Attribution for AI-Generated Content
    1. MCQ about attribution
  1. Module 2 Overview
  2. A Framework for Responsible Use of Generative AI
    1. MCQ on technology risk
  3. Who Stands to be Harmed
    1. Given scenarios, select entity most at risk of harm
  4. A Taxonomy of Harms for Generative AI
    1. Given scenarios, select category of harm that best fits the situation
  5. Sources of Generative AI Risks
    1. MCQ on identifying main cause of harmful outcomes in scenarios
  1. Module 3 Overview
  2. Generative AI and the Learning Process
    1. MCQs about general opportunities/risks
  3. Learners Have Different Needs
    1. Describe your own reaction to generative AI, then consider how others might react
  1. Decoding the Syllabus
    1. Identify acceptable actions based on syllabus examples
  2. Appropriate Use: Generative AI as Study Aid or Learning Companion
  3. How to Build an Effective Chatbot Prompt
    1. Prompting a Study Sidekick: Prompt Copilot to generate practice materials
    2. MCQ: identify how to make a prompt better
  4. Responsible Use: Decide, Verify, Cite, Rectify
  5. Responsible Use: Lateral Reading
    1. Use lateral reading to evaluate generative AI output
    2. MCQ on lateral reading strategies