This assignment, by Kevin Yee, Kirby Whittington, Erin Doggette, and Laurie Uttich, is from the book ChatGPT Assignments to Use in Your Classroom Today.
The summary from the site explains:
Assign students to generate a specific output, then explain whether ChatGPT’s output is factually correct.
Key Features of This Assignment
- Correctness Evaluation
- Students are tasked with generating specific outputs using ChatGPT and then assessing the factual accuracy of the generated content. This task helps in understanding the potential inaccuracies that can arise from relying on language prediction models.
- Understanding Generative AI Limitations
- The assignment educates students on the inherent weaknesses of large language models like ChatGPT, emphasizing the importance of not relying on them unquestioningly. It highlights how these models mimic speech without true understanding and can produce convincingly false information.
- External Corroboration Requirement
- Students are encouraged to provide external sources to corroborate the correct answers, enriching the task and ensuring a thorough evaluation of the information. This step reinforces critical thinking and the necessity of cross-referencing facts.
Find the full version of this resource on the University of Central Florida OER repository.