TBR AI EXCHANGE
Submission Date
Submitter’s Name/Email
Institution/School
Department/Discipline
Activity Purpose (assessment, data collection, classroom management, etc.)
AI Tool(s)
Activity Details
Faculty often spend hours manually converting Word documents, Publisher test banks, and existing assessments into Brightspace (PAWS) quizzes. The process requires extensive formatting, creation of answer keys, assignment of point values, development of feedback and hints, and adherence to D2L markup rules. These repetitive tasks can significantly increase course preparation time and limit the development of additional assessment resources for students.
To address this challenge, I developed and implemented an AI-assisted workflow that combines generative AI tools such as ChatGPT and Microsoft Copilot with the D2L Question Import Tool. The workflow enables faculty to transform existing test banks and assessment documents into clean, properly formatted, upload-ready Brightspace quizzes within minutes.
Using carefully designed prompts, AI generates D2L-compatible markup, answer keys, feedback, hints, point values, difficulty levels, and grading logic for a variety of question types, including Multiple Choice, True/False, Matching, Ordering, and Short Answer questions. This significantly reduces the technical burden associated with quiz creation while preserving faculty control over instructional design and content quality.
I designed, tested, and implemented the workflow and delivered a hands-on faculty development session, QuizSmart with AI: Build PAWS Quizzes in Minutes, during Southwest Tennessee Community College’s Summer Institute 2026. The session demonstrated how faculty could use AI to automate quiz formatting and assessment development while maintaining academic quality and instructional alignment.
Following the session, I distributed supporting materials, tutorials, prompt examples, and implementation guides to faculty across the institution and provided ongoing support for adoption and troubleshooting.
The workflow enables faculty to convert existing assessments into Brightspace quizzes in minutes rather than spending hours manually creating and formatting questions. By automating repetitive formatting tasks, faculty can focus more time on instructional design, assessment quality, and student learning.
The approach also supports the creation of larger question banks, additional practice opportunities, and more diverse assessment resources. Faculty participants were provided with a practical, repeatable workflow that can be immediately applied to current and future courses, creating ongoing gains in productivity and efficiency.
Human oversight remains central to the process. Faculty are instructed to review, verify, and validate all AI-generated content, answer keys, feedback, and grading logic before deployment. This ensures accuracy, alignment with course outcomes, academic integrity, and responsible use of AI in assessment development.
Because the workflow uses readily available AI tools and existing Brightspace infrastructure, it can be adopted across disciplines and institutions. The process provides a scalable and repeatable model for leveraging AI to improve faculty productivity while maintaining instructional quality and ethical AI practices.
Please download the files before using the application: Download the D2L Question Importer Tool files .
Two sample prompts are provided below.
Convert the following quiz questions into the D2L Question Import format. Use the correct markup syntax for multiple-choice, true/false, matching, ordering, short-answer, and image-based questions. Return the output in plain text, ready for import into Brightspace.Based on the attached Testbank from the publisher, rewrite the following questions without changing the format to match the correct answers with the attached Testbank. Write the question number in the same line as the question. Do not make anything bold. Revise the hint right after the “@ ” in every question so that the hint does not tell the correct answer directly. At the end, give me a count of how many questions have answers that did not match the Publisher.Please note that these prompts are only a starting point. Depending on the complexity of your questions, question types, images, formatting requirements, or desired grading options, you may need to refine, expand, or revise the prompt through multiple iterations to achieve the desired output. Effective prompt engineering often involves providing additional instructions, examples, or constraints to guide the AI toward producing import-ready content.
Using AI to generate and validate question markup can significantly reduce the time required to build assessments while maintaining accuracy and consistency. However, I strongly recommend reviewing all generated output before importing it into Brightspace to ensure correctness and alignment with your instructional goals.
If you have questions about the D2L Question Importer Tool, Brightspace quiz imports, regular expressions, or AI-assisted quiz development, please contact me at smmasum@southwest.tn.edu. I would be happy to help.