J. Drew Stephen
University of Texas at San Antonio
In this assignment, students in a lower-division Music and Film course enrich their understanding of historical practices and gain critical AI literacy by conducting an AI-simulated interview with a film composer who was active during the Golden Age of film in the 1930s and 40s. Students select and research a composer to gain biographical and contextual knowledge, design five interview questions, and engage with a generative AI assistant that has been prompted to assume a historically accurate role of the composer. Students gain critical AI literacy by evaluating the responses for bias and accuracy and by reflecting on the pedagogical and ethical benefits or drawbacks of the learning activity. I taught the assignment in the 2025 spring and summer semesters.
Learning Goals
Original Assignment Context: A lower-division Music and Film course satisfying the Core Curriculum Component Area Requirements in the Creative Arts Domain
Materials Needed: A reliable reference resource such as Grove Music Online, Wikipedia, or a course textbook to conduct initial research; access to a generative AI assistant such as ChatGPT (OpenAI), Claude (Anthropic), Copilot (Microsoft), or Gemini (Google) to conduct the interview
Time Frame: The assignment is intended as one of several inter-related activities assigned over a one-week time frame in a module on films produced in Hollywood’s Golden Age. Students should plan to devote 1-2 hours to researching the composer, conducting the interview, and evaluating and reflecting on the responses.
Overview: A persistent challenge in teaching a large online asynchronous course like Music and Film, which I typically teach with an enrollment of 100 students, is fostering meaningful student engagement in the material. I am particularly interested in having students achieve an awareness of how cultural contexts shaped the practices, experiences, and creative activities of film composers. To address this challenge, I designed an assignment in which students use a generative AI assistant to simulate an interview with a composer active during Hollywood’s Golden Age in the 1930s and 40s. This activity was included in a module on Golden Age film music and intended to reinforce key concepts from the lessons and assigned readings through interactive and reflective engagement.
Students were expected to enhance their learning by engaging with a generative AI assistant, but I also wanted them to have enough prior knowledge to evaluate the responses. To this end, I had them research the composers beforehand and create five interview questions that were evaluated for thoughtfulness and relevance by me and by the AI assistant as part of the process.
The five composers I selected for this activity were chosen based on the module topic—all were active during Hollywood’s Golden Age, and their work appears in the module’s assigned repertoire. I also selected these composers because they are among the best known and most thoroughly researched figures in film music history. This intentional selection increases the likelihood that large language models (LLMs) will generate accurate and contextually rich responses, since these composers are more frequently represented in the training data used to build LLMs. To obtain consistent and successful responses, I designed and provided the prompt for students to use. This had the added benefit of providing students with a model of a successful prompt demonstrating clarity, context, and perspective.
One of the learning objectives for the assignment was for students to gain a richer understanding of film music practices in the 1930s and 40s, but I also wanted students to acquire critical literacy in the agile and ethical use of generative AI. This was achieved by having students reflect on their engagement in the activity by considering the accuracy of the responses, look for bias and especially the tendency of AI to ignore or overlook controversies, and comment on any historical, pedagogical, or ethical concerns about the potential imperfections of AI representations of historical figures. I also included an extra-credit opportunity in which students were asked to design their own AI prompt that could be used to improve or enhance learning in the course.
Students clearly enjoyed the assignment. I used generative AI to evaluate the student reflection responses from the summer 2025 semester to determine that 90% were positive (students expressed enjoyment, found the assignment engaging, and believed it enhanced their understanding), 8% were neutral (students acknowledged the assignment’s value but expressed reservations about AI’s accuracy, biases, or effectiveness) and 2% were negative (students found the assignment unhelpful, inaccurate, or were strongly against using AI for this purpose).1 My evaluation of student reflections also identified the following most frequently mentioned insights:
Many of the students were fully aware of the pedagogical value of the activity and expressed this value in their reflections:
Whereas there is a need in many situations to verify and validate all AI responses, that was less crucial in this assignment. The AI assistant was simulating rather than recreating responses from the historical figures which left a generous amount of space for creative responses and interpretations. In the student responses, and in my own testing of the activity, I did not encounter any critical errors or misinformation. The viewpoints and sentiments may not have been spoken by the actual historical figures, but they could all credibly be attributed to them. What emerged above all was an opportunity for students to connect on what felt like a personal level with a figure from the past to gain a richer understanding of musical practices and attitudes.
Assignment Overview
For this assignment, you will use AI to conduct a simulated interview with a Golden Age film composer. You’ll research your composer, prepare five thoughtful questions, and engage in a historically informed conversation using an AI assistant. Afterward, you’ll submit your questions, the AI’s evaluation of them, and a short reflection on the experience.
Learning Objectives
After completing this assignment, you will:
Assignment Instructions
Once you have selected an AI Assistant, enter the following prompt and begin your interview. Make sure you replace the placeholder text between the square brackets with the name of your composer. You should be asked to submit your questions one at a time. You are welcome to stick with the questions you prepared or modify them if the conversation goes in an interesting direction or if a new question occurs to you based on the responses to the previous questions. You will probably get an evaluation after the fifth question, but you may need to ask for it. If it is not generated automatically, just reenter the final sentence of the prompt. Here is the recommended prompt:
Assume the identity of the renowned film composer [insert the composer you selected here] working in Hollywood in the 1930s and 1940s. You have worked on numerous classic films and have a wealth of knowledge about the music and film industry of that era. I am a university student studying film music, and I have the opportunity to interview you. I will ask five questions, one at a time, about your experiences, techniques, and the impact of your work. Respond to each question as if you were the composer, staying in character and providing historically accurate and contextually relevant answers. After answering all five questions, please evaluate my questions for thoughtfulness and relevance.
Let's begin with my first question:
How to Earn Credit for this Assignment
To earn credit for this assignment you must submit the following: