How AI Helped Create Villains in My English Course

Tiffany Morin

Creativity, Invention, Collaboration

Last summer, I wanted to refine the final project in my English course. The course “How to Be Bad: What Makes a Good Villain in Fiction” explores different types of villains in film and literature. I wanted to turn their final paper into a scaffolded inquiry project to ensure that the students engaged with the material we studied. In previous semesters, students had two options for a final paper: write a traditional paper arguing for the most effective villain from our readings and films, or take a creative approach by inventing a villain and having them interact with villains from class. While the creative option allowed for interesting storytelling, I found that it often resulted in a different kind of engagement that didn’t always result in the same level of analysis.

To balance creativity with analysis, I redesigned the assignment so that every student would create their own villain but still engage in comparison with three villains from our syllabus. To help with the invention process, especially for the students who were not creative writers, I introduced AI as a brainstorming tool, providing students with a way to generate and refine ideas before moving on to analysis. They could use any platform they were comfortable with.

Rather than allowing AI to generate villains for them, I required students to write their own prompts to guide AI’s responses. This ensured that they remained in control of the creative process and forced them to think critically about what they wanted in a villain before receiving AI-generated input. Their prompts reflected the qualities they found most compelling in villains and the themes they wanted to explore.

Some prompts focused on appearance and deception:

• “Make the appearance of my villain someone unassuming that could easily blend in.”
• “Describe a villain who is outwardly charismatic but secretly controlling.”

Others explored psychological depth:

• “Give me a villain who believes they are saving the world, but their methods are horrific.”
• “Create a villain whose actions are driven by fear rather than power.”

Some students leaned into the backstory:

• “Make my villain’s past so cryptic that even they don’t fully understand it.”
• “Describe a villain whose origins are unknown, but rumors about them create fear.”

Others wanted to explore moral ambiguity:

• “Make a villain who starts as a hero but loses themselves along the way.”
• “Give me a villain whose ultimate goal is peace, but they go about it in the worst way possible.”

The goal was for AI to enhance their creativity rather than replace it. By using AI as a collaborator rather than a creator, students had the chance to interact with the responses, revising them, rejecting them, or expanding upon them to craft more nuanced characters. 

Once their villains were fully developed, students moved on to the analytical portion of the project, where they compared their villain to three villains from our class. They could write this comparison as a persuasive paper, or for those who chose the creative option, their villains had to interact with the villains from class, engaging in dialogue, confrontation, or alliances. Both options required analysis and textual support.

This approach could be applied in any discipline where students struggle with idea generation. AI didn’t do the analytical work for them. It simply helped get them started, allowing them to focus more on deep thinking and analysis. Students still had to craft arguments, compare themes, and use textual evidence to support their claims. 

By the end of the project, students had developed characters that could take on some of fiction’s most notorious villains. Seeing my students, over the course of the semester, engaged in a process that required them to think critically, refine ideas, and support claims, showed me that, when used thoughtfully, AI can be a great tool that can support learning instead of replacing it.