2 High Impact Practices: Writing with Generative AI
Kevin Brown
This chapter shares key lessons from Writing with Generative AI, a Writing Intensive course I developed at the University of Missouri using High Impact Practices (HIPs). Designed for upper-level undergraduates in the arts and humanities, the course focused on both writing about generative AI and writing with it. Students engaged in critical conversations about ethics, authorship, and creativity while using AI tools as active collaborators in their research and writing process.
At the center of the course was a semester-long research paper, modeled on the depth and structure of a thesis project. Students developed their papers in ten stages, each supported by in-class lab work, guided AI prompts, and peer feedback. They used ChatGPT throughout the process to help brainstorm, organize, revise, and reflect. To promote transparency and accountability, students submitted their chat transcripts along with their drafts.
This chapter walks instructors through each stage of the assignment. It includes example prompts, classroom exercises, and editable materials that can be adapted to a range of course goals and disciplines. The structure is designed to be flexible, with each part reinforcing critical thinking, writing skills, and responsible AI use.
By combining hands-on learning with reflective research practices, this course helped students develop stronger writing and a deeper understanding of how to work with AI thoughtfully. The approaches shared here can support instructors who are exploring how to integrate generative AI into their own writing classrooms with clarity, purpose, and impact.
Syllabus
This syllabus can serve as a flexible foundation for instructors looking to integrate generative AI into a writing-intensive course. Rather than offering a rigid model, it is structured to be modular and adaptable, allowing instructors to tailor it to their own discipline, level of technical familiarity, or student population.
The course is built around a balance of skill-building, reflection, and project-based learning. It guides students through foundational concepts, writing practices, and hands-on experimentation with AI tools. The major assignments and class schedule are designed to support sustained writing over time, including space for peer review, revision, and instructor feedback. This structure helps students develop confidence in both their writing and their ability to engage critically with AI technologies.
Instructors can adjust the course calendar to fit shorter semesters, modify the reading list to suit their field, or swap in alternative tools and exercises. The AI integration policy offers a clear but open-ended model that encourages ethical use without restricting experimentation. Embedded in the schedule are checkpoints for research development and revision, which can be used in any course that aims to scaffold student thinking and encourage iteration.
Whether you are teaching a composition course, a seminar in digital humanities, or an interdisciplinary elective, this syllabus provides a working model for incorporating AI in a thoughtful, transparent, and pedagogically sound way.
Research Paper
The research paper assignment in Writing with Generative AI provides a strong model for project-based learning focused on writing and critical engagement with technology. Instructors can use it to guide students through a structured writing process that includes feedback, revision, and hands-on collaboration with AI tools.
The assignment breaks a complex research project into smaller, manageable parts. Each step, such as the research question, thesis, and methodology, is scaffolded to help students develop their thinking over time. This design supports students with varying levels of writing experience and gives instructors clear opportunities to track progress and provide targeted support.
Flexibility is one of the assignment’s greatest strengths. Instructors can modify the timeline, shift emphasis to match their discipline, or revise the reflection and presentation components to suit their course goals. Requiring students to submit AI transcripts promotes transparency and encourages classroom conversations about how AI tools influence the writing process.
This assignment works best in upper-level undergraduate courses where students are ready to develop and defend original arguments. Instructors who value inquiry, ethical technology use, and iterative revision will find this structure adaptable and practical for a wide range of classroom settings.
GETTING STARTED
To help students begin working with generative AI in a low-stakes, collaborative way, the course opened with an in-class AI Scavenger Hunt. This activity introduced students to using large language models as research assistants while encouraging them to think critically about how these tools search, summarize, and cite information.
Instructors can use this exercise to start early conversations about reliability, authorship, and bias. Students are asked to locate a variety of academic sources using AI, including videos, podcasts, articles, and interactive archives. They then evaluate the quality and credibility of what they find. Working in teams, students quickly see that while AI can suggest useful leads, human judgment is still necessary to determine what sources are valid.
This activity also helps surface common issues that may come up later in the semester, such as hallucinated citations, vague sourcing, or mismatches between prompts and results. It sets a strong foundation for future assignments by asking students to create full citations and explain why they selected each source.
The scavenger hunt can be adapted to fit any course topic or discipline. It works especially well as a first-week assignment because it builds student confidence using AI tools while also reinforcing collaboration and critical thinking. By the end of the exercise, students are better prepared to treat AI as a research partner rather than a replacement for academic work.
Research Question
The research question is the foundation of the semester-long project, and this assignment gives students a structured way to develop it using AI as a thinking partner. Instructors can use this step to help students learn how to ask better questions, narrow their focus, and begin a research process grounded in curiosity and clarity.
This phase of the course works well early in the semester. It helps students shift from general interest areas to specific, researchable questions. The guided prompts encourage students to brainstorm with AI, test out variations, and reflect on clarity, relevance, and scope. Because they work directly with AI tools, students learn that strong input often leads to stronger output.
Instructors can customize this part of the assignment to suit different disciplines. The same prompting strategies can be applied whether students are researching ethics in AI-generated art, the impact of automation on literature, or how generative tools reshape authorship. The assignment also includes reflection checkpoints that prompt students to evaluate their questions before moving forward.
This activity supports a habit of critical thinking and revision early in the writing process. It also gives instructors a clear place to step in with targeted feedback before students invest too much time in a direction that may be too broad or unclear. As a result, students build a stronger base for the rest of their research and writing.
Literature Review
The literature review is often one of the most difficult parts of a research paper for undergraduates, especially when working across disciplines or engaging with unfamiliar source material. In this course, students used AI to help navigate that challenge by developing early research strategies, drafting summaries, and refining their academic tone.
Instructors can use this approach to teach the purpose and structure of a literature review while also modeling how AI can support (but not replace) scholarly reading and synthesis. The assignment encourages students to begin with broad keyword searches and then narrow their focus by working with AI to generate topic-specific subfields. This helps students build a more strategic, layered approach to research.
Collaborative review and revision are key parts of this phase. Students are asked to share and compare their AI-generated drafts, align tone across group sections, and evaluate what needs further development. This gives instructors a chance to emphasize critical reading and human judgment, showing where AI output may be incomplete or misleading.
This assignment is especially useful in courses where students are just beginning to engage with scholarly conversations. It helps them map the field, identify gaps, and prepare to situate their own argument within an ongoing academic dialogue. Instructors can adapt this framework for a variety of subjects by customizing the prompt set or the sample topics. Used well, this activity builds both information literacy and a stronger sense of research as conversation.
Methodology
For many students, writing a methodology section can feel abstract or overly technical. To make this process more accessible and reflective, this course used a recursive prompting exercise with generative AI. Instead of presenting a fixed format or predefined structure, the assignment guided students through a personalized, question-based conversation that led to a clear and well-supported methodology.
Instructors can use this model to help students build research awareness and agency. The recursive prompt encourages students to answer one question at a time, each connected to earlier parts of their paper, such as the research question, thesis, or literature review. This process strengthens the connections between ideas and helps students make deliberate choices about how they will carry out their research.
As students respond to each question, AI helps them consider possible methods, compare strategies, and clarify the purpose of their analysis. This approach is especially effective in the arts and humanities, where methods often involve interpretation, comparative study, or close reading. By building the methodology step by step, students can better connect the process of research to the goals of their argument.
This assignment also introduces methodology as a process of reflection and refinement. Research is not something students are expected to get right on the first try. It improves through questioning, feedback, and revision. Instructors can adapt the recursive prompt to suit different subjects or course levels. What makes it useful is not just the format, but the way it teaches students to think critically, explain their choices, and use AI as a tool for deeper understanding.
Analysis
The analysis section is where students bring their research to life. It asks them to apply their chosen method, interpret their findings, and support their argument with evidence. In this course, students used a structured set of AI-guided prompts to develop their analysis in direct connection with their methodology. This approach helped them stay focused, organized, and intentional in how they presented their ideas.
Instructors can use this assignment to help students move beyond summary or opinion and into deeper interpretation. The AI prompts guide students through questions like what to analyze, how to structure their analysis, and what insights their method is likely to reveal. This helps students connect the dots between their research question, method, and final argument.
The assignment also models alignment. Students are asked to check whether their analysis uses the same language and framework as their methodology. This reinforces the idea that every section of a research paper should work together and that strong writing often comes from consistency and clarity across parts.
This structure works especially well for students in the arts and humanities, where analysis may focus on texts, performances, images, or cultural trends. Instructors can easily adjust the materials and prompts to suit different course topics. Whether students are interpreting a Shakespeare play, a digital artwork, or a media campaign, this scaffolded approach helps them organize their thinking, stay on track, and produce deeper, more focused analysis.
Final Draft
As students near the end of their research paper, one of the most important steps is making sure their arguments are fully supported with reliable sources. In this course, students used generative AI to help with final editing, quote integration, and citation formatting. This process made it easier for students to spot where their writing needed support and to insert evidence in a way that felt natural and clear.
Instructors can use this stage to shift the focus from drafting to refinement. The assignment guides students through six key steps: identifying weak points in their argument, finding reliable sources, inserting quotes or paraphrases, adding clear attributions, formatting citations, and reviewing the overall flow. With help from AI, students learn to strengthen their voice without losing it.
This approach is especially useful for students who may struggle with academic research or citation styles. Instead of correcting mistakes after the fact, students work step by step with AI to build their confidence in sourcing and attribution. The result is a more polished paper that shows both original thinking and engagement with existing scholarship.
Instructors can adapt this framework to different citation styles, disciplines, or course levels. It also works well for group editing sessions or one-on-one conferences. By using AI to support final editing, students are able to focus on the details that make their argument stronger, their structure clearer, and their research more persuasive.
Conclusion
The design of the class Writing with Generative AI is grounded in High Impact Practices that promote deep learning through sustained engagement, collaborative work, and structured reflection. This model combines traditional research instruction with AI-supported writing strategies, offering students a clear, step-by-step path from early brainstorming to final revision.
Each stage of the course builds intentionally on the last. From developing a research question to integrating sources and refining a final draft, students move through a scaffolded process that emphasizes critical thinking, ethical decision-making, and active learning. Generative AI tools are used not as shortcuts but as partners in inquiry, helping students explore ideas, weigh choices, and revise with purpose.
Instructors can adopt the full sequence or integrate individual elements into their own teaching. The assignments, lab activities, and AI prompts are adaptable to a variety of disciplines, including composition, digital humanities, media studies, and the arts. With small adjustments, this structure can support both introductory writing courses and advanced seminars.
This approach enriches the writing process rather than replacing it. By centering student agency and using AI to support, not supplant, original thinking, instructors can create meaningful opportunities for students to engage in rigorous, creative, and reflective work. The strategies in this chapter are designed to be flexible, practical, and ready to adapt to the needs of any classroom.
Wishing you success as you explore new ways to teach with AI. May your students be curious, your prompts effective, and your outcomes rewarding.
