Before the Paragraph, After the Prompt: Collaborative Invention with Generative AI

Jeanne Beatrix Law
Kennesaw State University

This assignment guides writers to integrate generativeáAI into an intentional composing process. Both models presented were developed and tested in multiple contexts at a diverse public research university. First, students create their three-prompt portfolio, iteratively applying the RhetoricaláPromptingáMethod (RPM) to align purpose, audience, genre, and style. Then, they write a 300-word reflection, deploying the EthicaláWheeláofáPrompting (EWP) that audits AI exchanges for usefulness, relevance, accuracy, and harmlessness. Lastly, their work culminates in concept maps/outlines, visualizing revisions from prompt to edited output. The accompanying rubric rewards rigorous, human-directed editing. By cycling deliberately through successive prompts, writers plan rhetorical choices, approaching metacognition rather than one-click shortcuts. In student-reported data, we found this disciplined iteration keeps students in charge of human-AI collaboration, easing FOBR jitters and preparing students to craft responsible, polished prose in AI-infused learning environments.

---

Learning Goals: 

In completing this assignments, students will:

• Develop rhetorical awareness through iterative prompting;
• Engage AI as a co-author during invention;
• Apply ethical reasoning to collaborative writing using the Ethical Wheel of Prompting;
• Reflect on authorship, originality, and revision as recursive processes.

Original Assignment Context: ENGL 1101 (college composition with a focus on argument-writing) at Kennesaw State University, a public R2 institution, with a first-year student population of more than 10,000, and more than 45,000 students total across undergraduate and graduate programs.

Materials Needed

• Access to ChatGPT, CoPilot, or a similar LLM (can be free-tier)
• Rhetorical Prompting Method Handout (OER, CC-BY licensed)
• Ethical Wheel of Prompting Handout (OER, CC-BY licensed)
• Google Docs or Word for annotation

Time Frame: 

• One 50-minute class session for introduction, demonstration, and group practice
• One homework assignment (~60 minutes) for individual prompting and annotation
• One follow-up session for discussion and reflection
• Integrated into a 2-week invention scaffold for a larger writing project

Overview: This low-stakes assignment introduces students in first-year composition to rhetorical prompting and ethical collaboration with generative AI during the earliest stages of writing. Rather than treating large language models as post-draft editors or ghostwriters, this assignment reframes generative AI as a co-authoring partner in invention, using two open-educational frameworks: the Rhetorical Prompting Method (RPM) and the Ethical Wheel of Prompting. Students work through structured, recursive prompts to generate and refine ideas with a GPT, then reflect on the rhetorical choices embedded in the genAI’s responses/outputs, ultimately deciding what to keep, what to revise, and what to discard.

I first piloted this assignment in Spring 2023 at Kennesaw State University with two sections of ENGL 1101 (First-Year Composition I), revising it for clarity and reflection depth for the current iteration. Students reported greater confidence in topic discovery, genre awareness, and audience adaptation when engaging with this assignment as part of their writing process. They also voiced nuanced concerns about authorship and originality, prompting valuable classroom discussions about ethical boundaries and human agency in AI-assisted writing.

Not every student embraced the method—some resisted using AI altogether—but that resistance, too, became part of the reflection process. Ultimately, the assignment helped students see invention as dynamic and dialogic, whether with peers, sources, or machines. Because the frameworks are licensed under Creative Commons (CC-BY), this assignment can be remixed for varied writing contexts, including creative nonfiction, multimodal writing, or other professional communication courses.

---

Assignment

WhyáThisáMatters

Effective writing doesn’t begin with the first sentence; it begins with clear decisions about purpose, audience, genre, tone, and style and with mindful checks for usefulness, relevance, accuracy, and harmlessness. In this assignment you will practice both, infusing your process with generative AI and using two models to guide you.

Frameworks Students Will Need (Also Available as Separate Files: OER)

These frameworks were developed and tested with students like yours, who want to use generative AI effectively and responsibly. The Rhetorical Prompting Method (RPM) might look familiar to them, because it is a “riff” on the writing process they already know.

The second model, the Ethical Wheel of Prompting (EWP) helps students make sure they remain in control of their voice and ideas when they use generative AI. Instruct them to ask the four questions going through the Wheel, knowing that it’s OK to return to previous steps in their prompting process if they aren’t sure that the outputs meet their needs vision. Knowing when an output just doesn’t “pass the smell test” is part of being an effective writer.

Assignment Task 1 – Creating a Rhetorical Prompt Portfolio (Using Rhetorical Prompting Method-RPM)

You want to be able to precisely design and refine prompts that are purpose-built, audience-aware, and genre-specific.

1. Choose a topic or idea you want to explore (option: instructor provides)
2. Set the Stage (RPM Stepsá1-4) – Decide writing purpose, audience, tone, genre.
3. Refine the Request (RPM Stepsá5-7) – Specify style, context, must-include facts.
4. Quick Polish (RPM Stepá8) – Scan output for length, sentence structure, grammar.
5. Annotate each transcript – identify RPM step used and one choice driven by EWP.

Template starter prompt:

 

Example using student-chosen topic:

For each of the following parts of the Rhetorical Prompting Method (RPM), students type their own ideas and words into an AI’s chat-box. It might look something like this, with italics representing what students type in the box. I theme my ENGL 1101 courses around The Atlanta Student Movement, using oral histories as primary sources. You can check out the OER oral histories here: https://soar.kennesaw.edu/handle/11360/2384. You can change the example below to align with your own course theme.

Interactions between students and AI:

Purpose: Please draft notes for of a Rogerian op-ed persuading the Georgia House Public Safety Committee to reconsider policing practices affecting undocumented college students.
Audience: Committee members and general public readers of the Atlanta Journal-Constitution.
Tone: Constructive, empathetic, solutions-oriented.
Genre: 250-word op-ed written to legislator but with general newspaper audience in mind first.
Style: Plain-language, 12th-grade readability; incorporate one rhetorical question.
Context: Connect the 1960 Appeal for Human Rights with 2025 concerns about campus policing.
Facts: Quote the grievance on “police intimidation,” include GA HBá1234 campus-deportation data, and reference RoslynáPope’s authorship.
Minor Editing: I’ll polish for clarity after you give me your draft.
**(EWPáREMINDER: reread before sending my prompt before hitting the “ENTER” button**

Assignment Taská2 – Reflection (Using Ethical Wheel of Prompting-EWP)

Students need to be able to ensure that their writing is truly theirs, along with their own ideas. By holding themselves and generative AI accountable in the following ways, they make sure that they remain in control of the editing and use of the AI’s outputs.

Write ~300 words answering the four EWP questions using the EWP as a model:

1. Did I reread my prompt before sending it?
2. Did I read the AI’s response carefully?
3. Did I check that response for usefulness, relevance, and accuracy?
4. Did I edit the output for usefulness, relevance, accuracy, and harmlessness?

Reflections demonstrate that students are reading and evaluating inputs and outputs, getting closer to metacognition when they make rhetorical editing choices.

Assignment Taská3 – Annotate with a Concept Map or Outline

In this task, students how visually how their topic/idea moved from Prompt → AI Output → Human Edits. Ask them to label each arrow with the RPM step and EWP check they apply. Focus on their actions as the human in control of the AI process. Table 1 gives an example of how to do this.

Beginnings of an Example Topic: An Appeal for Human Rights by Roslyn Pope (1960) is relevant in 2025 when we think about how law enforcement interacts with undocumented students on college campuses in Georgia.

Assignment Deliverables and Timeline

Assignment Assessment Rubric (abbreviated)

Quick Reference Checklists (also see handouts)

RPM mini-checklist:

• Purpose
• Audience
• Tone
• Genre
• Style
• Context
• Facts
• Minor Editing

EWP mini-checklist:

• Read my input   
• Read AI’s output    
• Checked usefulness/relevance/accuracy (URAH)   
• Edited for U/R/A/H

Submission checklist:

• 3áAnnotated Prompts (only v1 shown here; v2 &áv3 attached in LMS)
• 300-word Ethical Reflection
• Concept Map PNG + Table

---

Acknowledgements

This assignment is part of a larger body of work developing the Rhetorical Prompting Method and the Ethical Wheel of Prompting, both of which are openly licensed under Creative Commons (CC-BY) and available for reuse and adaptation. I thank the students in my general education and graduate courses at Kennesaw State University who courageously and critically engaged with generative AI in its most experimental classroom form, beginning in November, 2022.

Special thanks to the Generative AI Teaching and Learning Community at Kennesaw State University for its support and feedback, and to colleagues at Macmillan, where early drafts of this assignment were published and critiqued in the Bits on Bots blog series. The design of the Ethical Wheel of Prompting was informed by conversations about algorithmic authorship, epistemic humility, and human-centered prompting, many of which were shaped by work from the MLA-CCCC Joint Task Force on Writing and AI and the OpenAI Educators Community Group.

This project also draws from open pedagogy models and ethical prompting practices developed in collaboration with students and educators in the eight-course Coursera series "AI for Everyday Life, Education, and Professional Writing," which I co-authored with Dr. Tammy Powell and Brayden Milam in 2024.

Prompting materials used in this assignment were generated and edited using ChatGPT-4o and-o3, under my direction and in accordance with my ethical authorship practices. The textbook that is aligned to the assignment is the OER OpenStax Writing Guide (2021, Robinson).