Using AI as a Tutor for Writing Initiation

Victoria VanProoyen
Western Michigan University

Similar to the way in which a writing center tutor might initiate an intervention-focused tutoring session by being a writing expert, conference manager, and conversation facilitator (Mattison, 2019, pg.6), this lesson plan uses a teacher-created GenAI prompt to assist students in starting writing tasks. Developed from a case study in an “AI Writing” course at Western Michigan University, the lesson asks students to interact with GenAI platforms, such as Claude, to learn the first steps of an assignment and begin to critically evaluate AI-generated content. This lesson emphasizes the importance of ethical AI integration and personal input while also boosting students’ confidence and capability in beginning writing assignments. 

Learning Goals:

By the end of the lesson students should be able to: 

  • Understand and effectively use AI tools to assist in brainstorming, organizing ideas, and initiating writing tasks.
  • Develop the ability to assess and refine AI-generated content to meet academic standards and personal writing goals.

Original Assignment Context: Beginning of a course or as a scaffolding measure for students in need.

Materials Needed

  • Access to an AI platform such as Claude.ai, ChatGPT, MagicSchool, Perplexity, Google Gemini, etc.
  • Clear guidelines for the writing assignment, attached to the AI prompt.
  • (Optional) Prompt Log (spreadsheet or document template) to record AI prompts, responses, and observations.
  • (Optional) A brief survey where the student reflects on their interaction with the AI and/or provided prompt.

Time Frame

  • Lesson 1: Introduce GenAI tools, practice prompting, and cite GenAI use.
  • Lesson 2: Use given prompt and write draft (Optional: complete log and feedback/reflection collection).

Overview

Responding to the many misconceptions and negative associations regarding AI, Cao (2023) recommends that “Instead of perceiving generative AI as the one who ‘knows the answer,’ a more productive way is to view it as a ‘dialogue partner’” (p. 7). This sentiment aligns with Elizabeth Wardle's argument that student writers should avoid “answer-getting” dispositions and instead cultivate “problem-exploring” dispositions (2012). Both perspectives emphasize the value of engaging deeply with questions and exploration of ideas, rather than seeking immediate answers. 

This lesson plan was developed through an in-depth case study that investigated the potential of GenAI tools to aid students as they begin their writing tasks. As writing center studies have shown, initiation is a common struggle (CITE). The struggle to initiate is exacerbated in online learning environments where immediate teacher support is often unavailable. Through multiple iterations of GenAI prompts, the study identified key strategies for creating effective GenAI-assisted prompts for writing initiation. Feedback collected from a diverse group of testers led to more precise and helpful AI responses. This iterative process highlighted the importance of simplicity and clarity in prompt design. Ultimately, the process produced the following structured prompt: “Review the attached document. Then pretend to be a tutor to a struggling [grade level here] grade student. What is the first step the student should take according to the directions on the document? Please also offer help in getting started on that step.” This prompt is central to the lesson. It can be adapted for various writing tasks, guiding students to interact with GenAI in a way that simulates a tutoring session and making the GenAI's assistance as more personalized and actionable. By framing the GenAI interaction as a tutoring session, students can better understand the step-by-step guidance process. 

This approach demystifies GenAI's role, positioning it as a dialogue partner rather than an answer generator. This method encourages students to think critically about their writing process and engage more deeply with the AI's suggestions. 

Assignment

Step 1: Introduce AI Tools

  • Provide students with a brief overview of the chosen AI platform, including how to access, use, and cite it.
    • Tutor Prompting: Explain the concept of tutor prompting. Discuss how effective tutors guide their students through tasks by breaking down instructions into manageable steps, asking probing questions, and providing targeted feedback.
    • Reflection and Use: Provide opportunities for the student to engage with GenAI as a tutor and reflect on the usefulness of the responses. After receiving AI-generated guidance, students should reflect on the clarity and usefulness of the responses.
    • Citation: Explain the expectations for citing GenAI’s use in whichever format the course/assignment requires.

Step 2: Provide Prompt

  • Instruct students to use a prompt aimed at guiding the AI to help them start their writing assignment. 
    • Supply Prompt. As Mollick (2023) explains, “Structured Prompting is about turning the AI into a tool that does a single task well in a way that is repeatable and adapts to its user.” A different prompt may be provided but this is the one refined based on the case study data to work with any writing assignment: “Review the attached document. Then pretend to be a tutor to a struggling [grade level here] grade student. What is the first step the student should take according to the directions on the document? Please also offer help in getting started on that step.”

Step 3: Interact with GenAI

  • Students use GenAI to generate content using the given prompt. Optionally, students can use a log to document the GenAI’s responses and any adjustments they made to the prompts for improved results.

Step 4: Write Draft

  • Using the AI-generated content, students will draft their writing assignment. Emphasize the importance of personal input and critical evaluation of the AI’s suggestions.

Step 5: Reflect on Feedback and Reflection

  • Optionally, after completing their drafts, students complete a survey to provide feedback on GenAI’s effectiveness. Students reflect on how GenAI assisted their writing process and any challenges faced. When considering this optional  extension, assess how this lesson is being used (writing center context, content teacher assigned, tutor support). 

Step 6: Submit

  • Students submit their writing draft-and optional extras such as the prompt log or feedback survey.

Evaluation Criteria

  1. Students will be evaluated on their use of GenAI by reviewing the prompts and their integration into the writing assignment.
  2. Students will be evaluated on their ability to refine GenAI content to meet academic standards and personal writing goals.
  3. Students will be evaluated on the final writing draft based on coherence, structure, content, and adherence to assignment guidelines using a rubric.
  4. Optional Survey: Students will be evaluated on the depth of reflection on the GenAI's effectiveness and personal learning experience.
  5. Optional Log: Students will be evaluated on completing documentation requirements of GenAI prompts and responses, including observations and adjustments.

Acknowledgements

This assignment was developed as part of an “AI Writing” course at Western Michigan University, under the guidance of Dr. Brian Gogan. Special thanks to all participants who provided feedback on the AI prompts.

References

Cao, L., & Dede, C. (2023). Navigating A World of Generative AI: Suggestions for Educators. The Next Level Lab at Harvard Graduate School of Education. President and Fellows of Harvard College: Cambridge, MA. https://bpb-us-e1.wpmucdn.com/websites.harvard.edu/dist/a/108/files/2023/08/Cao_Dede_final_8.4.23.pdf 

Mattison, M., & Zebell, K. (2019). Taking Initiative: The Evolution of a Writing Tutor. WLN: A Journal of Writing Center Scholarship, 44(3-4). https://link.gale.com/apps/doc/A606944425/AONE?u=nm_p_oweb&sid=googleScholar&xid=258ff3c4 

Mollick, Ethan. (2023, November 1). Working with AI: Two paths to prompting. One Useful Thing. https://www.oneusefulthing.org/p/working-with-ai-two-paths-to-prompting#:~:text=Structured%20Prompting%20is%20about%20turning,and%20adapts%20to%20its%20user

Wardle, E. (2012). Creative repurposing for expansive learning: Considering "problem-exploring" and "answer-getting" dispositions in individuals and fields. Composition Forum, 26. Retrieved from https://compositionforum.com/issue/26/creative-repurposing.php