Small Bytes

Small Bytes Blog

The AI Small Bytes is a blog written by members of the AI Resources & Guidelines Committee. This blog will be updated periodically with new resources and information, and we hope that you will check back often. If you would like to talk about teaching with artificial intelligence and your concerns or ideas, please feel free to contact Lisa Low, AI Faculty Fellow, or Suzanne Tapp.

AI Teaching Resources

AI Tools for Research & Writing | University Libraries

Prompting Purposefully: Teaching AI Rhetorical Literacy in Grant Writing

by Dr. Rich Rice, Department of English, Technical Communication & Rhetoric

When generative AI first began to appear, I was interested more by what AI tools could reveal about writing itself than the content they could create. In my work with students preparing for careers in nonprofit and public-sector organizations, I noticed a gap: even as AI was fast becoming ubiquitous in professional grant writing contexts, few knew how to compose and collaborate with such tools strategically or ethically. Most attempted to use the technology as expedient; and others, distrusting it, kept it at arm’s length. What I saw was an opportunity to teach AI-use as a rhetorical act.

This semester I have developed new approaches for ENGL5391: Grant Writing for Nonprofit Organizations, where students learn strategies to identify community needs, articulate ideas for solving problems, and then write compelling proposals aligned with organizational missions. The genre of grant writing has long called for attention to audience, precision, and persuasive clarity. Now, with AI increasingly part of nonprofit workflows, students also need to learn algorithmic literacy, or the ability to understand ways in which the prompts they craft can shape their work.

Rather than thinking of AI as a static generator of text, these new approaches let students explore prompt engineering as a dynamic rhetorical process. In addition to deep diving SWOT and SMART analysis of problem-solution ideas, which are common tools in proposal writing, we experiment with three core techniques to better know what our intended audience knows and needs to know:

Mad Lib prompts, which foreground genre conventions by asking students to fill in rhetorical gaps in strategic ways;
Persona-based prompts identify the voice of program directors or community stakeholders in thinking through audience and tone; and
Flipped interactions ask AI what questions we should prompt it with, making our critical process more dialogical.

Specifically, Mad Lib prompts are partially structured templates with key rhetorical elements intentionally removed. Students “fill in the blanks” with purpose-driven choices such as audience, justification, genre moves, tone, and evidence. For example, “In our grant proposal, we seek to secure funding from [specific funder] to support [project focus], which responds to the pressing need for [identified community organizational problem]. By demonstrating [type of evidence] we show that our effort aligns with the funder’s priority of [funder goal]. In the end this proposal positions our team to [desired impact or outcome], furthering long-term sustainability for [target group].” Try it! Copy what’s in quotes above and fill in what’s in brackets for your next proposal.

Persona-based prompting enables students to assume a more plausible voice, purpose, constraints, and tone of a stakeholder. The audience could be anyone from a program director, a nonprofit leader, a faculty member, or a community advocate. As such, the approach can also allow writers more authentically to make meaningful rhetorical choices as they write on behalf of others. For example, a student might begin to dig in and deepen written understanding with an AI guide-on-the-side by feeding it something like: “As the Director of Community Literacy Initiatives, I need to convince city council members that this program makes neighborhoods more resilient. When writing to council members, I highlight our data on improved adult literacy rates and express myself in civic, solution-oriented tones.” Results provide writers with a model of language use that simulates a more authentic style.

And flipped interactions are particularly valuable, teaching students to ask AI what they should prompt next. It is in such a shift that the process becomes metacognitive and dialogic: students review what AI has suggested as next questions, identify gaps, and develop far more strategic ways of engaging writing with the tool. One student, while drafting a needs statement, fed AI, “What questions should I ask you so that I better understand systemic factors shaping this issue?” The tool suggested considering root causes, affected populations, existing interventions, and available data. Looking at the suggestions, she then noted which topics needed further investigation and which kinds of prompting might allow the tool to better support analysis.

Together, Mad Lib prompts, persona-based prompting, and flipped interactions allow students to dig. They learn to dig using AI’s suggested questions and directions as content to sift through. In the course I ask students to document precisely how they use AI not to police their use but as a reflective practice. When students express concern that AI “gets things wrong,” we treat such moments as teachable opportunities: places where rhetorical misalignment becomes visible and revision becomes an act of learning rather than correction.

What I’ve discovered is that AI does not reduce the intellectual labor of effective grant writing; it reveals it. Students iteratively develop sharper critical judgment in working with prompts, revising and editing AI-generated text, and comparing outputs against real-world models. They come to learn how to discern fundable writing from generic writing, where AI has inadvertently reproduced bias, missed key organizational values, or glossed over important community contexts. We’re building a new literacy powered by rhetoric and technology that “reverse engineers” logic models. In turn, this enables me to introduce more challenging writing tasks to students without overwhelming them: early drafting supported by AI frees them to expend energy on revision, strategy, and rhetorical nuance.

We’re working toward building valuable tool chests such as context engineering, which is less about the content itself and more about the conditions one creates to shape how the content is interpreted, such as situational framing, institutional positioning, and rhetorical moves to align with a funder’s mission. These artifacts will inform a forthcoming textbook I’m co-writing, Grant Writing Solutions Using Rhetoric and Technology, and contribute to scholarly work I’m writing on algorithmic literacy.

More broadly, this project speaks to Texas Tech’s commitment to thoughtful, research-informed AI pedagogy. I am inspired by work across campus, from the initiative on AI for Assessment and Feedback to the Scholarship of Teaching and Learning collaborations to the Library’s AI-Literacy Certificate program, each carefully scaffolding us to think differently about teaching in an era of AI-augmented writing. We are charged with adopting emergent “teach-nologies” to adequately prepare students for future workplaces while also modeling responsible, reflexive, and rhetorical use.

To me, for students to learn how to prompt with purpose is to learn how to write with purpose. Even more important than “digging” with AI is learning to work with intention so students cultivate learning that uncovers and builds meaning. We have the opportunity and responsibility of helping them understand how to use AI in ways that are ethical, strategic, and critically informed, supporting professional readiness. This is work that the community at Small Bytes explores, and I’m grateful for the chance to contribute to the conversation.

For Further Reading

Cole, R., Maher, L., & Rice, R. (2025). Rhetoric, technology, and the digital online media literacy assessment framework in grant writing. In T. Hallaq & C. Groshek (Eds.), Modern media literacy: Generative AI, social media, and the news (pp. 231–260). IGI Global. Print. https://doi.org/10.4018/979-8-3373-0872-2

Geroimenko, V. (2025). Key techniques for writing effective prompts. In The Essential Guide to Prompt Engineering: Key Principles, Techniques, Challenges, and Security Risks (pp. 37–83). Springer Nature Switzerland.

Li, R. (2025). Critiquing ChatGPT compositions: Collaborative annotation as an approach to enhancing students’ metalinguistic awareness of AI-generated writing. Thresholds in Education, 48(1), 7–24.

Stojković, N., & Wang, S. AI meta prompting as cognitive scaffolding in teaching academic writing. Social Science Research Network (SSRN). http://dx.doi.org/10.2139/ssrn.5295837

by Dr. Rich Rice, Technical Communication & Rhetoric, Fall 2025