GIFT-AI: Teaching the Game and Leveling the Field: Peer and AI Review + Reflection in a Business Writing Course

Synthesis: MacArthur et al. (2025) present the GIFT-AI approach — "teaching the game and leveling the field" — applying the Peer and AI Review + Reflection (PAIRR) model specifically to an upper-division Business Writing course (34 participating students at UC Davis in 2024). The model scaffolds major assignments so students receive peer review, then criteria-based chatbot feedback on the same draft, reflect on and critically assess both, and revise. The article shows how PAIRR builds AI literacy and writerly agency while leveling the playing field for students with differing preparation — particularly multilingual international students — and offers detailed implementation guidance for professional writing courses.

Key Findings

• The PAIRR model guides students in limited, guided use of chatbot feedback on drafts, requiring them to critically assess and reflect on both peer and AI feedback before revising — combining human-centered peer review with AI feedback in a way that protects writerly agency and builds AI Literacy.
• Students valued AI feedback for both higher-order and sentence-level comments but sometimes found it too general or "surface-level"; they appreciated peers' contextual and situational knowledge — e.g., "my peers looked at it from the employer's perspective… ChatGPT did not do that as much."
• Students did not simply trust AI's fluent prose: one quarter of coded reflections in the larger study expressed skepticism about or noted inaccuracies in AI feedback, a sign of developing AI literacy.
• For multilingual international students, the model addresses challenges of linguistic inadequacy, pressure to conform to standard language norms, and unfamiliarity with U.S. source-use norms — positioning AI as "an additional, supportive perspective" rather than a tool to "correct" deficient writing or a get-out-of-writing-free card.
• Three student profiles (Pengxi, high Self-Efficacy; Haoyu, low self-efficacy who initially over-relied on AI; Thuy, high-achieving first-generation student) illustrate how PAIRR encourages critical evaluation of AI feedback, recognition of peer-AI agreement, and preservation of students' own voices.

Study Design & Method

This is a Brief Research Report presenting the PAIRR model applied to a face-to-face and hybrid upper-division Business Writing course over a 10-week term at a public R1 university in the western U.S. The course (enrollment capped at 25) serves students majoring in management, economics, communications, and related fields. In 2024, 57% of students identified as Asian or Pacific Islander and 69% as multilingual; six were international students, 26% first-generation, 48% on financial aid. Students completed five major assignments (Job/Grad School Application, Internal Memo, Feasibility Study, Proposal, and a 5-minute Proposal Pitch), each requiring a draft, audience analysis, formal peer review by 2–3 peers, and revision. Data and findings draw on the larger PAIRR study (Sperber et al., 2025; N=654) and course-specific reflections, analyzed thematically following Saldaña in MaxQDA, with a focus on the Business Writing course and three multilingual international student profiles.

What this means for practice

• Instructors. Adopt the PAIRR sequence on major assignments: peer review of a full draft, then criteria-based chatbot feedback on the same draft, then a written comparison of the two, a revision plan, the revision, and a second reflection on which feedback changed what.
• Instructors. Assign the chatbot a supportive peer-reviewer role with explicit criteria and tell students it can be wrong: in the thematic coding of 131 reflections from the larger 654-student study, one quarter expressed skepticism about or noted inaccuracies in AI feedback.
• Instructors. Have students weigh both sources instead of picking one: when peer and AI feedback agreed, students found it reassuring, and when they differed, each source usually supplied complementary advice.
• Instructors. Route context-heavy judgment to people and sentence-level work to the tool: students reported that peers understood assignment and course context better, while AI feedback was constructive and actionable.
• Instructors. Treat guided AI use as an equity measure and keep it opt-in: underprepared students appear less likely to use AI and more likely to misuse it, so teach the criteria explicitly and let students who decline work with peer review alone.

Limitations

• As a Brief Research Report focused on one course, the study’s findings are largely descriptive and course-specific, and the article draws on the larger PAIRR study for its broader claims.
• The Business Writing course sample is small: 34 participating students of 46 enrolled.
• AI policies vary substantially across institutions and courses, as the authors note.
• The applicability of the model across different writing courses and contexts is discussed as future work rather than empirically demonstrated here.

Citation

MacArthur, M., Minnillo, S., Sperber, L., Whithaus, C., & Stillman, N. (2025). GIFT-AI: Teaching the game and leveling the field: Peer and AI Review + Reflection in a business writing course.