By Joe Gladd. Part of MyEssayFeedback.ai's Critical AI Literacy series.
Text-to-text prompting applications: generation, classification, summarization, translation, research, paraphrasing, content editing, brainstorming, process analysis. Good prompts are the difference between generic AI output and genuinely useful assistance.
Input: Straightforward commands produce generic outputs. Context: Background information dramatically improves quality — tell the AI about the assignment, audience, purpose. Follow-up: Conversations are iterative — refine, add constraints, ask for revisions.
Role Assignment: "You're an expert first-year writing instructor" produces better feedback than a generic request. Constraints: Set word count, format, audience level, style. Chain-of-thought: Ask the AI to think step by step for more accurate outputs. Multi-step prompting: Break complex tasks into numbered steps, waiting between each.
A multi-step approach: Step 1 — Set the role and explain you'll provide rubric, then instructions, then the essay. Step 2 — Provide rubric criteria (introduction, supporting ideas, cohesion, counterargument, pathos/ethos, citations). Step 3 — Explain feedback style (notice strengths first, then areas for development, reference specific paragraphs). Step 4 — Paste the essay. Step 5 — Follow up.
ChatGPT's feedback varies in quality and is not a substitute for human feedback. MyEssayFeedback.ai streamlines this — instructors set up prompts once, students receive structured feedback with built-in reflection workflows.