Treat Prompting Like a Thesis Statement: Build AI Literacy Into What You're Already Teaching

Here's something practical I've learned from watching faculty across disciplines wrestle with AI integration: you don't need a separate unit on ChatGPT. You need to treat prompting skills the same way you treat writing a thesis statement - - as a teachable skill that reinforces your course objectives rather than competing with them. The most effective approach I've seen doesn't look like technology training at all. A history professor asks students to submit their prompts alongside their AI-generated drafts, then evaluates both. An economics instructor has students refine their queries through three iterations, documenting how each revision improved the output. They're not teaching "How to use AI." They're teaching precision, critical thinking, and revision. The AI just makes the underlying skill visible. Start with this simple framework: ask students to explain what they wanted, show what they asked for, and reflect on what they got. That's the same analytical process you already value in your discipline, just applied to a new tool. When a biology student learns to prompt for methodological clarity, they're practicing the same thinking they need for lab reports. When a literature student revises prompts to get more nuanced analysis, they're doing close reading by another name. The key is specificity. Generic prompts yield generic results - - this is actually the teaching moment. Students quickly discover that vague questions get vague answers, and that forces them to articulate what they actually want. That's hard for many undergraduates, and it's exactly the skill that makes better researchers and writers. Don't fight the AI. Use it to make your existing learning goals more visible and practiced.