🛠️ Want AI-generated discussion questions to spark writing ideas? Try our Discussion Question Generator: perfect for pre-writing brainstorming sessions.

You’re watching a student stare at a blank Google Doc for twelve minutes. They type a sentence, delete it, type another, delete that too. Meanwhile, across the room, another student has already “finished” their essay:except it reads like ChatGPT wrote it, because ChatGPT did write it. These are the two failure modes of writing instruction in 2026: paralysis and outsourcing.

Here’s my position: banning AI from writing class is like banning calculators from math class in 1995. It’s a losing battle, and it misses the point. The question isn’t whether students will use AI:they will. The question is whether we teach them to use it in ways that make their writing better rather than nonexistent.

This guide walks you through a complete AI-enhanced writing workshop model that keeps human creativity at the center while using AI strategically at specific stages.

The Core Principle: AI Touches the Process, Not the Product

Let me be blunt about what this is NOT: it’s not “have AI write a draft and then students edit it.” That’s not writing. That’s editing someone else’s work, and students learn almost nothing from it.

The model I’m proposing uses AI at four specific touchpoints:

1. Brainstorming: AI as thought partner (expanding ideas)
2. Drafting: Human only (no AI allowed)
3. Revision: AI as feedback tool (targeted criteria)
4. Editing: AI as proofreader (mechanics only)

The draft itself:the thinking-on-paper, the voice, the argument:that stays human. Always.

Stage 1: Brainstorming with AI as Thought Partner

Most students don’t struggle with writing because they can’t write. They struggle because they don’t know what to say. AI is genuinely excellent at helping students find their ideas.

How it works in practice:

Give students 10-15 minutes with ChatGPT (free tier works fine) or SchoolAI ($0–$15/month for monitored conversations). Their job is NOT to ask AI to write anything. Their job is to have a conversation about their topic.

The prompt students use:

I'm writing a [type of writing] about [topic]. I'm in [grade level].
Don't write anything for me. Instead:
1. Ask me 5 questions about my topic to help me figure out what I actually think
2. After I answer, suggest 3 possible angles I could take
3. For the angle I choose, help me think of 3 specific examples or pieces of evidence

Remember: ask questions, don't write for me.

Why this works: The AI acts like a writing conference partner:the role you’d play if you had time to sit with every student individually. It asks questions that force students to articulate their thinking before they write.

Teacher moves during this stage:

• Circulate and glance at conversations
• Look for students who are asking AI to write (redirect them)
• Look for students whose AI conversation reveals they need more background knowledge (pull them for a quick conference)

Stage 2: Drafting: Humans Only, No Exceptions

This is the non-negotiable. During drafting, AI tools are closed. Period.

I know this feels old-fashioned. But here’s the research reality: the cognitive work of translating ideas into sentences is where writing skill develops. Skip that step, and you’re building a house without a foundation.

Practical enforcement:

• If you’re 1:1 with devices, use a monitoring tool like GoGuardian or have students work in a specific app (Google Docs with version history you’ll check)
• If you want to go low-tech, do drafting on paper. Yes, paper. Students write differently (often better) by hand, and it eliminates the AI temptation entirely.
• Set a minimum time, not a minimum length. “You will write for 25 minutes” beats “write 500 words” because the latter incentivizes filler.

What to tell students: “The brainstorming stage was about figuring out what you think. This stage is about figuring out how to say it. That’s a skill only YOU can develop. AI can’t do your push-ups for you, and it can’t do your drafting for you.”

Stage 3: Revision with AI Feedback on Specific Criteria

Here’s where AI re-enters:but in a controlled, pedagogically sound way. Students don’t ask AI to “make this better.” They ask AI to evaluate specific criteria that you’ve defined.

The key distinction: Revision is about ideas, structure, and argument. It’s NOT about fixing commas. That comes later.

The prompt students use (you customize the criteria):

Read my essay below. Do NOT rewrite it. Instead, give me feedback on these 
3 specific things only:

1. Thesis clarity: Can you identify my main argument in one sentence? 
   If not, tell me what's unclear.
2. Evidence quality: For each body paragraph, is my evidence specific 
   and relevant? Flag any paragraph where the evidence is vague or missing.
3. Counterargument: Did I address the strongest opposing view? 
   If not, what counterargument am I ignoring?

Be honest. Don't be nice just to be nice. Tell me what's weak.

[paste essay here]

Why this is better than peer review (sometimes): Let’s be honest:peer review in most classrooms produces feedback like “good job” and “maybe add more detail.” AI gives specific, actionable feedback every time. It doesn’t get tired, it doesn’t worry about hurting feelings, and it addresses exactly what you asked it to address.

But peer review still matters: I’m not saying replace peer review entirely. Human readers catch things AI misses:like whether a piece is boring, whether the voice feels authentic, whether an argument is technically correct but emotionally unconvincing. Use both.

Teacher moves during this stage:

• Require students to document the AI feedback they received AND what they chose to do about it (a simple two-column log works)
• Review the logs:this tells you more about student thinking than the final draft does
• Conference with students whose AI feedback reveals structural problems they can’t fix alone

Stage 4: Editing with AI as Proofreader

Once the ideas and structure are solid, NOW students can use AI for surface-level editing. This is the least controversial stage because it’s essentially what Grammarly ($0–$12/month) has been doing for years.

Tools for this stage:

• Grammarly (free tier catches most errors)
• ChatGPT with a specific prompt
• Microsoft Editor (free, built into Word/Edge)

The prompt for ChatGPT-based editing:

Proofread this essay for grammar, spelling, and punctuation errors ONLY.
Do not change my word choices, sentence structure, or ideas.
For each error, show me the original and the correction so I can learn 
from the mistake.

[paste essay here]

Critical teaching moment: Require students to accept or reject each suggestion individually. This builds grammar knowledge over time. If they just click “accept all,” they learn nothing.

Stage 5: Publishing and Reflection

The final stage is where students share their work:and reflect on how AI shaped their process.

Reflection prompt (assign alongside the final draft):

Answer these questions about your writing process:
1. What idea did the AI brainstorming conversation help you discover 
   that you wouldn't have found alone?
2. What piece of AI revision feedback did you disagree with? Why?
3. What's one thing you're proud of in this piece that is 100% yours?

This reflection serves two purposes: it builds metacognition about writing, and it gives you evidence that the student actually engaged with the process rather than outsourcing it.

Addressing Academic Integrity Head-On

Let’s talk about the elephant in the room. If students have access to AI during brainstorming, revision, and editing, what stops them from using it during drafting?

My honest answer: Nothing stops them completely. But this framework dramatically reduces the incentive to cheat because:

1. They already got AI help at the hard parts (finding ideas, getting feedback)
2. The drafting stage has a time component (they’re writing in class, where you can see them)
3. The reflection and process log create accountability
4. Version history in Google Docs shows the drafting process

Detection isn’t the answer. AI detection tools are unreliable (see our article on that topic). Building a process where AI use is transparent and structured is far more effective than playing gotcha.

What to say to students: “I’m not trying to catch you using AI. I’m trying to teach you to write. These stages exist because research shows this is how writing skill develops. If you skip the drafting stage, you’re only cheating yourself out of getting better.”

Sample Weekly Schedule for AI Writing Workshop

Here’s how this looks across a typical 5-day writing unit:

• Monday: Mini-lesson on the week’s focus skill + AI brainstorming session (15 min)
• Tuesday: Drafting day (full period, no devices or paper-only)
• Wednesday: Continue drafting + early finishers begin AI revision
• Thursday: AI revision day (with feedback logs) + peer review
• Friday: AI editing + final revisions + reflection submission

Tools and Costs Summary

Tool	Use Case	Cost
ChatGPT Free	Brainstorming, revision feedback, editing	$0
ChatGPT Plus	Faster responses, better feedback quality	$20/month
SchoolAI	Monitored student AI conversations	$0–$15/month
Grammarly	Editing/proofreading	$0–$12/month
Google Docs	Drafting with version history	$0

You don’t need to spend anything to run this workshop. The free tier of ChatGPT plus Google Docs gives you everything described above.

Common Pushback (and My Responses)

“Students won’t learn grammar if AI fixes it for them.” That’s why I require them to review each correction individually. And honestly:did students learn grammar from red-pen corrections on paper? The research says no. At least with AI, the feedback is immediate and explained.

“This takes too much class time.” Writing workshop has always taken class time. This model doesn’t add time:it restructures how existing time is used. The brainstorming stage replaces your old “think of a topic” time. The revision stage replaces ineffective peer review sessions.

“What about students without devices at home?” This entire model is designed to happen IN CLASS. That’s intentional. Writing workshop should be a classroom activity, not homework.

Related reading

• AI Writing Feedback for Teachers: Tools That Actually Help
• How to Detect AI in Student Work (And Why It’s Complicated)
• AI Classroom Policies: What to Include

FAQ

Do I need any special tools to get started with this?

For most AI applications, you just need a ChatGPT ($20/month) or Claude ($20/month) subscription. Some tasks benefit from specialized tools, but you can start with a general AI assistant and add specific tools as your needs grow.

How much time will this actually save me?

Most teachers report saving 3-8 hours per week once they’ve established their AI workflows. The first week is slower as you learn, but by week 2-3, the time savings compound. Focus on the tasks you do repeatedly: that’s where AI saves the most time.

Is the output quality good enough to use directly?

Rarely use AI output without editing. Think of AI as producing a strong first draft that’s 70-80% ready. Your expertise adds the final 20-30%: context, nuance, and accuracy that AI can’t provide. Always review before sending to clients or publishing.

What are the biggest mistakes teachers make with AI?

The top three: (1) not providing enough context in prompts, (2) trusting output without verification, and (3) trying to automate everything at once instead of starting with one workflow. Start small, verify everything, and expand gradually.

Will AI replace teachers?

No. AI replaces tasks, not jobs. The teachers who use AI will outperform those who don’t: they’ll handle more clients, produce better work, and spend less time on repetitive tasks. The value shifts from execution to judgment and relationships.