5.5 — AI as a Sparring Partner | AI for Real Life

Pressure-test your take before you submit.

If you've followed Module 5 honestly: you verified your sources, you read the relevant 10% closely, you took notes in your own words, you built a synthesis matrix, and you have a draft thesis. The lessons so far have been about getting to a take. This last lesson is about pressure-testing the take you have.

Four Sparring Prompts

Use them at different stages: 1 and 3 before drafting, 2 after outlining, 4 after a complete draft. Together they're the closest thing to having a smart reader available 24/7.

Prompt 1: Argue the opposite

Here's the thesis I'm planning to argue: [paste thesis statement, 1–3 sentences] And here's the outline of how I'd defend it: [paste outline] Your job: argue the strongest possible counter-thesis. Not "play devil's advocate." Actually try to make the case that I'm wrong. Use the same kind of evidence and reasoning a reasonable smart person on the other side would use. Don't pull your punches. Don't be polite. Don't end with "but ultimately your point is also valid." Make the case for the opposite as well as it can be made. I'll use your counter-argument to figure out where my thesis is weakest and where it needs strengthening.

Prompt 2: The skeptical professor in office hours

[Paste your draft thesis + outline.] You are a skeptical professor in [field] who has read drafts like this before. I'm in your office hours and you're going to push on my paper. Ask me 5 hard questions in sequence, the kind a real, slightly-tired professor would ask after reading a draft. Each question should: - Pinpoint a specific weakness in the argument or evidence. - Force me to commit to a sharper position than I'm currently taking. - Or expose where I'm hiding behind vague language. Don't ask easy questions. Don't ask questions that are basically restatements of what I've written. Ask the questions I'm hoping you won't ask. I'll answer them one at a time. After each of my answers, give me feedback: was that answer convincing? Where did it dodge? What would the prof follow up with?

Prompt 3: What's my paper missing? (run before drafting)

[Paste your draft thesis + outline + the synthesis matrix from Lesson 5.4 if you have one.] I'm about to start writing. Before I do, look at my outline against my source notes and tell me three things: 1. **What's load-bearing in my thesis but thinly supported?** Quote the part of my outline that's making a claim my source notes don't fully back up. Don't fill in the support; flag the gap. 2. **What objection to my thesis am I not addressing yet?** The 1–2 most likely counterarguments my outline doesn't mention. Don't make me address them; just name them. 3. **What's a perspective my sources don't cover?** A type of source I should look for, a viewpoint missing from my matrix, a stakeholder whose view isn't here. Be honest. I'd rather find these gaps now than after I've written 10 pages.

Prompt 4: What would change your mind? (run after the draft is written)

Here's the draft of my paper: [paste full draft] You're a smart reader who started skeptical of my thesis. By the end of reading the draft, did I move you? If yes, what was the most persuasive moment? If no, where did I lose you? Then: what's the single piece of evidence or argumentation that, if I added it, would make you most likely to be persuaded by the next draft? Don't suggest fluff. Name something specific that would actually change a skeptical reader's mind. I'm going to use your answer to decide what to revise.

Why all four are different

"Argue the opposite" gives you an opposing argument. "Skeptical professor" gives you the live questions you'll face. "What's missing" gives you the gaps before you draft. "What would change your mind" gives you the diagnostic on the finished work. Different tools for different stages.

The trap students fall into with sparring: and the fix.

🤷

Trap #1: Shallow acknowledgment

You read the counter-argument and add a sentence: "Some critics might argue X, however, [restate your thesis]." This is the AI pattern from Lesson 4.6: the both-sides hedge that lets you avoid actually engaging.

The fix: the counter-argument should change how you make your argument. If Claude's strongest objection is about scope, narrow your scope. If it's about evidence, find a specific source that addresses it. If it's structural, rebuild a paragraph.

🌀

Trap #2: Capitulation

You read the counter-argument and decide your thesis is wrong. You rewrite it to be safer, blander, less committed. Your paper softens into "well, both sides have a point."

The fix: the goal of sparring isn't to find the "correct" answer. It's to find the strongest version of your existing argument. If a counter-argument seems devastating, the move is usually to narrow the claim, not to abandon it. "X is true" might become "X is true under condition Y," which is a stronger paper, not a weaker one.

The mental model: sparring partner, not judge

The athletic metaphor matters. A sparring partner isn't trying to decide the championship. They're trying to make you better at your sport by giving you something hard to react to. You take the hits, learn what they exposed, and come back stronger next round. You don't ask the sparring partner who's right. You use the sparring session to figure out how to fight better. Run AI sparring the same way: it's input to your thinking, not a verdict on it.

Module 5: what you carry forward.

Five lessons, one habit: every claim survives scrutiny. The Citation Rule keeps fake sources out of your bibliography. The 5-Minute Workflow makes verification fast enough to actually do. The Structural Map prompt makes long PDFs readable in a weekend. The Synthesis Matrix turns notes into a thesis. And the four sparring prompts pressure-test that thesis before a professor does.

Phase 2 continues: Module 6 turns the camera at apps, resumes, and showing up.

Module 5 closes the academic-integrity flagship sequence. Module 6 is the next big payoff: college applications, supplementals, scholarship essays, resumes, cover letters, LinkedIn, mock interviews, and recommendation letters.

Continue to Module 6 → Apps, Resumes & Showing Up