Course: Using Artificial Intelligence (AI 101) · Silver Oak University (fictional sample) · Prof. Quinn
Module: Week 10 of 16 · Fall 2026 · in-person, two 75-minute sessions
Objective covered: Objective 4 — Critically evaluate and verify AI output — recognize hallucination, sycophancy, and bias, and run a reliable verification workflow.

This file holds two pieces: (A) the Module 10 Overview page ("Start Here") and (B) the Welcome Announcement that drips out when the module opens. Dates assume a Tuesday/Thursday lecture pattern with Week 10 meeting Tue Nov 3 and Thu Nov 5, and end-of-week work due Sunday Nov 8, 11:59 p.m.

(A) Module 10 Overview — Start Here

Welcome to Week 10: The Course's Central Discipline

This is your home base for the week. Read it first, then work the checklist below from top to bottom. Everything you need is linked inside the module. Bring your laptop to class — we run the verification workflow live.

Nine weeks in, you have been doing something every single week: catching the AI's mistakes. Catching a fabricated "fact" about campus in Week 1. Catching a hallucinated citation in Week 5. Catching sycophantic agreement in the Studio whenever you pushed back on the AI's plan. That thread wasn't accidental — it was preparation for this week. Week 10 is where we make that discipline explicit, systematic, and deep.

AI is confident by design. It generates fluent, specific, well-structured text whether that text is accurate or fabricated. This week you learn the predictable shapes of hallucination — the five recognizable forms that AI-generated falsehoods take — and you learn a four-step verification workflow you can run on any AI output, for any purpose, in any context. This is Skill 10, and it is the central discipline of the whole course.

The week's big question

"How do you actually know whether what the AI just told you is true?"

By Sunday you'll be able to name the five shapes of hallucination, explain sycophancy and how to counter it, and run a four-step verification workflow on any claim an AI makes.

By the end of this week, you can…

Use this as a checklist. If you can do all five out loud, you're ready for the quiz.

• [ ] Name the five hallucination shapes — invented citations, fabricated statistics, fake case law, wrong arithmetic, and fabricated quotes — and give an example of each.
• [ ] Explain sycophancy — why AI tends to agree with users even when they're wrong — and describe how to counter it in a conversation.
• [ ] Run the four-step verification workflow — ask for and check sources; cross-check in a second model; ask the AI to critique itself; verify in an authoritative external source.
• [ ] Explain why confident tone ≠ accuracy — and why asking the same model to check itself is not a complete verification strategy.
• [ ] Conduct a Hallucination Hunt — deliberately elicit fabrications, document their shapes, run the workflow, and report what was real vs. fabricated.

What's due this week, and when

Work these in order — each one builds toward the next.

AI policy reminder: you are required to use AI on the tutorial, discussion, assignment, practice, and Studio. AI is not allowed on the quiz, which checks that you personally understand the concepts. And this week more than any other — you will be catching the AI's mistakes as the assignment itself.

Late policy reminder: 10% off per day late. If life happens, reach out before the deadline.

How to succeed this week

• Do the Studio before the quiz. Running the Hallucination Hunt in the Studio will make the quiz vocabulary concrete and memorable — the five shapes will feel like things you have seen, not just definitions.
• Start the Studio early. The Hunt requires you to actually run verification steps — check citations in a database, cross-check in a second model — which takes real time. Budget 60–90 minutes.
• Come to class with a question ready. Tuesday's live demo will walk through a full fabricated citation being caught step by step. If you've read ahead, you'll see the method more clearly.
• For the discussion: take a real position. "When is it irresponsible to trust AI?" is a genuinely hard question. Come with an answer and be ready to defend it — the rubric rewards analysis, not a safe hedge.
• Remember: confident ≠ correct. This week's core idea in one sentence. Write it somewhere you'll see it.

(B) Welcome Announcement — Module 10

Release setting: post on the module's start day, i.e., Tue Nov 3, 2026. If your platform won't preserve the scheduled date on import, post as a draft labeled "Release: Tue Nov 3."

Subject: Week 10 — this week we go deep on catching AI's mistakes 🔍

Hi everyone,

Quick question before we start: have you ever caught an AI giving you a specific-sounding fact — a date, a citation, a percentage — that turned out to be wrong? How did you catch it? What did you do?

If the answer is "I didn't catch it" — this is the week that changes that.

Week 10 — Verification, Hallucination & Critical Thinking — is the central discipline of this course. Every week so far, you've caught one mistake in a Studio or assignment. This week we make that skill explicit, systematic, and deep. You'll learn the five predictable shapes of AI hallucination, why confident tone means nothing, what sycophancy is and how to fight it, and a four-step verification workflow you can apply to any AI output.

The Studio this week is the Hallucination Hunt — you deliberately elicit fabrications from an AI, document the shapes, run the full workflow, and write a report of what was real versus made up. It is the most eye-opening assignment of the term for most students. Start it before the weekend.

Three things not to miss:
1. Lecture Tutorial 10 — the verification concepts, taught by your own AI tutor. Due Sun Nov 8.
2. Studio 10 (Hallucination Hunt) and Quiz 10 — both close Sun Nov 8. Start the Studio early.
3. Discussion 10 — "When is it irresponsible to trust AI?" — initial post by Fri Nov 6.

One note: Week 11 begins the Cowork deep-dive (agents, projects, file automation). If you haven't installed the Claude desktop app yet, now is a good time to check that your laptop can run it — you'll need it next week.

Bring your laptop and a healthy skepticism to class Tuesday.

See you soon,
Prof. Quinn