Course: Using Artificial Intelligence (AI 101) · Silver Oak University (fictional sample) · Prof. Quinn
Module: Week 16 of 16 · Fall 2026 · in-person, two 75-minute sessions
Objectives covered: cumulative — Objectives 1–8 (Weeks 1–15): what generative AI is and how it works; effective prompting; modalities and tool selection; verification and critical thinking; Claude Cowork projects, files, skills, connectors, and artifacts; automation with scheduled tasks, dispatch, computer use, Chrome, and Excel; responsible AI: privacy, ToS, IP, bias, and ethics; integration and capstone.

This file holds two pieces: (A) the Module 16 Overview page ("Start Here") and (B) the Welcome Announcement that drips out when the module opens. This is finals week — it works differently from a normal week. Dates assume a Tuesday/Thursday lecture pattern with the Week 16 in-class review on Tue Dec 15; the Final window opens Mon Dec 14 and the exam is due six days later. Adjust day-of-week and times to match your section.

---

(A) Module 16 Overview — Start Here

Welcome to Week 16: Final Review, Exam & Capstone

This is your home base for the last week. Read it first, then work the checklist below from top to bottom. Everything you need is linked inside the module.

Heads-up: this is finals week, so it runs differently. There is no quiz, no discussion, no assignment, and no AI Build Studio this week — the comprehensive Final replaces all of them. The week is built to get you ready: we spend our class session reviewing the whole course, you work through a three-part prep kit, you submit and document your capstone, and then you sit the exam. The Final is cumulative over Weeks 1–15 (Objectives 1–8) — every AI concept, prompting technique, modality, tool, verification practice, Cowork feature, automation rule, and ethical principle we have covered is in scope.

---

The Capstone — Design, Build, Verify & Reflect

The capstone is the culminating synthesis of everything you have learned in AI 101. It is assessed under the Final; it does not create a new grade category.

What the capstone asks you to do:

1. Design a real AI-powered workflow or automation that solves a genuine problem in your academic or professional life. Ideally this runs in Claude Cowork, but a well-documented multi-tool workflow using any combination of the tools from this course qualifies. The problem and the solution must be real — not hypothetical.

2. Build it. The workflow must actually run. Document the tools and Cowork features you used (project, connected folder, skill, connector, scheduled task, etc.) and show at least one screenshot or step-by-step log of it working.

3. Verify it. Deliberately inspect the output. Find and document at least one error, hallucination, over-promise, or privacy risk in what the AI produced. Explain what went wrong and what you did to fix it — a revised instruction, a tighter constraint, a verification check added to the workflow. This is the embed-don't-trust step that the whole course has been building.

4. Reflect on the ethics. In 150–250 words, address: What data did you connect or provide to the AI? What did you choose NOT to automate, and why? What is the safe-use posture for this workflow (least privilege, approval checkpoints, what to never automate)? Would this workflow pass your own billboard test?

Scope note: The capstone should represent genuine work — something you would actually keep running. It is not graded separately; it is the practical demonstration that grounds your understanding of Objective 8 on the Final.

---

The Week's Big Question

"Across all eight objectives — how AI works, how to prompt it effectively, which tools to use, how to verify output, how to build and automate with Cowork, how to use AI responsibly, and how to put it all together — can I name the move each topic requires, catch the classic mistake, and demonstrate the embed-don't-trust discipline?"

By the end of the week you will have swept the entire Objective 1–8 arc, confirmed where your knowledge is solid, found and fixed any remaining gaps, completed your capstone, and shown what fifteen weeks of building, prompting, and verifying built.

---

By the End of This Week, You Can…

Use this as a checklist. If you can do all eight out loud, you are ready for the exam.

• [ ] Explain how AI works and its limits (Obj 1) — describe next-token prediction, the context window, and why AI can be confidently wrong; distinguish AI from a search engine; name what the Turing test does and does not prove.

• [ ] Apply effective prompting (Obj 2) — use conversation, content-providing, emphasis (Markdown, XML tags, CAPS), meta-prompting, the structured-prompt components, zero/one/few-shot examples, and simulations to get high-quality results; counter sycophancy; fix a weak prompt.

• [ ] Work across modalities and choose the right tool (Obj 3) — execute the record→transcribe→analyze workflow; match tools to tasks (chatbots, image, audio/music, video, research, coding); explain why no single chatbot does everything equally well.

• [ ] Verify AI output (Obj 4) — recognize hallucination shapes (invented citations, fabricated statistics, fake quotes, wrong arithmetic); identify sycophancy; run the four-step verification workflow; explain why asking the same AI to check itself is not reliable.

• [ ] Use Claude Cowork to build agentic workflows (Obj 5) — distinguish agent from chatbot; define project, connected folder, skill (SKILL.md), connector (MCP — open standard from Anthropic), live artifact (refreshes from connectors), and plugin (bundle); explain when you need a connector vs. a skill.

• [ ] Automate safely (Obj 6) — explain the scheduled-task constraint (awake computer + open app); describe dispatch; match computer use / Chrome / Excel to their correct surfaces; state the absolute money rule; identify prompt-injection risk and defensive habits; place approval checkpoints before irreversible actions.

• [ ] Apply responsible AI practices (Obj 7) — apply the billboard test; identify data types that must never be pasted (HIPAA, FERPA, PCI, proprietary); explain ToS and data-retention basics; describe the current IP/copyright landscape (with "not legal advice" caveat); explain AI bias; recognize academic integrity violations; troubleshoot context overload.

• [ ] Integrate and verify an end-to-end workflow (Obj 8) — design, build, document, and verify a real AI-powered workflow or automation; demonstrate the embed-don't-trust discipline; write an ethics reflection.

---

What's Due This Week, and What to Do

Work these in order — each one gets you ready for the next. This is the finals-week list; there is no quiz, discussion, assignment, or Studio here — the Final stands in for all of them.

#	Do this	Type	Due
1	Come to the in-class review (Tue Dec 15) and skim the Week 16 review slides (Deck 16) and the review lecture outline	Prep (ungraded)	Alongside class
2	Work the Study Guide (M) — the checklist of every concept across Objectives 1–8, with fresh practice items; do this first so you know what to drill	Prep (ungraded)	Before you sit the exam
3	Run the Exam-Prep Tutorial (N) — an adaptive cumulative review with one approved chatbot (ChatGPT, Claude, or Gemini); when you finish, submit the conversation share link	Exam-Prep Tutorial · graded (Lecture tutorials, 5% group)	Before the Final closes
4	Take the Practice Final (O) — sit it timed, like the real thing, then review every miss against the Study Guide	Practice · ungraded	Before you sit the Final (strongly recommended)
5	Complete and document your Capstone — design, build, verify, and write the ethics reflection for your end-to-end AI workflow	Capstone documentation · assessed under the Final	Before the Final closes
6	Sit the Final (L) — cumulative over Weeks 1–15 / Objectives 1–8; AI is not permitted	Final · graded (Final group, 25% of the course grade)	Window opens Mon Dec 14; due six days later

There is no Quiz 16, no Discussion 16, no Assignment 16, and no Studio 16 this week — the Final stands in for all of them. The Study Guide, Exam-Prep Tutorial, and Practice Final are your prep kit; the Final is what's graded.

A note on AI and the prep work: The Exam-Prep Tutorial is designed to be run with an approved chatbot — use it honestly. The tutor will sometimes give you incorrect Cowork feature descriptions, misstate the scheduled-task constraint, fabricate a citation, or claim AI models are fully unbiased; catching those errors is part of your preparation. AI is allowed only for the prep tutorial and the capstone build — not on the Final itself.

Late policy reminder: 10% off per day late — and the exam window is firm at the end of the term. If life happens, reach out before the deadline; I would much rather hear from you early than after.

---

How to Succeed This Week

• Review actively, not passively. Don't re-read notes — do the moves. Explain next-token prediction to a friend. Fix a weak prompt. Name the four Cowork terms from memory and match them to their definitions. State the scheduled-task constraint out loud. Apply the billboard test to a hypothetical scenario.

• Lean into Objectives 5–8. The midterm already tested Objectives 1–4, so those are your foundation. The Final weights the Cowork and automation material (Obj 5–6), the ethics and privacy material (Obj 7), and the capstone integration (Obj 8) most heavily.

• Use the prep kit in order. Study Guide → Exam-Prep Tutorial → Practice Final. The tutorial finds your weak spots; the timed practice final tells you whether you have fixed them.

• Use the verify-the-AI moment. The Exam-Prep Tutorial will make mistakes — they are built in as teaching moments. Catch them. That practice is exactly what the Final will test.

You have done the hard work across fifteen weeks. This week is about pulling the whole course together and showing it. Come to class ready to review out loud — and bring your questions. See you Tuesday.

---

(B) Welcome Announcement — Module 16

Release setting: post on the module's start day (offset = 0 days), i.e., Mon Dec 14, 2026 (the day the Final window opens) — not before. If your platform won't preserve the scheduled post date on import, post this as a draft labeled "Release: Mon Dec 14."

Subject: Week 16 — Finals week: the whole course, one last time

Hi everyone,

Here we are — the last week. This one is different from the rest: it's finals week. There is no quiz, no discussion, no assignment, and no AI Build Studio — the comprehensive Final takes their place. Everything this week is built to get you ready and then let you show what fifteen weeks built.

Here is the shape of it: our class session (Tue Dec 15) is a fast, complete review of the whole course — how AI works and why it is confidently wrong; the full prompting arc from sycophancy to simulations; the modality and tool landscape; the verification workflow; four weeks of Claude Cowork (projects, skills, connectors, artifacts, scheduled tasks, dispatch, computer use, Chrome, and Excel); the ethics and privacy framework; and the capstone integration. The Final is cumulative over Objectives 1–8; because the midterm already covered Objectives 1–4, the Final leans heaviest on the back half (Objectives 5–8) — Cowork, automation, ethics, and the capstone — but the earlier concepts are the ground the later ones rest on, so keep them sharp.

Your prep kit, in order: work the Study Guide first, then run the Exam-Prep Tutorial with an approved chatbot (ChatGPT, Claude, or Gemini) and submit the share link, then sit the Practice Final timed to find any soft spots.

The dates that matter:
1. Final — window opens Mon Dec 14, due six days later (25% of your grade; 25 items across all 8 objectives; AI not permitted).
2. Exam-Prep Tutorial — submit your chat share link before the Final closes.
3. Capstone — complete and document your end-to-end AI workflow and ethics reflection before the Final closes.
4. In-class review — Tue Dec 15; come with questions.

A word as we close the semester. When we started in Week 1, the whole promise was learning to use AI well — not to trust it blindly, and not to fear it, but to prompt it precisely, verify what it gives you, choose the right tool, build real automations, and apply good judgment about what to automate and what to keep human. Everything since has been that same instinct applied eight different ways: understand what the model is actually doing; prompt with structure and clarity; verify before you act; connect the right tools with the right permissions; protect what should be protected; and build something real. I have genuinely enjoyed watching you push back on AI over-promises, catch hallucinated citations, build working Cowork automations, and write ethics reflections that showed actual moral reasoning. That is the whole course. The Final is not about cramming — it is about naming the eight honest moves and showing you can do them. You can.

Open the Start Here / Module Overview page first — it lays out the whole week in order with every due date. Thank you for a tremendous semester.

Come with questions Tuesday,
Prof. Quinn

---