Course Challenge
Put all six modules together: a scenario-based knowledge check plus a hands-on capstone task where you take a real piece of work from blank page to verified, finished result.
How the Challenge Works
The challenge has two parts:
- Knowledge check — ten scenario questions covering every module. Aim for at least 8 out of 10. If you miss one, read the explanation and revisit that lesson.
- Capstone task — a hands-on exercise that uses the full workflow: collaborate, prompt with CRAFT, chain, verify, and use AI responsibly.
This course is free and self-paced, so there’s no grade to submit. The goal is simple: finish able to use AI confidently, effectively, and responsibly in your own work.
Part 1: Knowledge Check
1. (Module 1) A coworker gets a bland answer from their first AI prompt and says AI is useless. What's the best advice?
The collaborator mindset treats the first response as a draft. Better context and specific follow-ups are what turn a bland answer into a useful one.
2. (Module 1) You want AI's help to become a better presenter, not just to get one presentation done. Which request fits best?
Coaching mode — targeted feedback without a rewrite — builds your skill instead of replacing it.
3. (Module 2) Which is the safest way to get help summarizing a document containing employee salaries?
Salary data is confidential. Use approved tools with proper data agreements, or remove identifying details before sharing anything.
4. (Module 2) In a brainstorm, what should come first?
Diverge first (volume and variety), then converge with your criteria. Asking for 'the best' too early gets you one safe, generic answer.
5. (Module 3) Why can an AI produce a fluent, confident answer that is completely false?
Language models predict likely text. Plausible usually overlaps with true, but not always — that gap is where hallucinations come from.
6. (Module 3) For which task is AI most reliable?
Transforming material you supply is a core strength. Recalling specific quotes, recent data, or private information from memory is where AI is least reliable.
7. (Module 4) 'Write a newsletter intro, under 80 words, friendly tone.' Which CRAFT element is most clearly missing?
It has an Ask (write a newsletter intro) and a Format (under 80 words, friendly tone). But whose newsletter, for which audience, about what? Without Context the AI will guess.
8. (Module 4) The AI's draft is close but too formal and a bit long. What's the most efficient next step?
Specific feedback in the same conversation lets the AI revise its own draft precisely. That's the Tweak step in CRAFT.
9. (Module 5) You need a 12-page project proposal. Which approach will likely produce the best result?
Chaining gives you checkpoints to steer the work and catch errors early, instead of a shallow all-at-once result.
10. (Module 6) An AI agent has finished 'booking travel' and reports success. What should you do?
An agent's report of success isn't proof. Review the actions it actually took, especially anything involving money or commitments.
Part 2: Capstone Task
Choose one real deliverable you need in the next month — a report, a proposal, a training guide, a process document, a presentation, a newsletter. Then work through the full workflow:
- Collaborate (Module 1). Start by asking the AI to ask you 5 questions about the task before it does anything.
- Set up safely (Module 2). Before pasting anything, decide what’s safe to share. Remove or replace sensitive details.
- Know the limits (Module 3). List which parts of the deliverable depend on facts the AI can’t know or might get wrong. Plan to supply or verify those.
- Prompt with CRAFT (Module 4). Write a full Context–Role–Ask–Format prompt. Include “use placeholders instead of inventing facts.”
- Chain it (Module 5). Work in steps: outline → draft sections → critique from the audience’s point of view → revise. Create one visual if it helps. If the chat gets long, use a handoff summary.
- Take responsibility (Module 6). Check every fact and placeholder against real sources. Review for bias and tone. Decide how you’ll disclose AI assistance, if needed.
Self-assessment checklist. You’ve completed the capstone when you can say yes to all of these:
- The final piece is something I’d confidently put my name on.
- I verified every factual claim against a source outside the AI.
- I didn’t share anything I shouldn’t have.
- I can point to at least two places where my judgment changed the AI’s output for the better.
- I saved at least one reusable prompt template to my prompt library.
Congratulations — You’ve Finished AI Fundamentals
You can now:
- Work with AI as a collaborator, not a vending machine
- Set up and use a free AI assistant safely
- Explain how AI learns, where it fails, and where it shines
- Write effective prompts with CRAFT and a toolkit of techniques
- Chain prompts, carry context between chats, and generate visuals
- Use AI responsibly, handle bad responses, and supervise AI agents
Where to Go Next
- Keep practicing daily. Pick one recurring task each week and build a reusable prompt for it.
- Grow your prompt library and share the best entries with your team.
- Go deeper with other free courses in this catalog. AI & LLM Skills is a good next step if you want to see how language models work under the hood and start building with an AI API.