FAST-01 Lab Guide: Applied AI Literacy at Work

Provider: 7formations | Course: FAST-01 | Instructor: Wadood Chaudhary | Three labs (about 45 minutes each), one per session, plus the final applied task Audience: trainees (steps, worksheets) and instructors (setup, answer notes, timing). Instructor-only notes are marked Instructor.

All companies, people, IDs, prices, and messages in this guide are fictional. Labs run in the employer's approved AI tool (Microsoft 365 Copilot Chat, ChatGPT Business or Enterprise, Gemini for Google Workspace, or Claude for Work), in the 7formations sandbox, or on paper. Nothing in the labs connects to an employer's systems or uses real customer, patient, or employee data.

0. Lab environment and safety rules (read before Lab 1)

Tools

• Use only the AI tool your employer has approved, signed in with your work account. If no tool is approved, use the 7formations sandbox account issued for the course (fictional data only, spending-capped, disabled at the end of the cohort) or the paper versions of each worksheet.
• The instructor demonstrates in one tool only: the employer's choice. Where steps differ by tool, the guide gives a short note for each. Menus change often; the instructor confirms the current screens before each cohort.
• Paper fallback: every lab can be completed with printed AI outputs supplied in the course folder. What is scored is your judgment, not the tool.

Data

• Fictional data only. Massachusetts G.L. c. 93H and 201 CMR 17.00 treat a resident's name combined with a Social Security number, driver's license or state ID number, or financial account or card number as personal information. Never paste that kind of combination into any AI tool.
• No protected health information (PHI), no confidential business data (pricing, bids, unreleased financials, HR matters, source code, passwords).
• When you relate a lab to your own job, describe the task in general terms. "I answer customer questions about delivery dates" is fine. A real customer email is not.
• Shared-printer test: if you would not leave it on the office printer, do not paste it.
• If you paste something you shouldn't have, stop, tell the instructor, and do not "ask the AI to forget." Class practice mirrors what you'd do at work: report it.

Fictional organizations in this course

• Cape Harbor Inn: a 40-room inn in Chatham, MA (hospitality).
• Pioneer Valley Mutual Insurance Agency: an independent agency in Springfield, MA (financial services).
• Bayside Precision Parts: a 60-person CNC machine shop in Worcester County, MA (advanced manufacturing).
• Charles River Family Dental: a two-location dental practice (healthcare, used only with placeholders, never PHI).

Completion

• Each worksheet is scored Not yet / Meets / Strong against the criteria at the end of its lab.
• A "Not yet" worksheet is returned with comments and can be resubmitted once within the course window.

Lab 1 (Session 1): Personal AI Use Map, draft 1

Scenario. You are starting from your own job. Before anyone talks about tools, you need a clear picture of which tasks AI could help with, which need care, and which should stay with people. To calibrate, you first rate a fictional Cape Harbor Inn front-desk job, then your own.

Goal. Produce draft 1 of W1 Personal AI Use Map: 8 to 10 of your own weekly tasks, each rated good fit, use with care, or no AI, with a reason tied to risk, data, and checkability.

Data. Calibration card: "A week at the Cape Harbor Inn front desk" (fictional).

Steps

1. Calibrate (10 min). In pairs, rate the eight front-desk tasks: good fit, use with care, or no AI. Write one reason per rating. Use three questions: What data does it need? Can I check the result? Who is harmed if it's wrong?
2. Compare (5 min). The instructor reveals the reference ratings. Discuss any task where your pair disagreed.
3. List your own tasks (10 min). Write 8 to 10 tasks you do in a normal week, in general terms. Add rough hours per week.
4. Rate your tasks (10 min). Rate each task and write a reason. For every "use with care" task, write the check you would run (what you compare the output with).
5. Partner swap (5 min). Swap maps. Your partner challenges one rating. Change it or defend it in one sentence.
6. Try one (5 min). In the approved tool or sandbox, run one "good fit" task using fictional or general text only. Note one thing the output got right and one thing you would check.

Tool notes. Copilot Chat: use the work (Entra ID) sign-in so enterprise data protection applies; look for the shield or "Protected" indicator. ChatGPT Business/Enterprise: confirm you are in the workspace, not a personal account. Gemini: use the Workspace account, not a personal Google account. Claude for Work: use the Team or Enterprise organization.

Deliverable: W1 Personal AI Use Map (draft 1)

Completion criteria. At least 8 tasks listed; every task rated with a reason that mentions data, checkability, or harm (not just "saves time"); every "use with care" task has a named check; at least one rating changed or defended after partner swap. Meets for the course needs 5+ tasks with a correct, reasoned fit.

Lab 2 (Session 2): Data sorting game and a checked draft

Scenario. Pioneer Valley Mutual Insurance Agency (fictional, Springfield) is piloting an AI assistant for customer service. Before staff get access, every person must show they can tell safe data from restricted data and can catch errors in AI drafts.

Goal. Score 100% on the "never" cards in the data sort (W3), then apply the four-step check to three AI drafts and record keep, fix, or discard (W2).

Data. Part A uses the 20 data cards in the activity set (activities.json, F01-S2-A1) or the printed deck. Part B uses this fictional handbook excerpt and three AI drafts.

PVM Homeowners Handbook (fictional), section 4.2
- Water backup of sewers and drains is excluded unless endorsement WB-1 is purchased.
- WB-1 limit: $5,000 per occurrence; deductible $500.
- Report claims within 30 days of discovery by phone (413-555-0142) or the client portal.
- Temporary housing (Coverage D) pays up to 12 months when the home is unlivable after a covered loss.

Steps

1. Sort (15 min). In teams of three, sort the 20 cards into OK, remove first, and never. Score against the key. Any miss on a "never" card means re-sort the whole "never" pile until it is 100%.
2. Explain two (5 min). Each team explains one "remove first" card: what would you remove, and does the task still work afterwards?
3. Check three drafts (15 min). Individually, apply the four steps (source, facts, numbers, tone) to D1, D2, and D3. Record what you found at each step and your decision.
4. Rerun one (5 min). In the approved tool or sandbox, paste the excerpt, ask for a 3-sentence customer reply about sewer backup, and add: "Use only the excerpt. If something isn't there, say 'not in source.'" Check the output with the four steps.
5. Debrief (5 min). Which step caught the hardest error?

Deliverable: W2 Four-Step Check log and W3 Data sorting scorecard

Completion criteria. 100% correct on the "never" cards (retake until met); all four steps recorded for each draft; decisions match the reasons; the rerun output checked and recorded.

Lab 3 (Session 3): Final applied task, finished AI Use Map, and workstation checklist

Scenario. You now apply the course to your own work. You pick three task types from your AI Use Map, build and test a prompt for each on a practice version of the task, and finish with the final applied task and your workstation checklist.

Goal. Complete W4 Three tested prompts, the final applied task (prompt + four-step check + decision), the finished W1, and W5 Workstation data-safety checklist.

Data. Your own tasks described in general terms, plus a practice source you build: a public document, an approved template, or a fictional stand-in (for example, the Cape Harbor Inn FAQ excerpt below if your job has no safe source).

Cape Harbor Inn FAQ (fictional excerpt)
Check-in 3:00 PM; check-out 11:00 AM. Early check-in from 1:00 PM is $25 when available.
Shuttle to the Chatham ferry dock: 8:15 and 10:15 AM daily.
Pets: two dogs up to 40 lb in garden rooms only; $35 per night. No cats.
Cancellation: full refund up to 7 days before arrival; within 7 days, the first night is charged.

Steps

1. Pick three tasks (5 min). One good-fit and two use-with-care tasks from your map.
2. Build three prompts (20 min). For each: who you are and who reads it; the task; the safe source text pasted between markers; the format and length; what to do when something is missing. Run each in the approved tool.
3. Peer review (10 min). Swap one prompt and output. Your partner runs the four-step check and records keep, fix, or discard. Fix the prompt, rerun, and compare.
4. Final applied task (10 min). Choose your strongest prompt. Run it fresh, apply the four-step check, and write your decision and reason. This is scored.
5. Checklist (5 min). Complete W5 using the template below. Share one item with the group.

Deliverable: W4 Three tested prompts

Deliverable: W5 Workstation data-safety checklist (template)

• My approved tool(s) and how I sign in:
• Never goes in (my role's examples):
• Remove first (what I strip, and my placeholders):
• My four-step check before anything leaves my desk: source, facts, numbers, tone
• Who reviews high-stakes output (customer, money, safety):
• If something goes wrong, I tell: ______ (same day)

Completion criteria. Three prompts run on practice tasks with the output checked; at least one prompt improved after peer review; final applied task shows all four steps and a reasoned decision; W1 has 5+ correctly rated tasks; W5 is complete and specific to the trainee's role.

Appendix B: Rubric (Not yet / Meets / Strong)