A POSSIBLE SESSION FOR YOUR TEAM
Sample course outline — adapted after a planning conversation.
AI as an Employee
Design and test a supervised, repeatable AI responsibility using fictional records. Define when it can draft, when it must stop and who reviews its work. “Employee” describes delegation; a person remains responsible.
Proposed: two 120-minute working labs, each including a 10-minute break. Live deployment and production integration are outside this sample.
Proposed: a laptop and a team-approved text AI tool. Tool and access requirements agreed during planning; practice uses pasted fictional text. A shared text document or worksheet holds the manual run log; no scheduler, connector or autonomous execution is assumed.
This example is a starting point. Scope, timing, tools and deliverables are agreed together before a session.
WHO IT SERVES
Is this your next step?
Teams that already delegate and verify bounded AI tasks and want to explore an ongoing workflow.
Before the session
- Experience writing task instructions and checking AI outputs.
- Understanding of the sample workflow’s business rules and its human approval process.
- No coding required for the manual prototype shown here.
PRACTICE & TAKEAWAYS
Work you can try.
What you will practice
- Define a narrow AI role with permitted actions, stop conditions and a human owner.
- Run a manual prototype against normal, duplicate and exception cases.
- Record evidence for a go, revise or stop decision without treating a demo as deployment approval.
What you can leave with
- A role brief and manual supervised prototype.
- An evaluation case table and run log.
- A draft review, recovery and production-readiness question list.
PROPOSED AGENDA / 240 MINUTES
From first example to review.
Open each section for the exercise, output and review check. All practice uses fictional or anonymized material.
01Lab 1: Define the responsibility40 min
Purpose. Bound one repeat task before testing it.
Learning outcome. Write rules for drafting overdue-invoice reminders and escalating exceptions.
Explain or demonstrate. Contrast a draft-only role with authority to contact a customer.
Hands-on practice. Use fictional invoice-follow-up cases. Define approved input fields, human owner, drafting rules and stop conditions for missing payment data, disputes or duplicate requests.
Your output. A role brief and acceptance rules.
Review check. The role cannot send messages; missing data, disputes and duplicate drafts have explicit handling.
Prerequisites. Experience with checked AI task workflows.
Proposed time: 40 minutes, including the explanation, practice and review.
02Lab 1: Build a manual prototype80 min
Purpose. Make state and review visible.
Learning outcome. Simulate one repeatable run with a log and bounded review.
Explain or demonstrate. Walk through input → eligibility check → draft or escalation → human review → log. Include a 10-minute break after the demonstration.
Hands-on practice. Manually run one eligible and one disputed invoice through the instructions. Record invoice ID, input version, result and review state. Limit AI drafting to one attempt and one revision per case.
Your output. Two logged sample runs and a draft-only handoff.
Review check. Disputed case produces an escalation; eligible case produces an unsent draft linked to its source.
Prerequisites. Role brief from the first section.
Proposed time: 80 minutes, including the explanation, practice and review.
03Lab 2: Test failures and repeats70 min
Purpose. Check whether boundaries survive less tidy inputs.
Learning outcome. Detect missing information and avoid preparing a duplicate draft.
Explain or demonstrate. Show a repeated record with an existing draft log and a record missing payment status. Include a 10-minute break after the first two tests.
Hands-on practice. Run four cases: eligible overdue invoice, disputed invoice, missing payment status and invoice already logged with a pending draft. Compare each result with written expectations.
Your output. A four-case evaluation table with observed results.
Review check. Eligible case drafts; disputed and incomplete cases escalate; pending duplicate is held. Any mismatch is recorded.
Prerequisites. Saved role brief, prototype instructions and run log from Lab 1.
Proposed time: 70 minutes, including the explanation, practice and review.
04Lab 2: Review, recover and decide50 min
Purpose. Turn test evidence into a bounded next step.
Learning outcome. Revise a failed rule and document what must be checked before any live use.
Explain or demonstrate. Have a separate reviewer inspect source evidence and logs; explain that reviewer agreement does not prove correctness.
Hands-on practice. Review the four results. Tighten one ambiguous rule and rerun affected cases. If all rules pass, test and correct this seeded flawed rule: “Draft a reminder for every overdue invoice, even if disputed.” Rerun the disputed case against the corrected escalation rule. Write stop, recovery and human approval steps, plus a proposed cadence for later discussion.
Your output. Revised prototype, rerun evidence and a go, revise or stop recommendation for further supervised testing.
Review check. Recommendation cites test results; failure routes to a person, run limits are explicit and no live schedule or deployment is enabled.
Prerequisites. Four completed evaluation cases.
Proposed time: 50 minutes, including the explanation, practice and review.
TRY A REPRESENTATIVE EXERCISE
Route a disputed overdue invoice
Fictional practice material
Fictional case, as of May 15, 2026: INV-201, Elm Cafe, $420, due May 1, unpaid, disputed=true. Run log: no prior draft. Rule: disputed invoices must be escalated to the accounts owner; do not draft or send a reminder.
Example prompt
Apply this draft-only follow-up rule: disputed invoices must be escalated to the accounts owner, with no reminder drafted or sent. Case as of May 15, 2026: INV-201 Elm Cafe $420 due May 1 unpaid disputed=true; no prior draft. Return record ID, action, evidence and human next step. Do not infer a resolution or contact the customer.
Illustrative expected output
One acceptable example, not a guaranteed AI response.
Record: INV-201. Action: Escalate; no reminder drafted. Evidence: The supplied record is marked disputed=true. Human next step: Accounts owner reviews the dispute and confirms whether follow-up is appropriate. Nothing sent.
Check the result. The dispute rule overrides overdue status. The output escalates without writing a reminder, inventing a resolution or contacting anyone.
BEFORE WE MEET
What your team provides.
- Choose one recurring responsibility and name its human owner.
- Describe stop conditions, approval rules and the cost of a wrong action.
- Confirm allowed tools and data; bring anonymized exceptions or use supplied fictional cases.
- Identify who could evaluate permissions, logging and recovery before any later production work.
Use approved, anonymized examples or fictional material. Do not share confidential records or credentials in a planning inquiry.
ADAPT IT TO YOUR WORK
Choose a useful focus.
- Use a status-change monitor with a fixed supplied source instead of invoice follow-up.
- Add a separate draft reviewer while retaining one human decision owner and a fixed revision limit.
- Use an approved automation tool for a sandbox prototype only after features, permissions and time are confirmed.
We confirm the audience, business task, time and tool access during planning.