Interactive compliance & transparency lab
Build your integrity policy or AI disclosure log
Switch between teacher syllabus/assignment policy generation and student submission AI disclosure records.
Live Preview
Quick answer
Define, disclose, document, and defend
A fair simulation policy tells students which assistance is allowed for each stage, requires a short record of permitted support, preserves the original run evidence and revisions, and asks every student to explain part of the work. The goal is not to ban useful support. It is to keep the learning evidence visible and the student's decisions genuinely theirs.
1. Define
Label assistance as allowed, allowed with disclosure, or not allowed for this task.
2. Disclose
Record who or what helped, the purpose, and what was accepted or rejected.
3. Document
Keep settings, raw results, calculations, notes, drafts, and revision reasons.
4. Defend
Explain a decision, calculation, tradeoff, limitation, and next test.
Verifiable simulation integrity
1-Click scenario codes for verifiable baseline data integrity
To prevent AI data hallucination and ensure student evidence trails originate from legitimate runs, assign 1-click classroom scenario launch codes across all 18 business simulators. Standardized initial parameters allow instructors to verify student CSV exports and calculations against expected baseline ranges during Stage 2 & 3 audits.
Choose an assistance rule for each stage
Do not rely on a single sentence such as “AI is allowed.” Different stages create different learning evidence. Copy this matrix into the assignment and change any row to match school policy, student age, access, and the intended outcome.
| Stage | Allowed freely | Allowed with disclosure | Not allowed |
|---|---|---|---|
| 1. Ideation and question | Brainstorm business names, customer segments, or scenario ideas | Summarize articles or generate candidate hypotheses | Letting a tool pick the decision without student rationale |
| 2. Run and data collection | Using built-in simulation calculators or spreadsheet formulas | Converting raw numbers to a table or cleaning CSV format | Fabricating simulation results or using unrecorded runs |
| 3. Analysis and calculation | Using verified calculator tools to check manual arithmetic | Drafting alternative explanations of an observed pattern | Claiming an AI summary without checking against raw data |
| 4. Brief and presentation | Spell check, readability checks, accessibility formatting | Polishing student-written drafts or translating directions | Submitting an unedited AI-generated report or presentation |
| 5. Defense and debrief | Practicing questions with a peer, coach, or chatbot | Reviewing notes or transcripts during debrief preparation | Reading scripted AI text during live oral verification |
Printable student handout
Simulation assistance and AI disclosure log
Simulator and investigated decision:
| Tool or helper | Stage / purpose | Prompt or question used | What was accepted / rejected | How verified |
|---|---|---|---|---|
Verification statement: I confirm that the simulation results reflect actual runs, the calculations were verified against raw data, all material outside assistance is listed above, and I can explain each decision and tradeoff in my own words.
4-question oral verification check
When you want to verify independent student understanding quickly, ask each student to answer these four questions using their own evidence portfolio or decision brief. This routine takes 2–3 minutes per student and separates genuine learning from copied text.
- Decision: What did your team change, and what stayed constant?
- Evidence: Show one calculation or result that influenced the recommendation.
- Tradeoff: Which metric, stakeholder, or responsible constraint prevents a simple “maximize profit” answer?
- Limit and next step: What can the model not prove, and what fair test should come next?
Fairness check: assess accuracy, evidence use, and capacity to revise. Do not assess speed, eye contact, accent, confidence, handwriting, or disability-related communication differences. Provide wait time and alternate response modes when needed.
12-point transparent-work rubric
| Criterion | 3 — clear | 2 — partial | 1 — limited | 0 — missing |
|---|---|---|---|---|
| Evidence trail | Settings, results, formulas, and revisions are traceable. | Most evidence is traceable; one gap remains. | Only final values or fragments remain. | No usable evidence trail. |
| Disclosure | All material help is specific, complete, and connected to changes. | Help is named but purpose or changes are incomplete. | Disclosure is vague or added only after prompting. | Required disclosure is absent. |
| Verification | Student accurately explains and checks decisions, math, and claims. | Explanation is mostly accurate with one supported correction. | Explanation depends on prompts and cannot connect key evidence. | No evidence of understanding yet. |
| Judgment | Recommendation addresses tradeoffs, limits, guardrails, and a next test. | Recommendation addresses most elements. | Recommendation overclaims or misses key constraints. | No evidence-bounded recommendation. |
If evidence is missing or explanations conflict, pause the grade and request clarification under the school's normal procedure. A gap is a reason to investigate fairly, not automatic proof of misconduct.
Respond to concerns without guessing
- Return to the policy students received before the task.
- Describe the specific evidence gap or inconsistency without accusation.
- Review run records, calculations, drafts, revision history, sources, and disclosure.
- Invite the student to explain the work using the same oral-check criteria.
- Consider accessibility, language support, collaboration rules, and permitted tools.
- Follow school procedures and document the decision. Do not treat an automated detector as conclusive proof.
This classroom guide is not legal advice and does not replace school, district, institution, labor, privacy, accessibility, or academic-conduct policies. Use authoritative local policy when requirements differ.
Academic integrity and AI FAQ
Can students use AI for a business simulation assignment?
That depends on the published task policy. Distinguish support such as brainstorming or editing from evidence interpretation and final reasoning students must complete and explain. Require disclosure for any permitted material assistance.
How should students disclose AI assistance?
Record the tool or person, date, purpose, relevant prompt or request, what was used or rejected, and how the output was checked. The printable log above provides those fields.
How can a teacher verify individual understanding in team work?
Ask each student to explain one decision, calculation, tradeoff, and limitation using the team's actual evidence. Assess the reasoning rather than confidence, speed, or speaking style.
What information should never be entered into an AI tool?
Do not enter student names, grades, contact details, credentials, private business information, health or disability information, or other personal data. Use fictional or de-identified simulation evidence only.
Does an AI detector prove academic misconduct?
No automated indicator should be treated as conclusive proof. Review the policy, evidence trail, disclosure, student explanation, and school procedures before deciding.
Build a transparent assignment sequence
Publish the rules in the assignment template, preserve settings and results in the evidence portfolio, document individual work with the team contribution record, rehearse claims through the peer review protocol, and use the assessment and feedback toolkit for revision and grading.