AI Ethics, Bias & Disclosure at Work
Learners who may recommend, use or supervise AI in customer, people, marketing or analytical work.
What you will be able to do
- Identify bias, privacy, intellectual-property, and accountability risks.
- Classify AI use cases by potential harm and review need.
- Draft plain-language disclosure and consent language.
- Design escalation rules for high-risk outputs.
How this is assessed
Team AI-use policy for a real companySubmit a concise policy covering approved uses, prohibited uses, disclosure, data handling and escalation. Write it for a real organisation — your employer, a family business, a society you run — and for a reader who is busy.
| Grade | What earns it |
|---|---|
| Pass | Complete coverage of uses, disclosure, data handling and escalation. |
| Merit | Applied to real cases from the organisation rather than stated in general terms. |
| Distinction | Practical governance a manager could adopt unchanged — clear, short, and enforceable. |
Before you submit
- Approved and prohibited uses are both listed, with examples.
- Data handling names the three buckets and gives a workable alternative for the restricted one.
- Disclosure language is plain and specific to a moment in the customer journey.
- Escalation names a role, a trigger and a timeframe.
- For distinction: a manager could adopt it unchanged on Monday.
Handing it in
Marked by an AI governance, privacy, risk, HR or legal-operations practitioner experienced in translating policy into day-to-day workplace controls. You will need an account to submit, so that the mark has somewhere to land. Sign in or create one. The modules are open either way.
Modules
4 modules, self-paced. Nothing is timed and nothing is scored.
- 1Responsibility mapLocate the people affected by an AI workflow.
- 2Bias labTest outputs across varied scenarios.
- 3Data boundariesSeparate permitted, sensitive, and restricted inputs.
- 4Disclosure designNotices, approvals, and accountability rules.
This course stacks
Uses AI competently, knows what it does to the people downstream, and checks its output before anybody acts on it.
A credential is issued when all of its artifacts have passed. It records the courses, the bands and the dates, nothing else, and no score.