Learning objectives

Each objective states condition → behavior → criterion and uses an observable verb. Every objective appears in the alignment matrix and is assessed. Objectives were drafted with AI assistance and reviewed during the build; the structured review of objectives is documented in evaluation report.

Last reviewed: 2026-10-25

Workplace track (W)

Module 1 — What AI is (and isn’t)

  • W1.1 Given a short description of how an AI assistant produced an answer, the learner identifies the explanation as “predicting likely text” and corrects the misconception that the system “looked it up” or “knows facts”, identifying at least two of the three misconceptions listed in the module.
  • W1.2 Given eight work tasks from a logistics context, the learner classifies each as a good fit, a risky fit, or a poor fit for generative AI, with a correct classification on at least six of eight.

Module 2 — Checking outputs

  • W2.1 Given an AI-drafted customer reply that contains one factual error and one unverifiable claim, the learner identifies both and names the source needed to verify or correct each, correctly on both counts.
  • W2.2 Given an AI-drafted reply and the verification checklist, the learner applies the checklist and produces a corrected reply and a completed verification log, with all five checks addressed.

Module 3 — Data and privacy

  • W3.1 Given eight items of information that could be pasted into an AI tool (names, order numbers, anonymized totals, internal pricing, etc.), the learner classifies each as permitted / restricted / prohibited under the (fictional) Northwind data rule, correctly on at least seven of eight.
  • W3.2 Given the scenario “you already pasted customer data into an AI tool”, the learner selects the correct containment and reporting steps in the correct order, per the escalation path taught in the module.

Module 4 — Oversight and escalation

  • W4.1 Given four scenarios where AI suggests an action that affects a person (customer compensation, CV ranking, route planning affecting drivers, discipline), the learner decides for each whether a human must make the decision and identifies the correct escalation contact, correctly on at least three of four.
  • W4.2 Given a proposed AI use, the learner names at least two signals that mark it as high-risk (decisions about people, safety, legal effect; bias potential; scale), with at least two correct signals.

School track (S) — ages 14–16

Lesson 1 — How AI makes text

  • S1.1 After using the next-word interactive, the student describes in their own words how a chatbot produces text (it predicts a likely next word), correcting the misconception that it “knows” or “looks things up”.

Lesson 2 — When AI gets it wrong

  • S2.1 Given a chatbot answer to a history question, the student identifies at least two claims that need checking and verifies each with a reliable source, recording both on the fact-check worksheet.

Lesson 3 — Bias and fairness

  • S3.1 Given two AI outputs describing the same role or situation, the student identifies one way the outputs could be unfair to a group and explains how training data or output choices could cause it, with a cause that matches the example.

Lesson 4 — Using AI to learn, not to skip

  • S4.1 Given four homework scenarios, the student decides for each whether AI use would help learning or replace thinking, and writes two personal AI-use rules in the handout, with reasons for each rule.

Assessment coverage

  • Scenario quiz: 12 items → W1.1, W1.2, W2.1, W2.2, W3.1, W3.2, W4.1, W4.2 (items mapping)
  • Performance task (both tracks): W2.2, W3.1, W4.1 (workplace variant) / S2.1, S3.1 (school variant), scored with the rubric
  • Final project (school): S1.1–S4.1 applied together