AI Foundations and Literacy — Checkpoints
Read the checkpoint before starting a level. It defines what proficiency will require and prevents the modules from feeling like unrelated content consumption.
Verified AI-Assisted Task
Complete a bounded task and submit the goal, context, interaction evidence, finished result, verification, risks, human decisions, and a short explanation of the AI system involved. A **bounded task** is one with a clear start, a defined deliverable, and a scope you can complete in one focused session — for example, drafting a status update from approved notes, not deciding a project's funding.
Proficiency requires: accurate mental model, suitable interaction, traceable verification, responsible data use, and honest limitations. Self-assess against the rubric before submitting.
Comparative AI Use-Case Project
Frame a real use case, define success, compare at least two approaches including a non-AI option, test representative cases, map human responsibility, and recommend reuse, modification, pilot, or rejection.
Proficiency requires: independent framing, controlled comparison, evaluation evidence, proportionate safeguards, and documented reasoning.
Emerging Capability Recommendation
Investigate an unfamiliar capability or proposed system. Produce a plain-language explanation, evidence assessment, conceptual architecture, stakeholder and risk map, evaluation plan, oversight model, and adopt/pilot/monitor/disregard recommendation.
Proficiency requires: systems thinking, failure-aware evaluation, governance, source judgment, uncertainty, and conditions that could reverse the decision.