AI Literacy Framework for eLearning Design
eLearning teams must integrate AI literacy to protect critical thinking, using frameworks like CLEAR to set clear expectations and redesign assessments, as
Artificial Intelligence is now a widespread part of the learning experience, used by students and employees for tasks ranging from summarizing readings to drafting documents. The Digital Education Council's Global AI Student Survey found 86% of surveyed students use AI in their studies, with more than half using it weekly, based on responses from over 3800 students across 16 countries. For eLearning teams, this raises a critical question: how can programs allow useful AI support without weakening critical thinking or academic integrity?
Banning AI is often unrealistic. A better approach is to build AI literacy directly into the learning experience.
AI Literacy Extends Beyond Tool Training
Many organizations treat AI training as a technical topic, focusing on prompts and features. That is useful but insufficient. True AI literacy means learners can evaluate AI output, question assumptions, identify weak evidence, protect privacy, understand bias, and decide when not to use AI. It is about developing better judgment, not just producing faster work.
UNESCO's guidance emphasizes a human-centered approach, including policy development and responsible use. For eLearning teams, this means AI should be connected to learning outcomes and assessment design, not added randomly. A learner who uses AI to generate an answer may finish faster, but if they cannot explain the reasoning or apply the concept, learning has not truly happened.
The CLEAR Framework for Design
A practical method to build AI literacy is the CLEAR framework: Clarify, Limit, Evaluate, Apply, and Reflect.
Clarify Expectations. Every course should include a short, visible AI use statement explaining what learners can and cannot do. For example, an online writing course might allow AI for outlining but not for final submission.
Limit AI Use Where Human Thinking Matters Most. AI use should match the learning objective. Course designers must ask: what part of this task must come from the learner's own thinking? If the outcome is "write a personal reflection," AI-generated writing may reduce the assignment's value.
Evaluate AI Output. Learners need practice checking AI output for unsupported claims, missing context, biased wording, weak evidence, incorrect facts, overgeneralized conclusions, and sources needing verification. This turns AI into a learning object for practicing judgment.
Apply Knowledge In Real Contexts. Assessments should require application in ways difficult to outsource. Instead of a generic essay, ask learners to analyze a realistic scenario, choose a response, and explain their reasoning.
Reflect On The Learning Process. Short reflection prompts encourage transparency and metacognition. They help educators understand how learners use AI in practice.
Redesigning Assessments and Policies
The real risk is often unclear expectations, not AI use itself. Learners need to know when AI is allowed, when it is not, how to disclose its use, which tasks require original thinking, and how to check output for accuracy. Without clear guidance, learners guess, which weakens the learning experience.
A strong AI policy should be written in simple language and connected to specific activities. For instance: AI may be used to brainstorm project topics but not to submit a final reflection without personal analysis; AI output must be checked against course sources before submission.
Generative AI is forcing a rethink of assessment. Frameworks like the AI Assessment Scale help define different levels of AI involvement, from no use to full AI-supported production with human evaluation. Good assessment design should include clear AI rules, scenario-based tasks, explanations of reasoning, draft checkpoints, peer discussion, source evaluation, personal reflection, and practical application.
As AI-generated writing becomes common, many institutions explore detection tools. These can support academic integrity workflows but should be used carefully as one signal in a broader review process, combined with assignment design, learner reflection, and instructor judgment. Detection should not be treated as the only proof of misconduct due to risks of false positives and negatives. The best approach is prevention through better learning design.
