Page 283 - CyberSecurityEssentialsEnglish

Choose a real subject you will explain, learn, or train on. It might be a concept in science, a
skill at work, a university unit, or professional training. Do not begin with the tool. Begin with
a clear learning problem, such as: “participants memorize the steps but cannot tell when to

use each one”, or “students write correct definitions but cannot analyze a case”.

Define one core outcome that can be observed. Use a verb such as explain, apply, compare,
justify, or design. Then write the evidence that would convince you the outcome was

achieved. It might be a two-minute explanation, a decision table, a solution to a new case, a
prototype, or a live performance. Do not let “completing the activity” be the only evidence.
Design a first attempt before AI intervenes. The learner might write a prediction, solve part

of the problem, draw a map, or choose a decision and justify it. This attempt reveals the
starting point and prevents the tool from hiding the gap. Set its time and size so that it is
feasible and does not become a full test.

Choose one or two roles for the tool. It might be an explainer offering an alternative

example, a coach giving hints, a simulator, or a feedback assistant working to a criterion.
Write down what it should not do, such as not giving the final solution, not awarding a grade,
and not using personal data.

Define the sequence of interaction. What does the learner see first? What is the question?

When does the tool enter? When does the teacher intervene? What is the stopping point?
What happens if the tool gives inaccurate information or the connection fails? Every activity

needs a main path and a simple alternative path.

Add a moment of checking after the support. Do not use the same example; offer a transfer
case or ask for an explanation in the learner’s own voice. Then add feedback tied to a
criterion and a short opportunity to revise. Finally ask for a specific reflection: which idea

changed? What did the tool provide? And what did you decide?
Write a clear rule for AI use. State what is permitted, what is not, and what must be
disclosed. If the tool is a required part of the activity, provide a means of access or an

equivalent alternative, and make sure no sensitive content is requested.

Review accessibility. Are the instructions clear? Is there a text version of the media? Can the
activity be carried out on a phone or without a subscription? Are the examples varied? Are
there suitable options for demonstrating learning? Do not add every possible alternative;

focus on the barriers most likely in your audience.

Define only three indicators for the first trial: the quality of the evidence of learning, the time
to carry it out, and the amount of rework or help required. Gather learners’ comments on

where it helped and where it confused, then change one element in each cycle so that you
know the effect of the change.

282