AV Labs Learning studioTeach AI Judgment
A practical lab for university trainers. Start with one everyday decision.
2Review the proposed reply
Prepared exampleTick every sentence that needs attention. Leave supported sentences unticked.
Choose a reply you could approve
Check the facts, the promise being made, and the private information.
Choose a corrected reply before approving.
A human decision stays in the loop.
The review record below changes as you work. Approval never sends a message.
- Scenario opened
What you are trying: a local, scripted teaching sample with prepared replies and fictional information. It makes no AI calls, stores no learner records, and sends nothing. Reloading clears your activity.
Now help someone else make the call.
Move from using an AI tool to teaching a repeatable review habit. Trainers practise, explain their decisions, and adapt the lesson to their own subject.
What a learner should be able to do
- Point to the source that supports each important claim.
- Spot a wrong date, an invented promise, and private information that should stay out of a reply.
- Write a prompt that says what information AI may use, what it must not reveal, and when it should ask a person.
- Explain why an approved answer needs another review when its source changes.
- Teach the same review habit using an example from their own department.
A 15-minute teach-back
Work in pairs. One person facilitates; one is the learner. Swap roles in a second round.
- Minutes 0–3: introduce the source. Ask, “What would we need to check before sending this?”
- Minutes 3–8: let the learner choose. Ask for evidence before showing the feedback.
- Minutes 8–12: compare corrected replies; change the source and discuss approval.
- Minutes 12–15: the learner explains the rule in their own words and suggests a new example.
A 90-minute lab you can facilitate
One session within a larger training programme
Make the decision visible
Compare a confident answer with its source. Ask what could go wrong.
Review together
Complete one scenario in pairs. Explain every flagged sentence.
Improve the prompt
Specify allowed sources, privacy boundaries, and an escalation rule.
Teach it back
Two 15-minute rounds. Adapt the task for a different department.
Test the handover
Change a source, recheck approval, and agree what still needs practice.
Ready to use with support
The learner explains their decision with source evidence, removes private details, avoids promises they cannot support, and knows when to stop for human review.
Revisit before independent use
The learner trusts the wording without checking, treats a draft as permission to act, or keeps an old approval after the facts change. Repeat the example and explain the missing step.
Practical AI training that people can carry forward.
AV Labs connects hands-on practice with clear teaching materials, staff confidence, and simple automation that keeps responsibility with people.
- Prompt and source literacy
- Responsible everyday use
- Trainer facilitation practice
- Review and approval habits
- Simple workflow design
- Independent staff capability