For the officer who built it
You want to know it will not embarrass the shift
The test is five minutes and needs no equipment: three questions you can mark yourself, two it must refuse, and the gap question.
Build an AI companion · session 4 of 4
Five minutes of testing before the shift relies on it
Test the answers you can mark yourself, then test the questions it must refuse. Whatever passes both is usable - within its one job.
Six decisions on testing a companion: a question you can mark yourself, the admitted-gap answer, the access question it must refuse, a colleague's independent test, recording what passed, and when to stop using it.
Corporate pricing on request. Course funding application status: pending.
Session4 of 4 in Build an AI companion
Decisions6, one correct answer each
Marker1 printed point on the desk board
Recordthe five fields, same as every run
The scenario
Proving a companion behaves before the shift relies on it
Nobody hands a new officer the gate keys on the first morning. They are watched, asked questions with known answers, and given the post when they have shown they can hold it.
A companion should earn the same. The test takes five minutes and needs no equipment: three questions you can mark yourself, two it must refuse, one gap it must admit.
Write the results down. That sheet is what turns a clever app into something a supervisor can approve and a shift can rely on.
The six decisions are the ones that actually go wrong on a quiet shift. They are, in the order the run puts them:
- A question whose answer you already know
- The gap answer: does it admit what it does not know?
- The access question it must refuse
- A second test by someone else
- Writing down what passed, and when
- The failure that means stop using it
What changes, and for whom
Three people, three different changes
For the supervisor
Approval needs evidence, not enthusiasm
This session gives the supervisor a repeatable check with a written result, so approval is based on behaviour that was seen.
For the agency
Anything on shift must be verifiable
A companion that passed a recorded test, inside a written job and a rule sheet, is a controlled tool rather than an officer's private experiment.
These are descriptions of the work the training addresses. They are not client results and not a promised saving.
What the officer does, in order
One run, start to record
About ten minutes at the post, on the officer's own phone, with a supervisor on site.
Scan the marker at the desk board
One printed marker where the test sheet lives, beside the rule sheet.
Run the five-minute test
Three markable questions, two refusals and one admitted gap. Every answer written down.
Check the refusals
The overlay marks the questions a companion must hand back: access, identity, and anything confidential.
Write the record
One row, the same five fields as every run in the department.
The trainee list
Every session writes the same record, so four sessions make one page of that officer's training history, and the same five fields are asked of the newest officer and the most senior one - which is the part of it that is fair. The run is written to the agency's list in the five fields the checklist reads.
This page is the session. The run itself is on the player: open it here - the practice run needs no camera. It is session 4 of 4 in Build an AI companion.
A completed session produces one row, in these five fields
This is post training, so there is no assessor to satisfy - but the record matters anyway: it is how an agency shows what its officers were trained on, in what order, and when.
| Field | What it carries |
|---|---|
| Name | As written in the agency's own staff record. Nothing else is stored about the officer. |
| NRIC / FIN | The last four characters only. The full number is never entered and never leaves the phone. |
| Training provider | Whichever name the checklist route puts here: the agency's own name when the agency delivers the training in-house, or the service provider's name when a provider delivers it - in that case the checklist also asks the agency to submit the provider's invoice showing simulation training was used. The field is left blank on purpose: it is typed, not assumed. |
| Course title | States the technology: the training programme or session plus Augmented Reality (AR) Simulation Training. |
| Date of training | The date the run was completed, written from the device at the time of the run. |
The record lives on the officer's own phone until it is exported or printed, so the habit is: run it, then save the list. The house standard is 80 per cent, which is a fair bar for six decisions, and a failed run is a re-run rather than a mark against an officer.
What this package does not claim
The limits, stated on the page rather than found later
Every course here is honest about what it is and what it is not, because a package that oversells itself is the one that lets an agency down in front of an assessor. Read these before you book anything.
- This session does not certify any tool or product, and no test here substitutes for the agency's own approval process.
- It is not a SACE programme and does not satisfy a licence-renewal requirement.
- It does not cover every behaviour a companion can show. It covers the ones that matter at a post: accuracy on a known answer, an admitted gap, and a proper refusal.
- It is not a substitute for a post order. An approved companion sits inside the orders, not above them.
Questions about this session
Straight answers
How often should it be tested?
After any change to the rules or the information, and at each handover where a shift will rely on it. Five minutes, written down.
What if it fails a test?
Fix the rule, then test again - or stop using it for that task. An unruled companion is a risk, not a helper.
Who approves it for the post?
The supervisor, in writing, on the strength of the recorded test. That is what makes it part of the post rather than one officer's app.
Is this a SACE programme?
No. It is post training, and it does not satisfy a licence-renewal requirement.
Build an AI companion · session 4 of 4
Run it on your site before your officers do
Fix one marker at the real point, open the session on a phone, and see whether the overlay holds where your officers actually stand. That is a fair test, and it takes three minutes.
Corporate pricing on request. Course funding application status: pending.