What if my connection or device fails during a sitting?
What a candidate does when a connection or device fails during a certifying sitting, what the record keeps, how to report it, and what can be asked for before the sitting.
A connection can drop, and a device can stop, during a certifying sitting. This page sets out what a candidate does at that moment and afterwards. It also says what the record of the sitting keeps, and what a candidate can ask for before the sitting. Each answer names the section of AIRS-J/1 it rests on. Where the Standard sets no step, the answer says so.
What should I do at the moment it fails?
A candidate tells the person running the sitting at once. The candidate notes the time and what the screen showed when it failed, while it is still clear in memory. The candidate then follows the instructions of the person running the sitting. If nobody is running the sitting, the candidate goes straight to the steps below. The Standard sets no fixed procedure for the moment of a failure, so a note made at the time is useful when the candidate reports it.
May I switch to another device or another AI tool to keep going?
Do not open any AI tool outside the environment the sitting provides. Section 9.2 prohibits the use of AI tools outside the provided environment during a certifying session. Section 18.4 makes it a ground for suspending or revoking a credential. The Standard sets no rule on moving to another device, so the candidate asks the person running the sitting and waits for the answer. The AI inside the sitting is there to be used; the rule is about tools outside it.
Does the record show what happened?
The record keeps what the candidate decided and when. Section 8.7 says transcripts retain "the full decision trace": decisions, sources consulted, stated reasons, revisions and timings. Section 9.6 adds that every transcript records the scenario, rubric and instrument versions and the scaffolding level. That is a version record, not a record of failures. The Standard does not say how a failure itself is recorded, so the candidate's own note of the time and the screen matters.
How do I report it after the sitting?
The Standard sets no step for reporting a failure. The route AIR APAC offers is the contact page: the candidate writes as soon as possible and chooses the topic "Help with my assessment". The report gives the date, the time of the failure, what the screen showed and what the candidate was told to do. Once the result is issued, a material technical failure is a ground for appeal (Section 18.6), and an appeal is lodged within 30 days of the decision (Section 18.5). The steps are on the appeals page, and appeal is free for the first instance.
What does a review look for?
Section 18.5 says the review looks for "a wrongly applied rule or anchor, evidence the instrument extracted wrongly, a technical failure, or a procedural failure". The outcome and the reasons are given in writing (Section 18.5). This page does not say what a failure means for any particular result.
My sitting is for a job application. Does anything change?
Section 9.1 says an Assessment commissioned for a hiring decision is sat in person under supervision. Remote sittings for hiring open only when the approved instrument records the session (screen and camera) for its whole length. An appeal on a result released for hiring is decided within ten working days (Section 18.5). If the sitting mode that Section 9.1 requires was not met, that is also a ground for appeal (Section 18.6). The applicant receives the result no later than the employer receives its copy (Sections 12.6 and 12.9).
I am an employer, and an applicant reports a failure. What applies?
The applicant may appeal the result on the ground of a material technical failure (Section 18.6). The applicant may lodge the appeal within five working days of release. If so, the commissioning organisation does not communicate an adverse decision resting on the result until the appeal is decided (Sections 18.5 and 12.11). The result is one input, and the hiring decision is the employer's (Sections 12.6 and 12.9). The applicant can write to AIR APAC through the contact page.
I know my connection or device may cause a problem. Can I ask for support before the sitting?
Yes. Section 3 defines an access support as an approved change to "presentation, timing or response modality" that removes a barrier unrelated to what is measured. The Standard does not say whether a connection or device problem qualifies; AIR APAC decides each request in writing, with reasons (Section 8.9). A candidate asks at least 10 working days before the sitting; late requests are still considered, and the steps are under "How to ask" on this page. When the sitting is arranged, the candidate can also ask the person running it what to do if the connection or device fails.
Does AIR APAC keep track of technical failures?
Yes. Section 15.6 says AIR APAC monitors "access-support requests, decisions, use, technical failures and outcomes". This information is kept apart from scoring and is reported only in aggregate. A difference linked to an accommodation, an instrument or a delivery process is investigated (Sections 4.8 and 15.4).
Will an employer be told that I used an access support?
No. Under Section 12.6, "the employer is never told whether an access support was used". This holds for employees and for applicants alike. Applicants are told before they start how to ask for an access support (Section 12.9).
What this does not claim
- It does not say that a sitting resumes, restarts or is sat again after a failure, or that time is added. The Standard does not say so.
- It does not say that a failure never affects a result, or that a failure always voids one.
- It does not say that every connection or device failure is a material technical failure. The Standard does not define the term.
- It does not say that a weak connection or an old device qualifies for an access support. AIR APAC decides each request.
- It does not set a device or connection requirement. None is published.
- It does not give a time within which AIR APAC answers a report of a failure. None is published.
Judgment is how well a person decides once AI has given the answer. AIR APAC publishes the standard for judgment in AI-enabled work, certifies people against it through its official instrument, and stands behind every result.