What does the Standard measure, and what does it not measure?
AIRS-J/1 measures how a person decides when working with AI. This page lists what it does not measure, with the Standard's own reason for each, and what a result does not authorise.
Judgment is how well a person decides once AI has given the answer. AIRS-J/1, the AIR APAC Standard for Judgment in AI-Enabled Work, sets out how that is assessed and certified. Its own sentence about this is short: "This Standard assesses how a person decides, not what they know." (Section 2)
This page sets out what the Standard measures and what it does not measure, with the Standard's own reason for each. It also sets out what a result does not authorise. It follows AIRS-J/1 v2.3, in force 19 October 2026.
What it measures
The Standard covers "individual human judgment when working with, alongside, or in supervision of AI systems in a professional or pre-professional setting" (Section 2).
What is measured is how the person behaves when deciding. The Standard calls this "observable decision behaviour" (Section 2.1).
The Standard describes that behaviour in nine dimensions in four domains: Structure, Test, Engage and Perform (Section 5). The dimensions page lists each one.
What it does not measure, and why
| What the Standard does not measure | The Standard's reason |
|---|---|
| How ready an organisation is for AI | Out of scope (Section 2). The AI Readiness Index rates organisations, not people, on public signals. An organisation's Index position does not depend on its employees' certifications, and a certification has no effect on the Index (Section 16.1). |
| Certification of vendors and products | Out of scope (Section 2). Products and providers are assessed against a separate standard (Section 16.2). "A vendor certification says nothing about the judgment of any individual using the product." (Section 16.2) |
| Skill with any AI tool, model or vendor | "Prompt craft, model selection and tool fluency are not assessed." (Section 2.1) |
| Knowledge of a professional field | "Scenarios supply the domain content the candidate needs." (Section 2.1) |
| Intelligence or personality | "The construct is observable decision behaviour, not a trait." (Section 2.1) |
| Occupational competence, comprehensive AI literacy, management of people, or performance in an unbounded workplace | The Standard certifies "the exercise of judgment in AI-enabled work using domain content supplied within an approved assessment". It does not certify these things (Section 2.2). |
Training aligned to the Standard may teach how to brief an AI assistant and question it. "The quality of the prompt itself is not scored." (Section 2.1)
What "managerial" means here
Where the Standard uses the word "managerial", it has a narrow meaning. It "describes management of decisions, consequences and AI outputs" (Section 2.2).
It is "a construct label, not a claim that the holder occupies, or is qualified to occupy, a people-management role" (Section 2.2). A result makes no claim about whether a person can lead a team. The plain-language description on every credential and register entry must carry this scope (Section 2.2).
What a result does not authorise
A credential is not "a substitute for supervision, professional licensing or regulatory authorisation" (Section 2.1). The Standard gives the reason: "Nothing in this credential authorises a holder to act beyond their existing professional or legal authority." (Section 2.1)
A certified person therefore keeps the supervision, licensing and authority that already apply to them.
An employment-grade result is an Assessment result at Moderate or High confidence (Section 12.8). Where an employer uses one, the Standard calls it "AIR APAC's expert reading of the person's judgment" (Section 2.1). It also states: "The employment decision is the employer's." (Section 2.1, citing Section 12.6)
Practice, a training session and a result with too little evidence (Provisional or Insufficient confidence, Section 10) are development readings. They decide nothing about employment (Section 12.8).
Section 2.1 adds: "A holder or organisation that states a claim beyond its basis in this Standard is acting outside it."
Why attendance and self-reported confidence are not the measure
Section 1.1 starts from a change in work: "AI systems now produce the answer. They do not carry the consequence of acting on it."
A person still has to decide whether the work is fit to act on and who is entitled to be told. The person also decides what may not be delegated, and what happens when the work turns out to be wrong (Section 1.1).
Section 1.1 says the usual records do not measure this contribution: "Training attendance, tool proficiency certificates and self-reported confidence do not measure it."
For that reason the Standard looks at what a person does when deciding. It does not look at what they attended. An assessor may not score on inferred confidence (Section 7.3).
What this does not claim
- This page does not claim that AIRS-J/1 measures everything that matters in a person's work. Section 2.2 lists what it does not certify.
- It does not describe facilitator certification, which Section 20.6 covers.
- It does not claim that a result settles whether a person should hold a job. The employment decision is the employer's (Section 12.6).
- The Standard sets out the basis each claim about a result needs (Section 2.1). Every reading is scored by published rules and traces to what the person did. The validation report is published as the checks in the Standard are met (Section 11.7).
- The quotations come from AIRS-J/1 v2.3. Where this page and the full text differ, the full text governs.
AIR APAC publishes the standard for judgment in AI-enabled work, certifies people against it through its official instrument, and stands behind every result.