A digital inspection can return a precise number and still support the wrong decision. If a part is checked against a drawing revision that has already been superseded, the instrument records the dimension faithfully, the software produces a clean report, and the error travels onward quickly. Writing in Metrology News, behavioural scientist Gleb Tsipursky argues that connected, AI-assisted inspection makes this failure mode harder to notice, and that manufacturers should attach a "version passport" to every measurement that carries consequences.
His starting point is the technical promise of persistent identification. A Metrology News account published on 26 October described how Quality Information Framework data can keep the identity of a measured characteristic intact across product definitions, inspection plans and results. Traceability, in other words, is achievable. The management question is separate: whether each decision is actually made against the correct version of the requirement.
Correct arithmetic, wrong reference
The failure does not announce itself. AI tools can summarise an inspection report, highlight an apparent deviation or rank the cases that need review, but a model that receives results without version context may compare an observation against the wrong specification. Nothing in the arithmetic looks wrong, yet the conclusion does not hold.
What the passport would carry
Tsipursky is explicit that the passport need not be a new software product. It can begin as a bundle of fields travelling with the measurement:
- the approved drawing or model revision;
- the characteristic identifier;
- the inspection-plan revision;
- the instrument or method used;
- the result timestamp;
- the person or system that approved the disposition.
Why industrial AI raises the stakes
The National Institute of Standards and Technology's 2026 smart-manufacturing roadmap, as cited in the article, lists the integration of heterogeneous data and trustworthy operation among the continuing challenges for industrial AI. A version passport addresses one narrow slice of that problem: it makes the origin of a measurement and the conditions that apply to it visible to whoever acts on it next.
A pilot that fits before platform work
Quality teams can test the idea without integrating a new platform. Take a small set of completed inspection records — one genuine out-of-tolerance result, one conforming part, and one case where the revisions do not match. Ask an AI assistant to summarise the evidence, name the revision that applies and flag any missing provenance, then compare its answers with an experienced quality engineer's judgement. The scorecard matters more than a good-or-bad tally: whether the assistant catches a revision mismatch, whether it invents a tolerance that was never stated, and whether it can point to the approved source behind a conclusion. Those are precisely the errors that digital handoffs tend to conceal.
Ownership settled before the alert
Where records disagree, responsibility should already be assigned. A quality engineer might rule on whether a measurement is valid, design engineering on which definition governs, and manufacturing on whether production can continue. The escalation route has to be agreed before an alert appears, rather than negotiated during a shipment dispute.
Agreement is not the same as correctness
The argument departs from the familiar call for more comparison between AI and human inspectors. If both were handed an obsolete specification, perfect agreement would only reinforce the wrong decision with greater confidence. The reference itself needs scrutiny. Automation lets measurement data travel further and faster; a reliable chain of custody is what keeps drawing, feature, method and result together at the moment the decision is made.