A measurement can be taken correctly and still drive the wrong decision. When a part is checked against an obsolete drawing revision, the instrument records the dimension faithfully, the software produces a polished report, and an incorrect conclusion reaches everyone with impressive speed. In a Metrology News column, behavioral scientist Gleb Tsipursky argues that manufacturers adopting AI-assisted inspection should give every consequential measurement a "version passport."
He frames the gap as a side effect of connected metrology. A Metrology News account of persistent identification describes how Quality Information Framework data can preserve the identity of a measured characteristic across product definitions, inspection plans and results. The technical promise is traceability; the management challenge, as Tsipursky describes it, is ensuring that each decision follows the correct version of the requirement.
What the passport would hold
The passport does not have to be a new software product, he writes. It can begin as a set of fields that travel with a measurement:
- the approved drawing or model revision;
- the characteristic identifier;
- the inspection-plan revision;
- the instrument or method used;
- the result timestamp;
- the identity of the person or system that approved the disposition.
Why AI sharpens the problem
AI can summarize an inspection report, point to an apparent deviation, or recommend which cases need review. But a tool that receives results without version context may compare an observation against the wrong specification. The arithmetic can be correct while the conclusion is invalid. Tsipursky points to the National Institute of Standards and Technology's 2026 smart-manufacturing roadmap, which lists integration of heterogeneous data and trustworthy operation among the continuing challenges for industrial AI. A version passport targets a specific slice of that problem by making the origin and the applicable conditions of a measurement visible to the next decision-maker.
A small trial before platform integration
Quality teams can test the idea before connecting a new platform, he suggests. Pick a small set of completed inspection records: one legitimate out-of-tolerance result, one conforming part, and one example with mismatched revisions. Ask an AI assistant to summarize the evidence and require it to identify the applicable revision and flag missing provenance, then compare its answers with an experienced quality engineer's determination.
He advises against scoring the assistant only on how often it labels a part good or bad. The points to record are whether it catches a revision mismatch, whether it invents an unstated tolerance, and whether it can point to the approved source behind a conclusion — failures that digital handoffs often conceal.
Ownership when records disagree
Tsipursky also calls for clear ownership when records conflict. A quality engineer may determine whether a measurement is valid, design engineering confirms the governing definition, and manufacturing decides whether production can continue. The escalation route should be settled before an alert appears rather than negotiated during a shipment dispute.
He separates this argument from the familiar call for more AI-versus-human comparisons. Even perfect agreement between human and automated inspectors can reinforce a wrong decision if both were given an obsolete specification, so the reference itself needs scrutiny. Automation can move measurement data farther and faster, but a reliable chain of custody keeps the drawing, feature, method and result together at the moment of decision.
Tsipursky is chief executive of Disaster Avoidance Experts and author of The Psychology of AI Adoption at Work: From Resistance to Results (Georgetown University Press, 2026).