AIQI Consortium publishes first AI Conformity Assessment Use Case

The AIQI has published the first in a planned series of AI Conformity Assessment Use Cases, designed to help conformity assessment bodies (CABs) explore where AI could potentially support conformity assessment activities, while also considering the risks, limitations and controls that may be needed.

Use Case No. 1 – Use of AI to Support Management System Certification Decisions considers how AI might be used as a decision-support tool within the review and certification decision process. It explores how AI could analyse information such as audit reports, nonconformities and corrective actions to highlight risks, inconsistencies or gaps that may require the decision-maker’s attention. 

Alongside identifying potential benefits such as improved consistency, reduced risk of overlooking critical issues and more efficient review of complex cases, the use case considers risks including data quality and confidentiality, explainability, automation bias and potential misalignment with accreditation or regulatory requirements.

It also identifies potential control measures, such as human oversight, validation of AI performance, protection of client data, auditable records, staff competence and effective change management.

This is the first in a planned series through which the AIQI intends to examine different potential uses of AI across conformity assessment. The aim is not to tell CABs where or how they should use AI, but to help them ask the right questions: Where could AI provide useful support? What risks might it create? And what controls would be appropriate if it is introduced?