liveBook · Manning

Quality management system

As explained in AI Governance

An organized set of documented processes and responsibilities for maintaining the quality of an organization's products or services.

For a high-risk AI system, a quality management system (QMS) is the documented, repeatable operating framework that connects development, risk management, post-market monitoring, incident handling, and regulator communication. section 7.3.3

A working QMS must:

  • Cover the system’s lifecycle with documented, repeatable processes. section 7.3.3
  • Assign a specific person responsibility for AI quality and compliance; the AI Act does not require a named AI officer. section 7.3.3
  • Support prompt investigation and correction when monitoring identifies problems, including restricting or withdrawing the system if needed. section 7.3.3

Responding to discovered bias

If ClaimAssist V5 monitoring reveals systematic demographic bias, the provider must investigate and correct it; if the fix cannot be verified, the system must be taken offline. section 7.3.3

How AI Governance covers it

How deep it goes
Specifies required process areas and ongoing duties, while leaving detailed monitoring-plan requirements to further guidance.
Examples
Uses ClaimAssist V5’s discovery of systematic demographic bias to illustrate investigation, correction, and possible shutdown.
What you can do after
Operate documented, repeatable processes that coordinate high-risk AI development, monitoring, incident response, and regulatory communication.
How it connects
Links risk management, post-market monitoring, incident handling, conformity assessment, and technical documentation.

Related concepts

This explainer was generated by Manning's knowledge graph from the text of AI Governance. The section links point to the parts of the book it draws on.