QHSE professionals are increasingly turning to ChatGPT and similar tools for drafting help and quick answers. Here's what they're asking, and where AI genuinely helps versus where it can quietly get you into trouble.
Industry research on quality engineering has found a majority of organisations now using generative AI somewhere in their quality function — mostly for drafting support, nonconformity investigation prompts, and structuring procedures. That's a genuine productivity gain. It's also created a new set of questions we hear constantly from quality managers, IMS leads and QHSE consultants.
It can produce a plausible-looking first draft — a structure with Purpose, Scope, Responsibilities and Document Control headings — but it doesn't know how your business actually operates, so the content is generic until a real person edits it against your processes. Used as a drafting assistant to speed up the blank-page problem, then reviewed and corrected by someone who understands both the standard and your operation, it can genuinely save time. Used as a substitute for that review, it produces documentation that reads well and fails an audit, because auditors test whether you follow your procedures, not whether they're well written.
Not in any near-term sense. AI is increasingly used as a tool within quality management — drafting support, root-cause brainstorming, compliance-checklist prompts — but ISO certification audits are conducted by accredited human auditors, and the judgement calls in quality management still need a person who's accountable for the outcome.
Auditors don't check how a document was drafted — they check whether it's accurate, controlled, and actually followed. AI-assisted drafting is fine as a starting point. The risk is documentation that looks polished but doesn't reflect real practice, which shows up quickly at audit when an auditor asks a member of staff to demonstrate the process the document describes.
Yes, for revision-style prep — generating likely audit questions for a clause, checking your understanding of a requirement, or drafting an internal audit checklist to refine. It's not a substitute for a genuine internal audit programme or for knowing your own management system, since an AI tool has no visibility into what you actually do on your shop floor.
ISO 9001:2026 doesn't mandate specific AI controls, but its sharpened focus on risk-based thinking means that if your organisation uses AI in processes affecting product or service quality, that use falls within the same risk-and-opportunity thinking as any other process. Organisations already subject to the EU AI Act, or supplying customers who are, should expect quality and AI-governance requirements to increasingly overlap.
It can be a useful brainstorming partner for structured techniques like 5-Why or fishbone (Ishikawa) analysis — suggesting categories of cause a team might otherwise overlook. It can't replace the investigation itself: root cause has to be verified against real evidence from your process, not just plausible-sounding AI suggestions.
The main risks are plausible-sounding inaccuracy, generic content that doesn't reflect your actual process, and version control — an AI tool has no idea which of your documents is the current controlled version. Treat AI output as an uncontrolled draft that needs human review and formal document control before it enters your management system.
If your organisation develops or deploys AI systems, especially anything that could be classed as higher-risk under the EU AI Act, it's increasingly sensible to bring that governance into your existing management system rather than running it as a separate initiative — the same clause structure ISO 9001:2026, ISO 14001:2026 and ISO 45001 share makes this a natural fit for an integrated management system.
We use AI tools ourselves for the unglamorous parts of the job — first drafts, formatting, checking a clause reference. What we don't do is let a tool decide what "compliant" means for your business. If you want help building a quality management system that's genuinely yours, not a generic template, that's the conversation to have.
Standards & methodologies we work to
Sectors we work in
AI can draft a paragraph. It can't run your management review or sit in front of an auditor. Let's talk about what your business actually needs.
Talk to a consultant →