Is It Safe to Use AI for NMC Revalidation? Judging Any Tool Against The Code

A new wave of tools now promises AI-assisted revalidation: audit-ready portfolios, smart CPD tracking, reflective account drafting. Some are made by established companies. Some appeared last month. Either way, the question is not which tool is shiniest. The question is whether you can trust one with the record that keeps you on the register.

You do not need to take a marketing page's word for it. The NMC has already given you the yardstick. It is The Code, and any AI tool you use has to work within it, not around it.

This guide walks through the three parts of The Code that matter most when an AI touches your portfolio, and what the NMC's own revalidation rules say about who owns the final content. (If you are newer to this topic, our guide to what "NMC AI" actually means covers the background: what these tools do, and the NMC's plans for AI standards in the Code review.)

First: what the NMC actually requires of you

Revalidation rules do not mention AI. That cuts both ways. Nothing published bans you from using a tool to organise your portfolio or structure a draft. Nothing published endorses one either. What the rules do say, precisely, is this:

"You must have prepared five written reflective accounts in the three year period since your registration was last renewed or you joined the register. Each reflective account must be recorded on the approved form and must refer to: an instance of your CPD, and/or a piece of practice-related feedback you have received, and/or an event or experience in your own professional practice and how they relate to the Code."

Read that again: an event or experience in your own professional practice. The reflection has to be anchored in what actually happened to you. A tool can help you structure it. It cannot supply the experience, and it cannot be the one who relates it to the Code. That remains yours.

Test 1: Accuracy of records (The Code, section 10)

The Code says: "Keep clear and accurate records relevant to your practice."

When an AI drafts a reflective account from your voice note, ask where the words come from. A good tool structures what you said. A bad one embellishes, invents plausible-sounding details, or flattens what actually happened into generic reflection boilerplate. You cannot tell the difference unless you read the draft against your own memory of the event, line by line.

Before trusting any tool:

• Does it produce a draft from your input only, with nothing added you did not say?
• Can you see the draft and edit it directly, rather than accepting a polished version wholesale?
• Does it encourage you to verify the final text, or quietly nudge you towards "approve"?

If a tool's output drifts from what you actually experienced, that is an inaccuracy in a professional record. Whether a machine made the error changes nothing about who answers for it.

Test 2: Confidentiality (The Code, section 5)

The Code says: "Respect people's right to privacy and confidentiality."

Your reflective accounts, feedback records, and practice hours are professional records, and they can easily contain details that identify patients, colleagues, or workplaces. Where does that data go?

• Is data encrypted in transit and at rest?
• Is your content used to train models shared with other users? If the terms are unclear, that is your answer.
• Can you delete your data, and take it with you, when you leave?
• Does the tool discourage patient-identifiable information, and strip or flag anything sensitive?

Some tools ask you to paste reflections into a general-purpose chatbot. That is the weakest position: consumer AI services typically reserve broad rights over what you type, and you have little control over where it ends up. A purpose-built tool with a clear privacy policy is a different proposition, but only if you actually read the policy.

Test 3: Personal accountability (The Code, section 20)

The Code says: "You are personally accountable for your practice and for working within the limits of your competence."

This is the test that outranks the others. When your confirmation conversation happens, when a reflective discussion partner asks about an account, when the NMC samples your portfolio: it is you in the room, not the software. The tool does not attend the discussion. It does not answer the follow-up questions. It cannot be referred to the NMC.

Practically, that means the division of labour should look like this:

Safe to delegate to AI	Never delegate
Structuring your own account of your own experience	Inventing or padding the experience itself
Calculating CPD and practice hour totals	Deciding your reflection "sounds good enough" without reading it
Flagging gaps in your portfolio	The reflective discussion with a colleague
Reminders before your deadline	Confirmation, submission, and every judgement the NMC asks you to make

What this means for the tools you are looking at

AI-assisted revalidation is a reasonable idea, and the NMC's direction of travel suggests formal guidance is coming rather than a ban. The NMC's Code and revalidation survey drew over 12,500 responses, and initial findings published in November 2025 showed registrants want "clearer standards on the safe and appropriate use of artificial intelligence (AI) tools in nursing and midwifery." Employer roundtables in April 2026 showed support for principles guiding AI use in revalidation. The framework is being built. Until it arrives, The Code is the framework.

So judge every tool on the three tests above. An audit-ready portfolio means nothing if the accounts inside it are not genuinely yours. Smart learning means nothing if the confidentiality terms are vague. A free trial costs you nothing but an email; the wrong record costs you a lot more.

How Revalidation Copilot handles this

Revalidation Copilot drafts reflective accounts from your voice notes or text, tracks your CPD and practice hours, and helps connect your reflections to the Code's themes. It does not invent experiences, does not write reflections without your input, does not submit anything to the NMC, and does not store patient-identifiable information. Every draft is explicitly a draft: you review, edit, and own the final version before it becomes part of your portfolio.

If you are comparing tools, use the three Code tests on us too. We would rather you did than took our word for it.

Related reading

• What is NMC AI? How AI Helps UK Nurses with Revalidation
• NMC Revalidation Checklist 2026: Everything You Need to Submit
• How to Write Reflective Accounts for NMC Revalidation (With Examples)
• What Counts as CPD for NMC Revalidation? (35 Hours Explained)

References

1. Written reflective accounts: NMC revalidation requirements
2. The Code: NMC (sections 5.1, 10.1, 20.1)
3. Initial feedback on the Code calls for greater focus on AI, challenging discrimination and staff wellbeing: NMC, November 2025
4. NMC seeks feedback from employers and self-employed registrants on Revalidation changes: NMC, April 2026
5. Code and revalidation reviews: NMC, updated January 2026

Disclaimer: Revalidation Copilot is an independent tool and is not affiliated with, endorsed by, or approved by the Nursing and Midwifery Council (NMC). It uses AI to help nurses organise records and draft structured reflective accounts. All AI-generated content must be reviewed, edited, and approved by the user. The NMC does not endorse or require any specific AI tool. Always refer to official NMC guidance for your full revalidation requirements.

← Back to all articles