Issue 07. The Clinician’s AI Safety Check.

^{Clinically Intelligent is written by Juwon Akinyande in a personal capacity and is not affiliated with, endorsed by, or representative of Barts Health NHS Trust or any other NHS organisation. Content is for general educational purposes only. It does not constitute clinical, legal, or information governance advice. Before applying any guidance to your own practice, consult your Trust information governance lead, your Caldicott Guardian, your line manager, and your professional body.}

A short note this week. No tool. No workflow. Just something I think matters more right now.

I originally thought this would be a short downloadable resource attached to the newsletter. But the more I worked on it the more I realised it needed to exist regardless.

When I started using AI in clinical work nobody told me what I could put in. Nobody told me what I could not. I did not know what DCB0129 was. I did not know where the line sat for a clinician using consumer tools in an NHS setting. I learned by sitting with legal advisors, by writing six issues of this newsletter, and by noticing that every time I recommended a tool I had to do the same research about what kept that recommendation inside the rules.

The pattern that came out of that is the same pattern most clinicians on the ground are quietly working out for themselves. Often after they have already pasted something they probably should not have.

Whether organisations are ready or not, clinicians are already using these tools. We see patients every day. We are the people already using AI to save time on documentation, communication, and research. And almost nobody has given clinicians a practical framework for thinking more safely about where those boundaries sit.

This is what I wish I had been given when I started.

The examples and scenarios inside the guide are fictional and composite.

What it is.

A three stage nine question framework for NHS clinicians using consumer AI tools in clinical work. Run it before each session. Thirty seconds. Then decide.

It covers:

It also covers:

Access: The Clinician’s AI Safety Check

Who it is for.

Any NHS clinician using consumer AI tools at work. Especially anyone who has ever pasted something into ChatGPT and wondered afterwards whether they should have.

How to use it.

Run it before each session. Not once at the start of your career. Every time.

Thirty seconds before you act is the cheapest place to catch a mistake.

Keep it bookmarked. Share it with your team. If someone around you is using AI without a framework this is the framework.

This is not Trust policy and does not replace local guidance, IG procedures, or advice from your governance team. It is a practical starting point for clinicians who are already using these tools and need somewhere to begin.

That is this week’s issue.

Next week I want to talk about what seven issues of testing AI in clinical work has taught me so far. Some of it was not what I expected.

If someone you know would find this useful pass it on.

Clinically Intelligent drops every Wednesday. If you are not yet subscribed you can join free at clinicallyintelligent.com.

^{The AI tools discussed in Clinically Intelligent are consumer products. They have not been independently assessed by the author against DCB0129 or DCB0160 clinical risk management standards, and they may not be approved for clinical use by your employer. Before using any tool described in this newsletter in connection with your clinical practice, you must satisfy yourself that its use is permitted under your Trust information governance policy, your DSP Toolkit obligations, your professional registration requirements, and any applicable contractual terms with your employer. The author accepts no liability for use of any tool or workflow described in this publication. Patient identifiable information must not be entered into any consumer AI tool under any circumstances, irrespective of any guidance contained in this newsletter.}