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.
Earlier this month I was on a Teams call with an AHP Informatics Officer at a different site in my trust. He was talking about his work with his teams. Not AI. SharePoint. He is still trying to get people to move from OneDrive to SharePoint. Still explaining the difference between them. Still asking teams to stop defaulting to the wrong one. It is, in his words, an ongoing challenge.
Then he said something that has stayed with me. AI is not what they are focusing on. Not because it does not matter. Because people are still learning the basics.
That single sentence changes how you read every AI headline coming out of the NHS. Announcements about 505,000 Copilot licences. Ambient voice technology being introduced at trusts including my own. Reports on AI capability gaps. All of these assume a foundation that in most services is not fully in place yet. AI is being asked to grow on top of digital operation that is still consolidating.
What I have been seeing since coming back.
Since returning to work I have been paying close attention to where colleagues actually sit on digital. Most can send an email. Most can open a document. Fewer are confident on SharePoint. Fewer still have thought carefully about AI beyond the one rule almost everyone has absorbed, which is not to paste patient information into consumer AI tools. That rule is doing important work. It is also the entire content of most people’s AI training.
Outside of that rule, the picture is thin. What tools are approved. What tools are not. How to use them safely. When they help. When they do not. What clinical judgement should look like when the output is confident but wrong. Almost none of this is being taught in a structured way. People are working it out on their own or not at all.
I want to give you one small illustration of the wider promotion gap. NHS England has digital literacy resources available through Digital Learning Solutions. I only recently found out. Even now, trying to identify the specific course being pointed at as the official one is harder than it should be. I write a newsletter about AI in clinical practice for AHPs. If I only recently heard of these resources, and cannot easily tell you which one to start with, most working AHPs are in the same position. That is not a resource problem. That is a promotion problem.
At my own trust, ambient voice technology is beginning to be introduced. That is real progress and it belongs in this piece. Adoption is happening. It is just uneven, and the parts of digital work that people most need help with are not always the parts that get the loudest announcements.
What the report shows.
With thanks to Prabha Vijayakumar, National Chief AHP Information Officer at NHS England, for permission to share the National AHP Digital Maturity Assessment with readers of this publication. The full report is on the National AHP Virtual Hub on Future NHS. Colleagues can register for FutureNHS and request access to the workspace. It is worth reading in full. What follows is a summary of the parts most relevant to this piece.
The first AHP Digital Maturity Assessment was published in July 2025. It is the first time this has been measured at national level. The numbers that follow are the baseline. There is no prior data to compare them to.
The report measures AHP digital maturity across seven WGLL domains, with a workforce lens summary that groups the findings into five capability areas.
The strengths are real and worth naming. EPR Workflow Optimisation is at 63% nationally and 62% in London. General Digital Literacy is at 61% both nationally and in London. Data Analytics is at 52% nationally and 56% in London. These are areas where sustained investment over years has produced measurable capability. AHPs are showing up in these programmes even where formal digital roles do not exist for them. The London insight puts this precisely. AHPs frequently contribute to major digital programmes such as EPR change but often without formal titles or uplift.
The gaps are where the picture changes. AI and Decision Support Literacy is at 38% nationally. In London it is 33%, below the national average. Regionally, some areas are lower still. The report describes AI capability as low national readiness with limited awareness of AI tools and minimal structured training across professions.
AHP Digital Leadership is at 28% nationally and 28% in London. The lowest score in the entire report. The exact phrasing is worth quoting. “AHP digital leadership is inconsistent and often informal. Roles lack protected time, senior banding, and national coherence.”
The London insights add one more phrase that captures the pattern in three words. Investment tends to flow into large EPR and data platforms with AHPs “bolted on.”
Read those numbers and phrases together. AHPs are getting on with the digital work they can. What they lack is the specific support built for AI as distinct from general digital transformation. And they lack leadership structures formal enough to build that support at scale.
What is being measured and what is not.
The report uses the word competency. Ten domains of capability. Records. Transfer of care. Decision support. Medicines management. Data. And so on. These are demonstrable skills. You either can navigate an EPR or you cannot. You either can extract meaningful analytics from your service data or you cannot. Competency is measurable, and the framework measures it carefully.
The word literacy is used in the wider conversation as though it means the same thing as competency. It should not. Look at how the word is used in adjacent fields. Health literacy is not about whether a patient can log into a portal. It is about whether they can read information, evaluate it, and use it critically. Media literacy is about whether a reader can spot the difference between a reliable source and an unreliable one.
Applied to digital tools, literacy in this fuller sense means the ability to interpret what a tool is producing, judge whether to trust it, and use it critically. That is a different capability from operating the tool. Competency is operation. Literacy is judgement.
For most of the last twenty years, digital work in healthcare has been about competency. The tools were predictable. Log in. Complete the form. File the Datix. If you could operate them, you were digitally capable in the sense that mattered. The output was correct or not correct in a way you could verify quickly.
AI changes this. The tools are no longer predictable. The output looks confident whether or not it is correct. Operating an AI tool no longer tells you whether you are getting value from it. The skill that decides that is literacy in the fuller sense. Judgement. Knowing when to trust. Knowing when to check. Knowing when to override.
In the next decade, knowing how to operate AI will not be what sets you apart. Knowing how to judge what AI is producing will. The AHP DMA measures the first. The skill of judging AI is not yet being taught.
The gap is resources, not intent.
Everyone I have spoken to in this conversation wants the same thing. People who can use AI safely and thoughtfully. Leaders who can build what is needed. Workforces who feel supported to learn.
The leaders across trusts, regions, and nationally are actively working on this. The Digital AHPs Forum on Future NHS is one of the places where the work happens. There are also regional AHP Digital Forums which provide informal spaces to discuss digital, data, technology, and AI at a regional level. The AI Ambassador Network is another and has been building a genuine community of people who are trying to close the gap in practice rather than only in policy. Regional AHP digital officers are running programmes. National figures are producing reports like the one Prabha shared. Nobody is standing still.
What is missing is not intent. It is resource specifically directed at AI as distinct from general digital transformation. Most of the existing digital investment has gone into EPR programmes and general capability. That is exactly what the numbers show. EPR at 63%. AI at 38%. The investment produced the capability where it was directed. It has not yet been directed at AI in the same way.
The gap that people feel most is in practical, applied training that translates from policy documents into what to actually do at work on a Tuesday afternoon. Workshops rather than PDFs. Applied use cases rather than abstract principles. Time protected space rather than optional evening reading. Real examples of good and bad AI use in the clinical work AHPs actually do. That is the shape of what needs to exist next.
What you can do this week.
Three actions. All small. All honest about what is realistic.
If you manage a team. Ask them who is interested in digital or AI. Give those people some protected time to explore it, even an hour a fortnight. Do not wait for a national programme to arrive. The people who are curious now will be the ones who help their services adapt when the programmes do arrive.
For everyone else. Ask your line manager what AI or digital literacy provision exists at your trust. Ask about AI specifically, not just about digital in general. If the answer is nothing, that is data worth having. Raise it in supervision. Mention it in service meetings. Digital literacy resources are available through Digital Learning Solutions. There is also a free NHS England Digital Academy programme called Digital and Data Skills, designed for AHPs to build digital and data capability at their own pace. The resources exist. The gap is awareness of them and protected time to engage with them meaningfully. Whoever leads on digital or workforce at your trust needs to know that.
If digital or AI already interests you. Find out who leads digital in your organisation. It might be a Digital AHP lead, a CAHPIO, a CCIO, a CNIO, or another digital leader. Reach out and explore opportunities to get involved. If you are already contributing to digital work informally, discuss protected time with your line manager. Align what you are doing with your personal objectives and the trust digital strategy. This is how informal contribution becomes visible and supported.
Where this leaves us.
Everyone in this conversation wants the same thing. The support is being built. This publication is one small piece of it. The report is another. The Digital AHPs Forum is another. The AI Ambassador Network is another. The AHP Informatics Officer at my sister site who is patiently teaching teams the difference between OneDrive and SharePoint is another. Barts Trust introducing ambient voice technology is another. The work is happening. It is just not yet fast enough or connected enough to close the gap the numbers show.
Those of us who thrive in the next decade of AI will not be the ones who use it most. We will be the ones who bring judgement to it. That is a skill worth training for, and it is a skill nobody is yet teaching at scale.
That is all for Issue 13. Next week, back to a tool. 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.