Should You Use ChatGPT for Your NHS Supporting Statement?
It depends entirely on what you use it for. This guide explains where AI tools help, where they get NHS applications badly wrong, and what the panel is actually scoring.
Where AI tools can genuinely help
AI tools are not useless for NHS applications. Used in a specific, limited way, they can reduce the blank-page problem and help you organise your thoughts before you start writing. The issue is that most people use them to write the statement itself, which is where things go wrong.
Brainstorming your experience
If you are struggling to identify which of your experiences map to the person spec criteria, asking an AI tool to help you brainstorm is reasonable. You describe what you have done; it helps you see which experiences might be relevant. You still need to verify that against the actual criteria.
Structuring your STAR responses
If you have a rough account of an experience and want help identifying where the Situation, Task, Action, and Result elements are, an AI tool can point them out. It will not write better STAR responses than you can once you understand the structure, but it can help you see what is missing in a first draft.
Proofreading a draft you have written
Asking an AI tool to check your grammar and flag unclear sentences is a legitimate use. This is different from asking it to rewrite your statement. You want the corrections, not a replacement voice.
Everything above treats AI as a thinking aid, not a writer. The problem most applicants run into is that they skip these preparatory steps and ask the tool to produce the finished statement from a brief. That is where the failure modes begin.
Five ways AI-generated NHS statements fail at shortlisting
NHS shortlisting panels score each criterion individually. A statement that reads well but does not address the right evidence in the right order will score low even if it is grammatically perfect. These are the specific failure modes we see most often.
It does not know the person spec
A person spec for an NHS Band 6 Clinical Nurse Specialist is not a generic job description. Each criterion is a scored shortlisting question. An AI tool given a job title and a paragraph about your experience will write to general healthcare competencies, not to the specific wording of the criteria the panel is scoring. Even if you paste in the job advert, AI tools frequently miss the distinction between essential and desirable criteria and fail to address all of them in the required order.
It invents specific details
AI tools fill gaps with plausible-sounding specifics. Figures, timescales, patient numbers, and outcomes that you did not provide will appear in the draft because they make the text read more convincingly. These are fabricated. If you submit a statement claiming you managed a caseload of 40 patients per week and the panel asks about it at interview, you will not be able to support it. This is not a minor editing problem. It is a credibility risk.
NHS panels recognise the patterns
Hiring managers and shortlisting panels in NHS Trusts read hundreds of applications per round. AI-generated NHS statements have a recognisable structure and vocabulary: they open with "I am a dedicated and compassionate healthcare professional", they use phrases like "patient-centred care" and "collaborative working environment", and they close with something about being excited to contribute to the Trust's vision. This language does not score points. It signals that the applicant has not engaged with the criteria.
It does not understand per-criterion scoring
NHS shortlisting does not work by reading the whole statement and giving it an overall score. Each criterion is scored separately, often by different panel members. A statement that covers three criteria well and buries the fourth in a general paragraph will score zero on that criterion even if the statement as a whole reads well. AI tools do not produce criterion-by-criterion evidence maps. They produce statements that flow well as documents but are structurally wrong for per-criterion scoring.
It dilutes your voice with professional language
NHS panels are looking for evidence, not eloquence. A statement written in your own words, describing what you specifically did and what specifically happened as a result, is more persuasive than a polished paragraph full of sector vocabulary. When AI rewrites your experience into formal healthcare language, it often strips out the detail that made the evidence credible. The result is something that sounds more professional but scores less.
NHS panels score evidence, not polish
Every NHS supporting statement is shortlisted against the person specification for that role. The panel works through the essential criteria one by one and looks for evidence that you meet each of them. The evidence needs to be specific: what you did, in what context, and what the outcome was. That is the STAR structure (Situation, Task, Action, Result), and it is what NHS recruitment guidance requires.
A well-written AI statement typically scores poorly because it addresses the criteria generally rather than specifically. "I have extensive experience supporting patients through complex care pathways" is not evidence. "In my Band 5 role at [Trust], I coordinated discharge planning for a cohort of 12 patients across two wards, reducing delayed discharges by 20% over six months" is evidence. AI tools default to the first form because it sounds credible without requiring actual facts.
The word count on an NHS supporting statement is usually between 1,000 and 1,500 words. Every sentence needs to earn its place. Generic sentences about values, motivation, and enthusiasm use word count without scoring points. A statement that addresses every essential criterion with one strong STAR example each will consistently outperform a longer, more polished statement that does not map to the criteria.
This is the structural problem with using AI to write the statement: it optimises for readability, not for per-criterion scoring. If you submit what the AI produces without rebuilding it around the criteria, you are likely to be outscored by applicants who wrote less polished statements but addressed the person spec directly.
If you are going to use AI, use it this way
The following approach uses AI as a support tool while keeping you in control of the evidence and the structure. None of these steps ask the AI to write your statement for you.
List every person spec criterion before you start
Open the NHS Jobs listing and copy out every essential criterion. This becomes your scoring checklist. Your statement needs to address every one of them, in order, with evidence. Do this before you touch any AI tool.
Write a rough STAR example for each criterion yourself
Do not worry about the writing quality at this stage. You want the facts: what the situation was, what your specific task was, what you did, and what measurably changed as a result. This is the content the panel needs. Everything else is presentation.
Use AI to identify gaps in your STAR structure
Paste one rough STAR example at a time and ask the AI to tell you which of the four elements is weakest. Do not ask it to rewrite. Ask it to identify what is missing. This keeps you in control of the evidence while using the tool's pattern recognition constructively.
Write the improved example yourself
Take the feedback and write the improved version in your own words. The specific details (numbers, timeframes, outcomes) must come from you because only you know what actually happened. This is the version you submit.
Check the word count against the criteria count
Divide your available word count by the number of essential criteria. If you have 1,200 words and six criteria, each criterion gets around 200 words. If any one criterion is using 400 words and another is getting 50, your scoring will be unbalanced. Rebalance before you submit.
When it is worth getting a human to write it
Senior and specialist NHS roles
At Band 7 and above, the person spec criteria are more complex, the competition is higher, and the cost of a weak application is greater. The NHS Leadership Competency Framework becomes more prominent at this level, and evidencing leadership behaviours in STAR format requires a different approach than clinical competence evidence. This is where professional support has the clearest return.
Gaps, transitions, and cross-sector moves
If you have a gap in your NHS employment, are moving from a different sector, or are applying for an NHS role that does not directly match your current job title, you need to address the transferability of your experience explicitly. AI tools tend to present your background factually rather than strategically. A human writer can position your experience so that the panel sees the relevance rather than the gap.
Entry-level roles with large applicant pools
Entry and junior NHS roles can attract large numbers of applicants, many of whom have similar experience on paper. The shortlisting score is often what separates candidates who are called to interview from those who are not. If the person spec criteria are being addressed well and the STAR evidence is specific, a junior applicant can outscore a more experienced candidate who wrote a weaker statement. The investment is smaller at this level and the competitive difference can be significant.
More from our team
Why AI-Written CVs Get Rejected by UK Recruiters
How recruiters and ATS systems identify AI-generated CVs and what triggers rejection at the screening stage.
Read more →NHS Supporting Statement Examples and Professional Help
How we write NHS supporting statements, what the panel scores, and the options available if you want professional support.
Read more →Want your NHS supporting statement reviewed by a human?
Send us the person spec and what you have so far. We will come back with specific feedback on which criteria you are evidencing well and which need stronger STAR examples. Free, no commitment.