Home · Business · HR & People · Policies & handbooks
As of 13 August 2026, AI can only partly write a mental health policy for your UK employees.
This still needs a person who signs their name to it.
Can you do it?
5 minutesto a draft.
1 hourto something you’d act on.
Cost, all in£0
Skill neededchat-fluent
Who has to check ita professional
What the alternative costsThe supplied tool data gives no price for a UK HR professional or policy-writing service.
If this goes wrong, employees may rely on unclear support arrangements while your organisation treats an incomplete policy as sufficient.
What to actually do
Hand it to a person
The route this page recommends
Someone with a licence or accountable authority has to sign this before it counts.
Use a tool built for this
Second choiceDo it yourself
The distant thirdA chat interface, chat-fluent skill, and roughly 1 hour until you can act on the result.
How to actually do it
- Open the relevant mental health, employment and workplace guidance on GOV.UK, and open NHS 111 guidance for the employee support wording you may need.
- Gather your existing absence, equality, safeguarding, grievance, disciplinary, flexible-working and data-handling policies, plus your actual support contacts and occupational health arrangements.
- Fill in the organisation details in the prompt, paste the relevant policy extracts, and remove personal health information and unnecessary employee-identifying data.
- Paste the complete prompt into ChatGPT, Claude or Gemini and ask it to produce the policy and keep every unknown item marked [TO CONFIRM].
- Compare each factual statement and proposed commitment against your current procedures and the GOV.UK material you opened, then replace placeholders with confirmed contacts, routes and responsibilities.
- Send the draft, source links and implementation checklist to the policy owner and an HR or UK employment-law reviewer, then record approval before publishing it to employees.
- Publish the approved policy with the named contact route, brief managers on their responsibilities, and set the review reminder recorded in the policy.
Prompt
Draft a mental health policy for a UK organisation. Organisation details: - Organisation name: [name] - Sector and main work activities: [details] - Number and types of workers: [details] - Locations and working arrangements: [office, remote, hybrid, mobile or other] - Existing HR, absence, safeguarding and equality policies: [paste or describe] - Available support, contacts and occupational health arrangements: [details] - Policy owner and approval route: [details] - Review date or review interval: [details] Write a practical employee-facing policy with these headings: purpose, scope, principles, responsibilities, prevention and working conditions, recognising concerns without diagnosing anyone, how employees can ask for support, manager response, adjustments and return to work, confidentiality and privacy, absence and emergency escalation, unacceptable conduct, record keeping, complaints, monitoring, related policies, external support, approval and review. Use plain British English. Do not diagnose, assess or label any employee. Do not promise confidentiality where information may need to be shared for safety, legal or operational reasons. Do not invent contacts, services, legal duties, statistics or internal procedures. Mark every missing detail as [TO CONFIRM]. Distinguish policy commitments from optional good practice. Flag statements that need checking against current UK requirements and reliable government guidance, and include source links only when you can verify them. Treat this as a drafting aid, not legal advice. End with a short implementation checklist and a list of questions for an HR or UK employment-law reviewer. The employee-facing text must direct urgent or worrying symptoms to NHS 111 and must not provide diagnosis.
Open it prefilled in ChatGPT or Claude, or copy it into Gemini, which takes no prefill link.
What it gets wrong
- AI cannot know whether your stated support route works in practice or whether managers have the time and authority the policy gives them.
- AI cannot decide how your organisation should balance employee privacy, safeguarding, absence management and operational needs in an ambiguous case.
- AI cannot confirm that the policy matches your other HR documents, employment contracts and established procedures without a human comparing them.
- AI cannot take responsibility for consultation, approval, manager training or the consequences of an employee relying on the policy.
What caps this at PARTLY: legal accountability, judgement under ambiguity and verification cost.
How we scored this
Five axes, each scored nought to two by hand: ten means AI carries the task cleanly, and the thresholds that turn a total into YES, PARTLY or NO are published in the methodology. Each axis name links to its definition.
| Axis | Score (0–2) |
|---|---|
| Output | 2 |
| Inputs | 1 |
| Verification | 1 |
| Liability | 1 |
| Effort delta | 2 |
| Total | 7 / 10 |
The methodology and its thresholds are published in full.
FAQ
- Can AI write a mental health policy?
- Yes, it can produce a useful first draft with headings, procedures and plain-English wording. It cannot decide what your organisation can actually provide or confirm that the draft is suitable for UK employment practice, so have HR or an employment-law professional check it.
- Is an AI-written mental health policy legally compliant?
- You cannot assume that it is. Check every legal or procedural statement against current UK sources and your own policies, then obtain an HR or UK employment-law review before approval.
- What should a UK mental health policy include?
- It should explain its scope, responsibilities, routes for asking for support, manager responses, adjustments, absence and return-to-work arrangements, privacy, record keeping, complaints and review. It should also use accurate internal contacts and direct urgent or worrying symptoms to NHS 111 without attempting diagnosis.
- Can I use ChatGPT for employee mental health policies?
- You can use it to structure and draft a policy, provided you do not paste unnecessary personal health information into the chat. Treat the output as a draft, check it against GOV.UK and your organisation's procedures, and obtain appropriate HR or employment-law approval before publishing it.
Nearby answers
- Can AI write a social media policy for my UK business?YES
- Can AI create a fair dress code policy for my UK employees?PARTLY
- Can AI create a sexual harassment policy for my UK workplace?PARTLY
- Can AI create an equal opportunities policy for my UK business?PARTLY
- Can AI write a disciplinary policy for my UK employees?PARTLY
- Can AI write a hybrid working policy for my UK employees?PARTLY
Assessed by gpt-5.6-luna (gpt-5.6-luna) on 2026-08-13, second-checked by an independent model. Wrong somewhere? Email [email protected] and it gets re-checked.
The newsletter
AI news, new answers and product picks, straight to your inbox.