As of 13 August 2026, AI can only partly check your Attendance Allowance eligibility.

This still needs a person who signs their name to it.

Can you do it?

5 minutesto a draft.

30 minutesto something you’d act on.

Cost, all in£0

Skill neededchat-fluent

Who has to check ityou

What the alternative costsNo price for a benefits adviser is provided in the available tool data.

If this goes wrong: you misunderstand the care or supervision test, leave important difficulties out of your claim, or delay seeking support.

What to actually do

Hand it to a person
The route this page recommends
A person who owns the outcome does this end to end, worth it when the failure is dear.
Use a tool built for this
Second choice

Do it yourself

The distant third

A chat interface, chat-fluent skill, and roughly 30 minutes until you can act on the result.

How to actually do it

Open the current Attendance Allowance guidance and application information on GOV.UK, and keep the pages available for comparison.
Write down your age, UK nation, health conditions, when your difficulties started, and the care or supervision you need on both typical and bad days.
Gather relevant evidence you already hold, such as care information or medical correspondence, and note what each document actually supports.
Paste the prompt into ChatGPT, Claude or Gemini, replace every bracketed slot with your own facts, and ask it to produce the cautious eligibility screen.
Compare every criterion, missing fact and source link in the response with the GOV.UK pages, correcting any statement that is not supported there.
Use the resulting questions and evidence list to complete the official application or take the uncertain points to a welfare rights adviser before submitting it.

Prompt

I want a cautious eligibility screen for Attendance Allowance in the UK, not a final decision and not professional advice. Use the current GOV.UK Attendance Allowance guidance as the primary source and state the page or link you relied on. Do not guess about rules, rates, qualifying periods or my circumstances.

My details:
- Country and nation within the UK: [England, Scotland, Wales or Northern Ireland]
- Age and State Pension age position: [details]
- Health conditions or disabilities: [details]
- When each condition or difficulty began: [details]
- Help I need with personal care, including what happens on a typical bad day and typical day: [details]
- Supervision or support I need to stay safe: [details]
- How often the help or supervision is needed, and at what times: [details]
- Help I currently receive from another person: [details]
- Supporting evidence I have: [details]
- Anything that makes the account uncertain or complicated: [details]

Create a table with these columns: published criterion, facts I supplied, what is missing, and what I should verify. Separate facts from assumptions. Explain which parts appear potentially relevant and which cannot be assessed from the information given. Do not say that I am definitely eligible or definitely ineligible. List the exact questions I should answer on an application and the evidence I should gather, without inventing evidence. Finish with the current GOV.UK application route and questions to ask a welfare rights adviser if any important point remains unclear.

Open it prefilled in ChatGPT or Claude, or copy it into Gemini, which takes no prefill link.

What it gets wrong

AI cannot decide how the legal tests apply to borderline or fluctuating care and supervision needs.
AI cannot observe your daily difficulties or know whether your description leaves out important support needs.
AI cannot confirm that a medical letter or other evidence proves the point you need it to prove.
AI cannot replace the DWP decision-maker or accept responsibility for the claim.
A confident summary can make you stop checking the current GOV.UK rules.

What caps this at PARTLY: judgement under ambiguity, verification cost and stakes of error.

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	1
Inputs	2
Verification	1
Liability	1
Effort delta	2
Total	7 / 10

The methodology and its thresholds are published in full.

FAQ

Can ChatGPT tell me if I qualify for Attendance Allowance?: Only partly. It can organise your circumstances against the current GOV.UK criteria and show what information is missing, but it cannot make the DWP decision or reliably resolve every borderline case.
What questions does Attendance Allowance ask?: The questions focus on how your health conditions or disability affect your personal care and safety, including the help or supervision you need and how often you need it. Use the current GOV.UK application guidance for the exact form and do not rely on an AI summary alone.
Can AI help me fill in an Attendance Allowance form?: Yes, it can turn your notes into a clearer draft and identify missing examples, provided you check every sentence against your real experience. It must not invent symptoms, care needs, frequency or evidence.
Should I ask someone before claiming Attendance Allowance?: Ask a welfare rights adviser if your care needs are difficult to describe, fluctuate, or do not fit the published examples clearly. This is not professional advice, and a serious or disputed case needs a welfare rights adviser to check the application before you send it.

Nearby answers

All Benefits & entitlements rulings

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.