Patient complaint responses and AI privacy: what a complaint letter gives away in ChatGPT or Claude.

Replying to a patient complaint is hard to write: it has to be factual, kind and careful about what it admits. Practice managers often ask ChatGPT or Claude for a first draft, pasting in the complaint, the timeline and the internal findings. The letter that results names the patient, describes what went wrong with their care, says what the practice will change and tells the patient where to complain next. This page marks every item in a routine complaint response that the Privacy Act, Good medical practice and the NSW health records law bear on, and shows how the same letter is drafted on Ermos inside the practice.

The patient complaint response

Sent by the practice manager, usually with the doctor’s input, after a patient raises a concern. In NSW the Health Care Complaints Commission deals with complaints about health service providers if the patient wants to take it further.

Kestrel Lane Medical Practice
14 Kestrel Lane, Wollstonecraft NSW 2065 · 30 October 2026

Dear Mr Nguyen,

Your complaint of 26 October 2026. Thank you for telling us what happened. I am sorry for the worry it caused you and your family.

What happened.

What you told us.

Your care.

What we have changed.

If you are not satisfied.

Yours sincerely,

An invented complaint response. Phone numbers are placeholders. Hover or tap any highlighted item for the analysis; the full list is below.

What leaves the practice

  1. 1
    The patient’s name and address

    A named patient and a home address in a letter about his care. Personal information that is also health information, because the practice holds it as a health service provider. Privacy Act guide.

  2. 2
    The complaint, in the patient’s terms

    What the patient said went wrong, including a spouse seeing a text. His account of a delayed result, and the family detail he gave to explain the harm. Patient data guide.

  3. 3
    The timeline and named staff

    Dates, a clinician, an absence and a receptionist, setting out exactly how a result was missed. Sensitive to the patient, to the practice and to two named employees. Patient data guide.

  4. 4
    The patient’s care and the referral

    That he saw a doctor and was referred urgently to a named specialist, and that the doctor discussed the consequence of the delay. Health information. Patient data guide.

  5. 5
    The changes the practice has made

    Internal process fixes after an incident, ending with this patient’s contact preference. They describe a weakness and the response to it, and the practice may want to control who reads that.

  6. 6
    The notice to the medical defence organisation

    That the practice has told its defence organisation about the complaint. Sensitive to the practice, and a matter for the doctor and the organisation to decide how to describe.

  7. 7
    Where to complain next

    The HCCC deals with complaints about health service providers in NSW. Public information with no patient detail in it.

  8. 8
    A fee waiver and a meeting

    An offer made to this patient. A commercial gesture, and sensitive to the practice if it were to travel.

  9. 9
    The author and contact details

    Names the practice manager and a direct line. Ties the letter to its source.

What happens in Frontier Models (ChatGPT, Claude)

  • The whole document is transmitted to the provider’s servers. On personal plans the user cannot choose where it is stored or processed, and for the major providers that has generally meant outside Australia, under that country’s legal process. Some business plans offer storage in Australia, but the provider still receives the content.
  • On consumer tiers, prompts can be retained and used to train future models.
  • The provider’s staff and systems can reach the content under the provider’s own policies.
  • Once sent, the practice no longer controls it. Retention and deletion happen on the provider’s terms.

The same document on Ermos

Ermos Dominion

The document is drafted on a unit in your office, from your own templates and the patient record. Every marked item stays on the unit, which keeps working with the internet switched off. Nothing in it is transmitted to an AI provider, so there is no disclosure to analyse: the confidentiality work is the same as for your file server.

Ermos Edge

Your documents, the index and your access boundaries stay on your own instance in an Australian data centre. When a staff member asks for this draft, only their question and the excerpts they are permitted to see go to a named Australian provider, who is contractually barred from keeping the content or training any model on it. Nothing is processed outside Australia.

Frontier Models (ChatGPT, Claude)Ermos EdgeErmos Dominion
Where the document is processedProvider servers, generally offshore on personal plansYour instance in an Australian data centre; answers generated in AustraliaThe unit in your office
Who can read itProvider staff and systems, under policyYour staff, within the boundaries you set. The named Australian provider processes only permitted excerpts, under contract, and keeps nothingYour staff only
Is it keptOn the provider’s termsKept on your instance; the provider is barred from retaining itStays on the unit
Can it train a modelOn consumer tiers, yesNo, by contractOnly your own unit learns from it, on your hardware

Part of AI data leakage in healthcare practices.

Common questions

Do ChatGPT and Claude use the information in a patient complaint response?
Yes, a pasted complaint response is sent to the AI company's servers and processed there to produce a reply. As at October 2026, whether a conversation on a personal plan (ChatGPT Free, Plus or Pro; Claude Free, Pro or Max) can be used to train the models depends on a setting each user controls. Business plans (ChatGPT Business and Enterprise, Claude Team and Enterprise) and the developer APIs sit under commercial terms, and OpenAI and Anthropic each say they do not train on that content by default (OpenAI, Anthropic). On either product, the thumbs up and thumbs down buttons can send a whole conversation back for training. This letter names the patient and his address, sets out how a result was missed with two named staff, describes an urgent referral, and says the practice has told its medical defence organisation.
Is it safe to paste a patient complaint response into ChatGPT or Claude if we remove the names?
No, because the timeline still points to one patient and one practice. Removing names rarely makes a document anonymous. Information stays personal information while the person is reasonably identifiable, and the OAIC warns that information can be at risk of re-identification even when it has been de-identified or anonymised. A blood test collected on 12 October, a result that reached the doctor on 14 October and a text sent at 9.12am on 21 October fit only one case. The care described is health information, which is sensitive information under the Privacy Act. The Office of the Australian Information Commissioner (OAIC) recommends in its guidance on commercially available AI products that organisations do not enter personal information, and particularly sensitive information, into publicly available generative AI tools, because of the significant and complex privacy risks.
Can ChatGPT or Claude be trusted to draft a patient complaint response?
Not without a person checking every fact. Language models can produce fluent text with details that are not in the source, wrong dates or references that look real but are not. In this letter a wrong date for the result, a wrong name for a staff member, or an invented promise about the fee waiver or the meeting would do the most harm. The practice manager and the doctor should check each item against the record.
How can a medical practice use AI for a patient complaint response without sending it to ChatGPT or Claude?
Ermos is built so the document stays under your control. On the Dominion AI unit it is drafted inside your office and none of it is sent to an AI provider. On Ermos Edge your documents and index stay on your own instance in an Australian data centre, and only the question and the excerpts the user is permitted to see go to a named Australian provider that is contractually barred from keeping them or training any model on them. The response is drafted from the complaint and the patient record, using the practice's own letter template. The practice manager and the doctor check the facts and sign it.
Who deals with complaints about GPs and practices in NSW?
The Health Care Complaints Commission deals with complaints about health service providers in NSW. Its website has sections for complaints about registered health practitioners, non-registered practitioners and health organisations. A practice response that tells the patient where to go next is common, and the letter itself carries the patient's health information.
Is a complaint response health information?
It is likely to be. Section 6FA of the Privacy Act counts information about a health service provided to an individual as health information, and also other personal information collected in providing a health service. A letter describing a patient's missed result, referral and consultation falls within that.
Can an AI tool draft a complaint response?
It can help with tone and structure, but a person must check every fact and decide what the practice admits and promises. Ahpra says practitioners must apply human judgment to AI output. Where the draft is written matters because the timeline, names and clinical detail all go in with the request.

See Dominion answering from your own documents.

A demonstration takes 30 minutes, uses no client data, and comes with no obligation.

Book a demo