Consultation notes and AI patient confidentiality: what clinical notes give away in ChatGPT or Claude.

A consultation note is written fast, for the record, and it says more than the patient would say to anyone else: how much they drink, what frightens them, what a parent died of. Doctors paste notes into ChatGPT or Claude to tidy the shorthand, to turn them into a summary for a colleague, or to draft the follow-up letter. This page marks every item in a routine GP consultation note that the Privacy Act, Good medical practice and the NSW health records law bear on, including a line recording an apology to the patient, and shows how the same note is written up on Ermos inside the practice.

The GP consultation note

Written by the treating doctor during or straight after a consultation. Good medical practice 10.5 asks for records made at the time of events or as soon as possible afterwards, accurate, up to date and held securely.

Consultation note
· Dr C. Eze · Thursday 22 October 2026, 11.40am, 25 minutes

S.

Social.

O.

A.

P.

Result communication.

An invented consultation note. Hover or tap any highlighted item for the analysis; the full list is below.

What leaves the practice

  1. 1
    Patient name and date of birth

    The key that ties the whole note to one person. Personal information held by a health service provider, and health information once it sits in the clinical record. Privacy Act guide.

  2. 2
    Symptoms and a family cancer history

    A symptom history and a parent’s cancer. Health information is sensitive information under the Privacy Act, and the family detail bears on the patient’s own risk. Patient data guide.

  3. 3
    Alcohol, smoking, work and a secret from a spouse

    Lifestyle facts the patient gave in confidence, including something his wife does not know. Good medical practice 4.4 asks doctors and their staff to hold patient information in confidence. Patient data guide.

  4. 4
    Examination findings and pathology results

    Findings and laboratory values against a named patient. Health information, and the core of the clinical record under Good medical practice 10.5. Patient data guide.

  5. 5
    The assessment

    The doctor’s reasoning about what this could be, including a possible cancer. Clinical judgement that a person must own, and sensitive to the patient before the investigations are done.

  6. 6
    The plan and the safety net

    Referral, treatment and the advice given about when to seek emergency care. A record that protects the doctor and the patient, so it belongs to the practice.

  7. 7
    A recorded apology and a complaint right

    That a result was communicated late, that the doctor apologised and that the patient was told he may complain. Sensitive to the patient and to the practice, and the start of a complaint file. Patient data guide.

  8. 8
    The billing item and the recall

    An item number and a follow-up date. Low risk alone, but against a name it reveals the service delivered and when the patient will be back.

  9. 9
    The signature and time stamp

    Records who wrote and signed the note and when. The doctor’s own act, which a draft cannot replace.

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. See how this works in practice: Clinical documentation on Ermos.

Common questions

Do ChatGPT and Claude use the information in a GP consultation note?
Yes, a pasted consultation note 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 note records a patient's drinking that his wife does not know about, a father's bowel cancer, a positive faecal occult blood test and a recorded apology for a late result.
Can we use ChatGPT or Claude for a GP consultation note if we remove the names?
Not safely, because the clinical story still points to one patient. 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 site supervisor in Chatswood with iron deficiency anaemia, a father who had bowel cancer at 62 and a wife who does not know how much he drinks describes one household to anyone who knows him. Everything in the note 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.
Does deleting the chat remove a GP consultation note from ChatGPT or Claude?
Not completely. Deleting a chat removes it from the user's history, but OpenAI and Anthropic each say they can keep a deleted chat on their systems for up to 30 days, and longer where safety or legal obligations apply (OpenAI, Anthropic). Both products also have memory features that can carry details from one conversation into later ones, and on a personal account each user decides whether they are switched on (ChatGPT, Claude). A deleted chat that held this note would still have held the abnormal results, the plan for an urgent bowel referral and the apology, and the practice could not tell the patient the note had gone.
How can a medical practice use AI for a GP consultation note 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 note is drafted from the patient record and the practice's own consultation note template. The treating doctor checks every entry and signs it.
What does Good medical practice say about records and confidentiality?
Section 4.4 says patients have a right to expect that doctors and their staff will hold information about them in confidence, unless the law requires or permits release. Section 10.5 covers medical records: accurate, up to date, made at the time of events or soon after, held securely against unauthorised access, and kept for the legally required period.
Can AI tidy a doctor's note without breaching confidentiality?
The tidying is fine as a task. The destination decides the risk. If the note goes to a provider that stores it on its own terms, the patient's information has been disclosed to a third party, which APP 6 and, in NSW, the health records principles on disclosure (HPP 11) both govern. On a unit in the practice the note stays with the practice.
Who checks an AI-written note?
The doctor. Ahpra's guidance says a practitioner using an AI scribing tool is responsible for checking the accuracy and relevance of records created using generative AI, and the RACGP asks GPs to review scribe output for false positives and negatives. A tool can draft; the entry in the record is the doctor's.

See Dominion answering from your own documents.

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

Book a demo