After-hours handover notes and AI privacy: what a GP handover gives away in ChatGPT or Claude.

At the end of the week a GP writes a handover for the after-hours doctor: who might call, what to watch for and what has already been decided. It is written fast, often from memory, and sometimes tidied in ChatGPT or Claude before it is sent. A handover holds several patients at once, and it concentrates the most serious matters the practice is managing. This page marks a routine Friday handover and shows how the same note is drafted on Ermos inside the practice.

The after-hours handover note

Written by the GP at the end of a session for the after-hours or deputising doctor who covers the practice’s patients overnight or over the weekend.

Wombeyan Street Family Practice
8 Wombeyan Street, Eastwood NSW 2122 · (02) 5550 0161

After-hours handover · Friday 6 November 2026, from Dr Theodora Vasilakis

1. .

2.

3.

An invented handover note. Every patient is fictional and the mobile number is masked. Hover or tap any highlighted item for the analysis; the full list is below.

What leaves the practice

  1. 1
    A patient’s name, age and home address

    Identifies an elderly woman and where she is being cared for. Privacy Act guide.

  2. 2
    A terminal diagnosis and palliative medicines

    A cancer diagnosis and the medicines in her syringe driver. Health information is sensitive information under the Privacy Act. Patient data guide.

  3. 3
    A resuscitation decision and advance care directive

    An end-of-life decision, and where the document is kept. Patient data guide.

  4. 4
    A family disagreement

    Two named children, a mobile number and a dispute about her care. Private family detail Good medical practice 4.4 asks doctors to hold in confidence. Patient data guide.

  5. 5
    A child and his mother

    A four-year-old and his parent, identified by name. Privacy Act guide.

  6. 6
    A child protection concern and report

    An injury, a suspicion and a report to the Child Protection Helpline. Among the most sensitive information a practice can hold. Patient data guide.

  7. 7
    An instruction to protect the report

    Who must not be told. If it leaked, the person it names would learn that a report was made. Patient data guide.

  8. 8
    A pending result

    A suspected diagnosis and a scan against a named patient. Patient data guide.

  9. 9
    The standard closing line

    Routine wording with nothing about any patient.

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 general practitioner (GP) practices. See how this works in practice: Care plans on Ermos.

Common questions

Do ChatGPT and Claude use the information in an after-hours handover note?
Yes, a pasted handover 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 covers a palliative patient's resuscitation order, a family dispute, a child protection report and a scan result still to come.
Is it safe to paste a handover note into ChatGPT or Claude if we use initials?
No, because the details still identify each 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 91-year-old on a syringe driver in a named street and a four-year-old with a report to the Child Protection Helpline are recognisable without their names. 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.
What if a GP uses their own ChatGPT or Claude account to tidy a handover note at the end of the day?
The practice cannot see what was sent from a personal account. On a personal plan the user cannot choose where conversations are stored or processed. Some business plans let an organisation choose to store data at rest in Australia (OpenAI), which still leaves the content with the AI company. Both companies can also review content that is flagged for safety or policy reasons. A tired GP who pastes this note sends a child protection report and an instruction about who must not be told outside the practice's own records, so the practice could not say where that information went.
How can a GP use AI for an after-hours handover 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 handover is drafted from the patient records the GP selects, in the practice's own format. The GP checks it before it goes to the after-hours doctor.
Can a GP share patient information with the after-hours doctor?
Sharing with the doctor who will care for the patient is part of providing the health service. An AI tool used to tidy the note is a different recipient, outside that care.
Why is a handover note riskier than a single patient's note?
It concentrates several patients and the most serious matters the practice is managing into one short document, so a single paste discloses all of them at once.
Is a suspected child protection matter covered by the Privacy Act?
Yes. It is health information about the child and personal information about the family, and it is sensitive information. APP 11.1 requires reasonable steps to protect it from unauthorised access and disclosure.

See Dominion answering from your own documents.

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

Book a demo
Keep reading
AI data leakage in general practitioner (GP) practices AI for general practitioners (GPs) AI and patient data All professions