Healthcare Clinical Supervision Logs: Use case and AI Confidentiality
A provisional psychologist keeps a log of every supervision session: the clients discussed, what she learned, what her supervisor said and what she plans to do next. It is written late in the day and often pasted into ChatGPT or Claude with a request to tidy the summaries, draft learning goals and complete the reflection. A clinical supervision log is meant to be de-identified, but a plumber from Neutral Bay whose brother has died is not anonymous, and a full name slips in. This page marks each item in an invented supervision log that the Privacy Act, NSW health records law and the Psychology Board’s Code of conduct bear on, and shows how the same log is worked on Ermos inside the clinic.
The clinical supervision log
Completed by the supervisee after each session and signed by the supervisor. It records the clients discussed, the reflection, the supervisor’s feedback and the learning goals, and it sits beside the clinical record.
Iluka Street Allied Health
31 Iluka Street, Cremorne NSW 2090 · (02) 5550 0176
Supervision log · October to November 2026 · Confidential
| Supervisee and supervisor | Cleo Armitage, provisional psychologist on the internship program, registration PSY0000000001 (placeholder). Principal supervisor: Meera Joshi, Board-approved supervisor, Iluka Street Allied Health.1 |
| Wed 7 Oct · 60 min · in person | C1: A.B., 44, a plumber from Neutral Bay, wife Karen, whose brother died by suicide in September. Session 2 of 6. We discussed grief versus depression and whether to ask about risk at every session.2 |
| Wed 21 Oct · 60 min · in person | C2: I wrote up Mrs Cheryl Bonnett’s intake. She is 67, newly widowed and lives on Lagoon Street, Cremorne. Meera suggested a ten-minute check on her sleep at every session.3 |
| Wed 4 Nov · 90 min · video | C1 again. He told me he has thought about “not being around”, with no plan or intent. Meera asked me to complete the risk assessment form in the session, agree a safety plan and ring her afterwards if anything changes. We discussed whether to tell his GP.4 |
| Wed 18 Nov · 60 min · in person | Reviewed the log, the safety plan for C1 and my goals for December. |
| My reflection | I froze when he said it. My dad’s gambling taught me to go quiet when people are upset. I want to work on this and I have not told my own therapist.5 |
| Supervisor feedback | Meera: good formulation and direct risk questions. Needs to slow the intake down and write her notes the same day. On track for the risk assessment competency and not yet for case formulation across more than one client.6 |
| Concern | The wall of Room 3 is thin and I could hear Callum’s session in Room 2. Ingrid has been told.8 |
| Learning goals | Complete the risk assessment form in session. Write notes the same day. Read the intake template again. Practise a safety plan script.7 |
| Client information | C1 knows I am supervised by Meera and that she sees my notes. C2 has not been told yet. To do before our next session.10 |
| Hours and sign-off | Four sessions, 4 hours 30 minutes in total (60, 60, 90 and 60 minutes). Signed by Meera Joshi on Wed 18 Nov 2026.9 |
An invented log. The clinic, the supervisee, the supervisor and the clients are fictional and the registration number is a placeholder. Hover or tap any highlighted item for the analysis; the full list is below.
Supervision log entry drafter
An Agent+ that turns a supervisee's session notes into a log entry in the layout your practice uses. It separates clients discussed, reflection, supervisor feedback and learning goals, and flags details that could identify a client so the entry can be reworded.
What it does
- Reads the supervisee's session notes
- Drafts the entry in your supervision log layout
- Flags details that could identify a client
What stays with your people. The supervisee reviews the entry and the supervisor reads and signs it. Any view of the supervisee's competence stays with the supervisor.
An Agent+ is built once and anyone in the practice can run it. It works from your own documents and runs on Ermos Dominion, the AI unit in your office, or on Ermos Edge, your own instance in an Australian data centre.
What leaves the practice
- 1The supervisee, the supervisor and the program
Ties the log to two named practitioners. The Psychology Board describes the internship as involving supervision by a Board-approved supervisor, with one principal supervisor. Code of conduct 10.1(b) asks a supervisor to oversee the supervisee’s service-related behaviour and give appropriate feedback, and 10.1(f) says a supervisor might be held responsible for a supervisee’s conduct if she fails to supervise to the standard reasonably expected.
Frontier Models (ChatGPT, Claude) Two named practitioners and a clinic; low risk alone Ermos Edge Stays on your instance in Australia; leaves only inside a permitted excerpt, per question, to the named provider Ermos Dominion Stays on the unit in your office Use case Supervision log entry drafter. Drafts the supervision log entry from the supervisee's session notes, sorting clients discussed, reflections and learning goals into your practice's log layout. See the use case - 2A client described by initials who is still identifiable
Initials, an age, a trade, a suburb, a wife’s first name and a brother’s death are together enough to identify him. The Privacy Act covers information about an individual who is reasonably identifiable (s 6(1)), and a health service provided to him is health information (s 6FA(a)(iii)). Code 3.3(h) says not to disclose, transmit or share any person’s information, even if the person is not directly named or identified, without first getting written and informed consent. Privacy Act guide.
Frontier Models (ChatGPT, Claude) A client who can be identified from the details, and his brother’s death, transmitted Ermos Edge Stays on your instance, but a person must still remove identifying detail from any log shared elsewhere Ermos Dominion Stays on the unit, but a person must still remove identifying detail from any log shared elsewhere Use case Supervision log entry drafter. Drafts the supervision log entry from the supervisee's session notes, sorting clients discussed, reflections and learning goals into your practice's log layout. See the use case - 3A full name that slipped in
The log is meant to be de-identified, but the supervisee has written a client’s full name, age and street. Sharing a client’s information with a supervisor is part of the service. Sending it to an outside tool is another recipient, so APP 6.1 and HPP 11(1) ask for consent or an exception, and Code 3.3(a)(iii) asks that before disclosing information clients be informed about its nature and purpose. Patient data guide.
Frontier Models (ChatGPT, Claude) A client’s full name, age and street transmitted Ermos Edge Stays on your instance in Australia; leaves only inside a permitted excerpt, per question, to the named provider Ermos Dominion Stays on the unit in your office Use case Supervision log entry drafter. Drafts the supervision log entry from the supervisee's session notes, sorting clients discussed, reflections and learning goals into your practice's log layout. See the use case - 4A risk discussion and a safety plan
Thoughts of not being around, the absence of a plan or intent and what the supervisor asked the supervisee to do. Sensitive information (s 6(1)). Code 8.5(a) asks for accurate, factual records and 1.2(j) asks psychologists to consult another practitioner when appropriate, which is what the session was. A drafted summary can overstate or soften the risk, and Ahpra says practitioners must apply human judgment to any output of AI. Patient data guide.
Frontier Models (ChatGPT, Claude) A risk discussion and a safety plan transmitted Ermos Edge Stays on your instance, but a person must still judge the risk and write the plan Ermos Dominion Stays on the unit, but a person must still judge the risk and write the plan Use case Supervision log entry drafter. Drafts the supervision log entry from the supervisee's session notes, sorting clients discussed, reflections and learning goals into your practice's log layout. See the use case - 5The supervisee’s personal reflection
A confession about her own father’s gambling and a decision not to tell her own therapist. It is personal information about the supervisee and a third party, and not about a client. Code 10.1(d) asks supervisors to recognise the power imbalance in a supervision relationship, and the log is written for that relationship rather than for an outside tool.
Frontier Models (ChatGPT, Claude) A supervisee’s personal confession about her family transmitted Ermos Edge Stays on your instance in Australia; leaves only inside a permitted excerpt, per question, to the named provider Ermos Dominion Stays on the unit in your office Use case Supervision log entry drafter. Drafts the supervision log entry from the supervisee's session notes, sorting clients discussed, reflections and learning goals into your practice's log layout. See the use case - 6The supervisor’s assessment of her competence
Code 10.2(a) asks psychologists assessing a provisional psychologist to be honest, objective and constructive and not to put clients at risk by an inaccurate or inadequate assessment. A tidy-up by an AI tool can soften or sharpen the verdict, and the assessment is the supervisor’s alone.
Frontier Models (ChatGPT, Claude) A supervisor’s verdict on a named trainee transmitted; a rewrite can change its meaning Ermos Edge Stays on your instance, but the verdict is still the supervisor’s Ermos Dominion Stays on the unit, but the verdict is still the supervisor’s Use case Supervision log entry drafter. Drafts the supervision log entry from the supervisee's session notes, sorting clients discussed, reflections and learning goals into your practice's log layout. See the use case - 7Learning goals
General tasks such as writing notes the same day and practising a safety plan script. No client is named and nothing here is a result.
Frontier Models (ChatGPT, Claude) Generic goals with no client named; harmless to send Ermos Edge Not about where data goes Ermos Dominion Not about where data goes Use case Supervision log entry drafter. Drafts the supervision log entry from the supervisee's session notes, sorting clients discussed, reflections and learning goals into your practice's log layout. See the use case - 8A concern about overheard sessions
The wall between two treatment rooms is thin and a named colleague’s session can be heard. Code 3.3(c) asks psychologists to provide surroundings that enable private and confidential consultations. APP 11.1 and HPP 5(1)(c) ask for reasonable steps to protect information from unauthorised access or disclosure, and APP 11.3 adds that these steps include technical and organisational measures. Privacy Act guide.
Frontier Models (ChatGPT, Claude) A concern naming a colleague and a room transmitted Ermos Edge Stays on your instance in Australia; leaves only inside a permitted excerpt, per question, to the named provider Ermos Dominion Stays on the unit in your office Use case Supervision log entry drafter. Drafts the supervision log entry from the supervisee's session notes, sorting clients discussed, reflections and learning goals into your practice's log layout. See the use case - 9Hours and sign-off
Four sessions that total 4 hours 30 minutes, signed by the supervisor on a date. The total can be checked against the entries. The names of the supervisor and the date are low risk alone.
Frontier Models (ChatGPT, Claude) Totals and a date; harmless to send beyond the names Ermos Edge Not about where data goes Ermos Dominion Not about where data goes Use case Supervision log entry drafter. Drafts the supervision log entry from the supervisee's session notes, sorting clients discussed, reflections and learning goals into your practice's log layout. See the use case - 10What clients have been told
Code 10.3(e) asks that, when appropriate, clients are informed that provisional psychologists work under supervision and share information with supervisors, and 10.3(d) asks that the scope of a provisional psychologist’s role is made clear and documented. A client who agrees to a supervisor seeing his notes has not agreed to an outside AI tool seeing them. Privacy Act guide.
Frontier Models (ChatGPT, Claude) A consent gap described; sending the log to ChatGPT or Claude would not fix it Ermos Edge The duty to tell clients how their information is used still applies Ermos Dominion The duty to tell clients how their information is used still applies Use case Supervision log entry drafter. Drafts the supervision log entry from the supervisee's session notes, sorting clients discussed, reflections and learning goals into your practice's log layout. See the use case
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 Edge | Ermos Dominion | |
|---|---|---|---|
| Where the document is processed | Provider servers, generally offshore on personal plans | Your instance in an Australian data centre; answers generated in Australia | The unit in your office |
| Who can read it | Provider staff and systems, under policy | Your staff, within the boundaries you set. The named Australian provider processes only permitted excerpts, under contract, and keeps nothing | Your staff only |
| Is it kept | On the provider’s terms | Kept on your instance; the provider is barred from retaining it | Stays on the unit |
| Can it train a model | On consumer tiers, yes | No, by contract | Only your own unit learns from it, on your hardware |
Part of Use cases in healthcare practices. See how this works in practice: Clinical documentation on Ermos.
Common questions
Do ChatGPT and Claude use the information in a clinical supervision log?
Is it safe to paste a clinical supervision log into ChatGPT or Claude?
What if a supervisee uses her own ChatGPT or Claude account for a supervision log?
How can a psychology practice use AI for a supervision log without sending it to ChatGPT or Claude?
Do clients have to be told that a provisional psychologist is supervised?
What does the Code of conduct say about a supervisor's responsibility?
Is a supervision log covered by the Privacy Act and NSW health records law?
See Dominion answering from your own documents.
A demonstration takes 30 minutes, uses no client data, and comes with no obligation.
Book a demo