• Home |
  • Artificial Intelligence (AI) Use Policy

Artificial Intelligence (AI) Use Policy

Bronte Family Doctors (BFD)

Policy owner: Practice Manager
Approved by: Practice Principal
Effective date: November 2026
Review date: November 2027

Book your IUD discussion

30-45 MINUTE APPOINTMENT

Your consultation may include

  • Review of your current health conditions
  • Weight history and previous weight loss attempts
  • Nutrition and eating patterns
  • Physical activity
  • Sleep and stress
  • Medications that may contribute to weight gain
  • Hormonal factors, including perimenopause and menopause
  • Assessment of cardiovascular and metabolic health
  1. Purpose and Scope

Bronte Family Doctors supports the responsible use of artificial intelligence (AI) to improve clinical documentation, administrative efficiency and patient care, while maintaining patient safety, confidentiality and professional standards.

This policy applies to all staff, GPs, contractors, locums and students.

AI is a support tool and must never replace clinical judgement or professional accountability.

  1. General Principles

All AI use at BFD must comply with Ahpra guidance, relevant privacy legislation and professional obligations.

  • Accountability: Users remain responsible for the accuracy, appropriateness and safety of all AI-generated content.
  • Transparency and consent: Patients must be informed when AI is used in their care, with consent obtained where required.
  • Privacy: Patient information must only be processed through approved systems with appropriate safeguards.
  • Safe use: Users must understand the purpose, limitations and potential inaccuracies or biases of AI tools.
  • Human oversight: All AI-generated content must be reviewed before being used, saved or shared.

Only AI tools approved by the Practice Principal may be used for practice purposes. New tools or AI features must be assessed before implementation.

  1. Approved AI Tools

Clinical AI Scribes

BFD permits the use of approved healthcare-specific AI scribes, currently:

  • Heidi Health
  • Lyrebird Health

These may be used by authorised clinicians to assist with drafting consultation notes, referrals and other clinical documentation, subject to patient consent and clinician review.

AI scribes must not independently make clinical decisions, diagnoses, treatment recommendations or Medicare billing decisions.

General-Purpose AI

General AI tools, including ChatGPT, Claude, Copilot and Gemini, may be used for approved non-clinical tasks such as:

  • Practice management, policies and administrative communications
  • Staff education and general research
  • Drafting website content and patient information
  • Preparing non-confidential meeting summaries and business documents

Identifiable or potentially re-identifiable patient information must never be entered into general-purpose AI tools. This includes clinical notes, results, images, referral details and patient correspondence.

Confidential staff, financial or commercially sensitive information must not be entered into unapproved AI systems.

AI meeting transcription or recording requires prior approval and consent from everyone participating. General AI tools must not record identifiable patient discussions.

  1. Patient Information and Consent

Patients are informed about the use of AI scribes through BFD’s website, Privacy Policy, patient information materials and practice notices.

Before using an AI scribe, clinicians must:

  1. Explain that the tool listens to the consultation and generates draft clinical notes, including how patient information is handled.
  2. Obtain verbal consent at the beginning of each consultation and document this in the clinical record.
  3. Obtain consent from all other participants in the conversation, including family members, carers, interpreters and students.
  4. Obtain consent from an appropriate parent, guardian or substitute decision-maker where the patient cannot provide valid consent, recognising the capacity of mature minors to consent.
  5. Respect the patient’s right to decline or withdraw consent at any time, without affecting their care.

The scribe must not be activated until consent has been obtained. Clinicians should consider pausing or discontinuing its use during sensitive discussions.

If consent is declined or withdrawn, standard clinical documentation procedures apply.

  1. Safe Clinical Use and Documentation

AI-generated clinical documentation is considered a draft until reviewed and approved by the treating clinician.

Clinicians must:

  • Review and correct notes during or promptly after the consultation, before saving them to Best Practice, sending documents or using them for Medicare billing.
  • Verify clinical details, including medications, dosages, examination findings, diagnoses and management plans.
  • Add relevant observations, non-verbal findings and clinical reasoning that may not have been captured.
  • Ensure documentation is accurate, complete and attached to the correct patient.
  • Confirm that documentation and Medicare billing meet professional and MBS requirements.
  • Manage draft information according to the approved provider’s data-retention arrangements.

Clinicians remain personally responsible for the accuracy of clinical records and decisions, regardless of AI involvement.

Each clinician must use their own authorised account and comply with BFD’s cybersecurity and access requirements.

If an AI tool is unavailable or unsuitable, clinicians must continue documentation using standard practice procedures.

  1. Privacy, Security and Approval of AI Tools

All AI use must comply with BFD’s Privacy Policy and applicable legislation, including:

  • Privacy Act 1988 (Cth) and Australian Privacy Principles
  • Health Records and Information Privacy Act 2002 (NSW)
  • Surveillance Devices Act 2007 (NSW)

Before approving an AI tool that processes patient information, the Practice Principal must ensure an assessment has been completed covering:

  • Clinical suitability, accuracy and intended use
  • Data collection, storage, retention, deletion and any overseas processing
  • Privacy protections, encryption and access controls
  • Use of information for AI training or other secondary purposes
  • Provider security, breach notification and contractual terms
  • Patient information and consent arrangements
  • Integration with practice systems and business continuity

Provider assessments, approvals and agreed data-retention settings are maintained separately in BFD’s AI Provider Assessment Register.

AI tools must not be used with patient information until approved.

  1. Training, Monitoring and Incident Management

All staff and contractors must be familiar with this policy as part of induction.

Clinicians using AI scribes must understand the tool’s limitations, privacy arrangements, consent requirements and procedures for reviewing documentation.

The Practice Manager is responsible for:

  • Maintaining the register of approved AI tools and user access
  • Coordinating training and communicating relevant updates
  • Monitoring AI use and identifying emerging risks
  • Coordinating incident reporting and provider communication

The Practice Principal oversees clinical safety and reviews AI-assisted documentation periodically for accuracy, completeness and recorded consent.

Any AI-related error, privacy concern, suspected data breach or patient complaint must be reported promptly to the Practice Manager.

Incidents are managed under BFD’s incident management, privacy and data breach procedures, with appropriate corrective actions documented.

  1. Responsibilities and Policy Review

Practice Principal: Approves AI tools, oversees clinical safety and reviews the policy.

Practice Manager: Coordinates implementation, staff training, provider assessments, access and incident management.

Clinicians: Obtain consent, verify all AI-generated clinical documentation and maintain professional accountability.

All staff and contractors: Use only approved AI tools, protect confidential information and report concerns.

This policy is reviewed annually, or sooner following significant changes to AI technology, legislation, professional guidance or practice procedures.

Related documents: Privacy Policy; Cyber Security Policy; Data Breach Response Plan; Incident Management Policy; AI Provider Assessment Register; AI Scribe Patient Information Sheet.

Secret Link