What Exactly Is an AI Scribe?
An AI scribe is a tool that uses artificial intelligence to automate clinical documentation. During a consultation, the software listens to the conversation between a doctor and patient, then transcribes and summarises it into structured clinical notes.
Proponents say this technology is a game-changer, designed to tackle one of the biggest sources of physician burnout: administrative work. By handing over the note-taking, doctors can theoretically focus more on the patient in front of them, improving eye contact and engagement rather than being buried in a keyboard.
The Promise of a Lighter Load
The appeal for many doctors is clear. The administrative burden of medicine is immense, with many physicians spending hours after clinic closes catching up on paperwork—time often referred to as “pyjama time”. AI scribes promise to significantly cut this documentation time, potentially reducing stress and burnout. According to a poll by the Royal Australian College of General Practitioners (RACGP), the use of these tools by doctors nearly doubled in just over a year, rising from 22% in August 2024 to 40% in November 2025. Companies offering the technology claim it has been used hundreds of millions of times globally as doctors seek relief.
The Core Issue: Privacy and Data Security
This rapid adoption has outpaced regulation, sparking significant concerns from Australia's federal health department. A central issue is the privacy of incredibly sensitive health information. When an AI scribe is used, the conversation is processed by third-party software. Questions abound: Where is this data stored? Is it on- or offshore? If data leaves the country, it may be subject to foreign laws, creating risks for patient data security. Some providers have been criticised for a lack of transparency, marketing their tools as privacy-compliant without offering clear details on how data is handled. This lack of clarity can erode patient trust, especially if they are not properly informed.
Accuracy, Errors, and Who Is Responsible
Beyond privacy, there is the critical question of accuracy. AI models can make mistakes. They can misinterpret accents, omit critical details, or even 'hallucinate'—that is, invent information that was never said. In a medical context, even a small error can have serious consequences for patient safety. This raises a thorny legal and ethical question: if an AI-generated note contains a mistake that leads to harm, who is liable? The RACGP has made it clear that GPs remain fully responsible for the accuracy of patient records, even if they are generated by an AI. This means every AI-generated note must be carefully reviewed by the clinician, a step which itself takes time.
The Regulatory Catch-Up
Currently, the oversight of AI scribes in Australia is fragmented, handled by a patchwork of regulators including the Therapeutic Goods Administration (TGA) and the Office of the Australian Information Commissioner. As it stands, most AI scribes are not considered 'medical devices' by the TGA because they don't offer diagnostic advice, placing them in a regulatory grey area. However, this is under review. Health authorities have warned that these tools have “little oversight,” and there are growing calls to establish clear safeguards. Patients are also caught in the middle, with some reporting they have been told to find another doctor if they do not consent to the use of an AI scribe.
