Doctors’ soaring use of AI scribes prompts Australian government warning over privacy | Health

The federal health ministry has expressed concerns about the use of AI printers by doctors as the health regulator considers the need for safety measures regarding the technology.

AI writing tools record, transcribe and summarize conversations between doctors and patients for medical notes and have exploded in popularity in the last 18 months.

The use of AI printers by doctors in Australia has almost doubled from 22% in August 2024 to 40% in November 2025, according to an online survey from the Royal Australian College of General Practitioners (RACGP).

Companies offering the technology to practitioners say the technology has been used hundreds of millions of times around the world in the past 18 months alone, as doctors seek to ease the administrative burden of patient consultations.

The federal health department has identified concerns about the increased use of technology. Predicts Senate briefing in February 2026 Documents obtained by Guardian Australia Under freedom of information laws, the ministry said AI authors “have little control” and expressed concerns about the use of such tools in healthcare.

Digital printers are only classified as “medical devices” if they serve a therapeutic purpose and are therefore regulated by the Therapeutic Goods Act.

“Anecdotally, some are marketed outside regulatory tools (e.g., not a medical device) or privacy-compliant, often with limited transparency,” the ministry said. “Some suppliers may be unaware that their cloud platforms are sending data outside of Australia, which could pose a risk to the security of patient data.”

The ministry said some providers were advertising a 30% increase in income for healthcare professionals without additional working hours or patient visits. [Medicare Benefits Scheme] costs”.

AI writers can increase clinicians’ productivity and potentially reduce their burnout, but they are subject to the same limitations in terms of quality and accuracy as other large language models, the department’s AI advisory group said in an April briefing paper.

“This has implications for patient safety, clinical accountability and the integrity of data held in the national digital health infrastructure.”

The department also noted that there are significant differences in how clinicians and practices obtain consent from patients for the use of printers.

“Our position is that informed consent requires consumers to understand the benefits and limitations of the technology to which they are consenting.”

Dr. general manager of the Consumer Health Forum. Elizabeth Deveny said the documents show consumers and the government are raising the same questions about the technology.

“It is good that the ministry has identified these as key issues. The next question is whether the existing measures are strong enough and consistent enough in daily practice,” he said.

“If [scribes] “Free up clinicians time…the public deserves to know whether that time means better care, better access, or just more billable activities, which doesn’t help us with the burnout problem.”

Guardian Australia has previously reported concerns from patients that doctors are not asking for enough consent when using AI tools. A psychiatrist in Melbourne refused to accept patients who did not agree to the use of an AI printer.

Deveny said the forum is increasingly hearing from patients who are told they will need to find a different provider if they do not consent to the use of AI printers.

Oversight of AI printers in Australia is carried out by various regulators, including the Therapeutic Goods Administration, the Australian Health Practitioner Regulation Agency (Ahpra) and the Office of the Australian Information Commissioner.

A health department spokesman said the TGA was reviewing digital printers in Australia in part to determine whether they should be classed as medical devices and subject to oversight, and the report would be published in the coming months.

“The TGA is working with industry associations and individual companies to ensure the regulatory requirements are understood and we plan to publish a summary of the findings from the review in the coming months,” the spokesman said.

Privacy commissioner Carly Kind said in a speech in May that her office was “closely monitoring” the rollout of AI typing technology, meeting with the RACGP ethics committee and typing technology providers.

“We continue to engage with non-governmental organizations concerned about the rollout of the printers, shortcomings in GPs’ implementation of AI printer approval protocols, and lack of disclosure about the printers in their privacy policies,” he said.