By David Swan
Publication Date: 2026-03-23 00:00:00
An AI medical scribe used in Australian clinics has been manipulated by security researchers to go off script and create identity theft guides. However, the misbehaving bot was unable to access patient data.
Mindgard, a US-based cybersecurity company, says a Heidi Health bot used for clinical documentation could be freed from its ethical restrictions in minutes with the right prompts, demonstrating the risks to Australian companies if they quickly deploy AI tools.
Heidi Health said the vulnerability was identified and fixed internally before Mindgard contacted it, and that the compromised tool…
