Melbourne's 847 Aged Care Facilities Serve 62,000 Residents. Most Still Run on Paper.

Maria lives in Brisbane. Her mother, Rosa, is in a residential aged care facility in Kew. Rosa has early-stage dementia and a heart condition that requires daily medication monitoring. Before AI, Maria's weekly routine included three or four phone calls to the facility, each one an exercise in frustration. She would call during lunch and be told the nurse was doing rounds. She would call at 3pm and get voicemail. She would send emails that took days to receive a response. On the occasions she did get through, the information was vague: “Your mum had a good day” without any specifics about medication, meals, or mood.

The anxiety was constant. Maria found herself checking her phone compulsively, half-expecting a call with bad news. She took a week off work to fly to Melbourne for a care plan review meeting that lasted forty minutes. She spent more time at the airport than in the meeting. It was unsustainable, and the guilt of not being physically present was affecting her own health.

Now, the AI sends Maria a daily summary at 7pm Brisbane time. It tells her that Rosa took all three medications today, ate 80% of her lunch, joined the morning art therapy session, and had a 15-minute conversation with another resident in the garden. When Rosa had a minor fall last Tuesday, Maria received an alert within 4 minutes — before the incident report was even filed. She could see exactly what happened, what the clinical response was, and that Rosa was assessed as uninjured. For the first time in two years, Maria sleeps through the night.

David manages a 90-bed residential aged care facility in Heidelberg. He has been in aged care for eighteen years and loves the work — or at least, he loved what the work used to be. These days, he arrives at 6:30am to review overnight incident reports, spends most of the morning on AN-ACC documentation, fields phone calls from families who cannot reach nursing staff, attends a quality meeting at 2pm, and leaves at 7pm having not spoken to a single resident. His clinical background is in wound care. He has not assessed a wound in months.

The Royal Commission reforms brought necessary change, but the documentation requirements have been implemented with blunt-force intensity. Every interaction needs recording. Every observation needs documenting. Every concern needs a paper trail that can survive an unannounced audit. David's registered nurses are spending more time typing than caring. His personal care workers, many of whom have limited digital literacy, struggle with the electronic systems and resort to scribbling notes on paper that need to be transcribed later — if they are transcribed at all.

Since deploying AI, David's facility has automated 60% of its routine documentation. Medication administration records are generated automatically from the dispensing system. Care observations are captured through voice-to-text during rounds, transcribed, categorised, and filed by the AI without nurses touching a keyboard. AN-ACC assessments are pre-populated with data the AI has already collected from daily interactions. David's last unannounced audit resulted in zero non-conformances — a first in the facility's history. He assessed three wounds last week. He remembered why he got into this work.

Mr Nguyen is 86 years old. He arrived in Melbourne from Vietnam in 1982 and worked as a carpenter in Springvale for thirty years. He speaks conversational English for everyday interactions, but when it comes to understanding his medical conditions, medications, and care plans, he needs Vietnamese. His wife passed away three years ago, and his three adult children share the responsibility of overseeing his care at a residential facility in Noble Park. One son lives in Springvale, one daughter lives in Reservoir, and one son lives in Ho Chi Minh City.

Before AI, the family struggled with a communication gap that was both linguistic and logistical. Mr Nguyen would nod along during English-language medical consultations without fully understanding his treatment changes. His children would call the facility and navigate conversations with staff who spoke no Vietnamese, relying on his Springvale-based son to translate over the phone — a process that was slow, error-prone, and exhausting for everyone involved. The son in Vietnam received updates weeks late, filtered through multiple retelling, and lived with constant worry that something important was being lost in translation.

The AI now conducts Mr Nguyen's daily welfare checks in Vietnamese. It asks about his pain levels, sleep quality, and appetite using language that feels natural and respectful, with the correct Vietnamese honorifics for an elderly man. When his cardiologist adjusted his blood pressure medication, the AI explained the change to Mr Nguyen in Vietnamese and sent a detailed summary to all three children — in Vietnamese for the son in Ho Chi Minh City, in English for the daughter in Reservoir, and a mixed Vietnamese-English version for the son in Springvale who prefers both. The family, for the first time, is on the same page.