Back to deakin.edu.au
AI-enabled technologies are already being used in health, but the people most likely to be affected by them are rarely involved in decisions about whether or how they’re used.
The AI in Health Trust Lab exists to help give people with lived and living experience, a genuine, ongoing say on whether AI solutions in health are genuinely worthy of their trust – not to persuade communities to trust them.
Trust has to be earned – through transparency, genuine community engagement and ongoing accountability.
The AI in Health Trust Lab is a community-centred space where different people come together to ask what AI in health should have to answer for. Why was this thought to be needed? Who was include in decisions about design? Should this be trusted? Who does it serve? Who is accountable if it fails? We are deliberately slowing down to scrutinise, not racing to accelerate.
The goal is not more trust in AI-enabled technologies. It is better-placed trust – given deliberately to what has earned it and withheld from what has not.
AI-enabled technologies are being used across health right now – in screening, triage, scheduling, note-taking and decisions about care. Some of the promised solutions are real. Earlier detection, better prediction, more time with a person instead of a screen.
But so are the risks. Systems that don’t work well for the people the health system already underserves. Decisions not clearly explained. Errors in outputs. Lack of transparency around the handling of personal information.
And only 30% believe current laws and safeguards are enough to make its use safe. For First Nations communities and others who have experienced institutional harm, distrust of AI sits alongside a broader and well-founded mistrust of health systems. That’s not a problem to be solved by convincing people. It’s a signal that trust has not been earned yet.
We focus less on whether AI-enabled solutions work in a technical sense, and more on whether the decisions about their design, adoption, implementation and use are safe, fair and accountable to the people they affect.
Leading with the voices of people with lived and living experience, and working with researchers, health services, industry, government and funders, we name the signals of trustworthiness – uncovering the difference between what has been promised and what has been proven.
The AI in Health Trust Lab turns community insight into evidence to inform better practice. We help define what ‘worthy of trust’ looks like.
Established by Health Voices Victoria with funding from Deakin University, the Trust Lab brings community voice into conversations and decisions it does not normally reach.
It creates a space where communities, researchers, health services, government, industry and funders can recognise shared concerns and build evidence together.
The foundations of the Trust Lab are being built through co-creation with an Activation Group, meeting regularly throughout 2026. Members bring a mix of lived and living experience and sector insight – community members, researchers, advocates, people working in health services, policy thinkers, and people who design and build AI enabled solutions.
This dynamic group of thirteen people are co-creating the Trust Lab’s position, the principles that underpin the work, our priorities, and how we will evaluate whether it’s working.
In the second half of 2026, we’re widening who we talk to through our series of Community Conversations. We’ll be talking to people across Victoria and online, far and wide, no matter your stance on AI in health, whether you’re cautious, curious, sceptical or excited. We will be seeking out people who are too often left out or ignored. These conversations will shape where the Trust Lab goes next and sharpen our approach to meet community’s expectations and needs.
Sign up to hear what the Trust Lab is doing, what we’re learning, and when there’s an opportunity to get involved.
January 2026 – Kick-off Health Voices Victoria team gathered key collaborators and set the plan for 2026 and 2027.
April 2026 – Activation Group recruitment More than 110 people applied for around a dozen places – a clear signal of how much interest there is in this work.
May 2026 – Co-creation begins Monthly sessions commenced to shape the Trust Lab’s position, principles, priorities and evaluation framework. Continuing through to November.
August 2026 – Public launch The AI in Health Trust Lab launched on 24 August 2026 at Deakin Downtown.
Second half of 2026 – Community conversations Conversations with community groups across Victoria and online.
Q1 2027 – What we heard Our first Community Insights Report, and the evaluation framework that holds us to account.
Computer systems designed to do things that would usually need human intelligence – recognising images, understanding speech, making decisions. Most modern AI learns by finding patterns in very large amounts of data.
We use this rather than “AI” alone, because in health it is rarely the AI on its own that matters. It is the whole system it sits inside – the product, the process, and the decisions people make with it.
Trust is something you give. Trustworthiness is something a system, person or organiastion has to demonstrate. Our aim is not to persuade you to trust AI – it is to help make AI genuinely worthy of it.
We use this instead of “stakeholder”, a term with colonial roots in the practice of claiming land by driving stakes into the ground. It means everyone affected by this work.
Experience of the health system, and of health conditions, held by people who have been through it – including those who continue to live with it every day. We recognise this as essential expertise in improving health and wellbeing.
We use “lab” in the non-traditional or scientific sense: a collaborative space where diverse people explore issues, develop ideas and test them together – closer to a living lab or policy lab than a clinical one.
No, we have a neutral stance on AI in health. Some of the promise is real. We are against these technologies being adopted without the people they affect having a say, and without evidence that they work for the people they claim to help.
Yes, and we particularly want to work with you. You do not need to understand how AI works to have a view on whether a machine should make a decision about your health, or who should be answerable when it gets it wrong.
We do not hold regulatory power. What we can do is build community-verified evidence of what trustworthy looks like and put it in the hands of the people who make these decisions – and in the hands of communities asking them to do so. That may be a slower route than enforcement. It is one of influence and it is the honest one. We are developing an Evaluation Framework so we can report on our impact.
Looking to work with the AI in Health Trust Lab?
