Australia ranks last in the world for trust in AI — and the reason is not that we don't use it. It's that we use it without trusting, training in, or understanding it. That gap is the real story.
- Only 30% of Australians believe AI's benefits outweigh its risks, the lowest of any of the 47 countries studied.
- This is not a usage problem. Around half of Australians use AI regularly; trust is lagging behind use, not the other way round.
- The real gap is confidence and literacy: just 24% of Australians have had any AI training, against a 39% global average.
- For leaders, low national trust is not weather that passes. It is the adoption climate you already lead.
Australia ranks last in the world for trust in AI, and the reason is not the one most leaders assume. It is not that Australians avoid the technology; half of us use it regularly. It is that we use it without trusting it, without training in how it works, and without believing the institutions around it are ready. Australia leads the world in scepticism and trails it in confidence. That gap is the real story, and it is sitting inside your organisation right now.
I get asked whether this is just Australian cynicism. It is something else. It is a coherence problem, a mismatch between head and gut: a country using a tool faster than it has made sense of that tool. And the same pattern plays out in every boardroom where I sit.
Does Australia really rank last in AI trust?
It does. In the 2025 Trust in AI study from the University of Melbourne and KPMG (a global professional-services firm), researchers surveyed more than 48,000 people across 47 countries. Only 30% of Australians said the benefits of AI outweigh its risks, the lowest score of any nation surveyed. Just 36% are willing to trust AI, while 78% are concerned about negative outcomes. We sit near the bottom of the world on acceptance, excitement and optimism.
Read that again. Bottom of forty-seven. This is not a rounding error. It is a national posture.
Is it that Australians simply do not use AI?
The opposite is true, and this is the part that matters for leaders. Around 50% of Australians use AI regularly and roughly two-thirds say their employer uses it. The technology already sits inside the building. What is missing is the human readiness around the tool: only 24% of Australians have undertaken any AI training, against a 39% global average, and more than 60% report low AI knowledge. Trust is lagging behind use. People are operating tools they have not been trained to understand, and then, sensibly, they do not trust the result.
The workplace numbers sharpen the picture. In an August 2025 EY survey of Australian workers, 68% now use AI in their jobs, yet only 35% have received any formal AI training from their employer and 54% say they are not confident using AI at work. Two-thirds want more training than they are given. Use is racing ahead; support is dragging behind.
Leading inside a low-trust market?
The AI Strategy Session turns the trust and readiness gap into a ninety-day plan you can take to your leadership team.
Book your Strategy SessionWhy does national trust matter to my organisation?
Because trust is the substrate of adoption; it is the ground the whole effort stands on, the bedrock beneath every pilot and rollout. The Australian Responsible AI Index, run by research firm Fifth Quadrant with the National AI Centre in 2024, found that 78% of organisations believed they were using AI safely and responsibly, while only 29% actually were when measured against responsible-AI practices. That is a forty-nine-point gap between confidence and reality, and it is exactly the kind of dissonance that erodes trust further the moment something goes wrong.
The encouraging signal: readiness is climbing from a low base. Cisco's 2025 AI Readiness Index found 22% of Australian organisations are now fully ready to deploy AI, up sharply from just 4% the year before. The movement is real. But a fully-ready minority inside a low-trust majority is a leadership challenge, not a technology challenge.
Australia leads the world in AI scepticism and trails it in AI confidence. We are using the tool faster than we have made sense of the tool.
What do leaders actually do with this?
You stop treating trust as a communications problem and start treating it as an operating condition. Three moves, in order:
- Close your own confidence-to-reality gap first. Be honest about how responsibly you actually use AI, not how responsibly you believe you do.
- Invest in literacy before you invest in the next tool, because an untrained team cannot trust what it does not understand.
- Make trust a measured outcome, not a slogan. Track training, confidence and responsible-use practice the way you track any other operating metric.
Trust is not won with a policy. It is built by people who know what they are doing. Low national trust is not weather that passes. It is the climate you already lead. The organisations that win the next decade in this market will be the ones that build trust deliberately, from the inside, starting with the leader.
| Measure (Australia) | Figure |
|---|---|
| Believe AI's benefits outweigh the risks | 30% (lowest of 47 countries) |
| Willing to trust AI | 36% |
| Have had any AI training | 24% (vs 39% global) |
| Workers given formal AI training by their employer | 35% (EY, 2025) |
| Organisations fully AI-ready | 22% (up from 4% in 2024) |
| Believe they use AI responsibly vs actually do | 78% vs 29% |
Frequently asked questions
Does Australia really rank last in AI trust?
Is the problem that Australians do not use AI?
How AI-ready are Australian organisations?
About the author
British technology futurist, AI keynote speaker and advisor. Thirty years across enterprise technology and AI strategy, helping leaders navigate the future of work. The futurist who died.