There is a comfortable story that gets told about Australia and artificial intelligence: that we are laggards, sceptics, a nation dragging its heels while the rest of the world races ahead. A recent piece in SMBtech pushes back on that framing, and it is worth taking seriously. The argument is not that Australians hate AI. It is that we are not yet convinced, and those are very different things.
The distinction matters because it changes the policy question. Hostility calls for persuasion campaigns and pushback. Hesitation calls for evidence, guardrails and a reason to trust. On the current numbers, Australia is squarely in the second camp: curious, increasingly exposed to the technology at work and at home, but unwilling to hand over the benefit of the doubt.
The trust gap is real, and it is measurable
The clearest read on national sentiment comes from the trust research led by the University of Melbourne with KPMG, whose global studies have repeatedly placed Australia among the more sceptical advanced economies when it comes to trusting AI systems and the organisations that deploy them. Australians report using AI tools in growing numbers, yet a large share say they are wary of the outcomes, worried about accuracy, and unconvinced that adequate safeguards are in place. Reported trust in AI has, if anything, softened even as usage has climbed, the opposite of what you would expect if familiarity bred comfort.
That pattern is the crux of the SMBtech piece. Adoption and trust are moving in different directions. People are trying the tools because they are embedded in the software they already use, not because they have been won over. It is participation without endorsement.
CSIRO’s work through the National AI Centre paints a complementary picture on the business side. Its Responsible AI research has found that many Australian organisations believe they are handling AI responsibly while lacking the practices, governance, testing, human oversight, to back that belief up. The gap between confidence and capability among businesses mirrors the gap between usage and trust among the public. Both are a form of unconvinced.
Two ways to read the same numbers
There are broadly two camps in how this gets interpreted, and both have a point.
The optimists, many in industry, and some inside the research agencies, argue that scepticism is healthy and temporary. A wary public that demands evidence is exactly the public you want before a powerful technology becomes load-bearing across the economy. In this reading, Australia’s caution is not a drag on progress but a quality-control mechanism. Once the tools demonstrably work, once the failures are rare and the safeguards visible, trust will follow adoption rather than lag it. The task is not to change minds but to earn them.
The pessimists counter that hesitation has a cost, and Australia is paying it. If businesses hold back on serious AI investment because leaders are unsure of the rules, the ethics or the return, the country risks falling behind on productivity at precisely the moment the technology is reshaping global competitiveness. Treasury and industry groups have spent the past two years framing AI as one of the few genuine levers on Australia’s stalled productivity growth. From that angle, a nation that is “not convinced yet” is a nation leaving output on the table while others bank it.
Both readings can be true at once. Caution can be prudent and expensive simultaneously. That is what makes this a genuinely hard national question rather than a simple one.
Why this is an Australian story, not a global one
It would be easy to treat AI sentiment as universal, the same anxieties everywhere, but the Australian version has its own texture. We are a services-heavy economy with a large public sector, a concentrated corporate landscape, and a workforce that has watched automation debates play out before. There is also a specific institutional trust dimension: Australians tend to reserve judgement not just on the technology but on the companies and governments deploying it, many of them offshore. When the model was trained elsewhere, the data lives elsewhere and the accountability is unclear, “not convinced” is a rational stance rather than a technophobic one.
Our universities and CSIRO sit at the centre of this. They are among the most trusted institutions in the country, and they are doing much of the domestic research on how AI actually performs and where it fails. That gives Australia an unusual asset: credible, local, non-commercial voices who can either build public confidence or validate public doubt. How they are funded and heard over the next few years will shape whether the trust gap narrows or hardens.
The Australian stakes
For Australian businesses, the practical lesson is that adoption is not the same as buy-in. Rolling out a tool does not mean staff or customers trust it, and treating deployment as the finish line risks a backlash the first time the system gets something important wrong. The organisations that pull ahead are likely to be the ones that invest in the unglamorous scaffolding, governance, transparency, human review, that turns reluctant users into confident ones.
For workers, the wariness is partly self-protective and partly unresolved. Australians are broadly open to AI that assists them and far more sceptical of AI that replaces or surveils them. Where employers land on that line will do more to shift sentiment than any marketing.
What’s next
The policy backdrop is finally catching up. The federal government’s consultation on safe and responsible AI, including proposed mandatory guardrails for high-risk uses, is the clearest signal yet that Canberra reads the public mood as demanding rules before enthusiasm. If those guardrails land well, clear, proportionate, enforceable, they could be the thing that converts hesitation into cautious adoption. If they stall or arrive muddled, the trust gap is likely to persist.
The through-line is straightforward. Australia is not standing in the way of AI. It is waiting to be shown a reason to believe in it, and, for now, keeping its hand on the wallet and its scepticism intact. Whether that is wisdom or a missed decade depends entirely on what the technology, and the people deploying it, do to earn the trust they have not yet been given. You can read the original argument at SMBtech.
Sources: SMBtech, “Australia Doesn’t Hate AI. It’s Just Not Convinced Yet.”
