Australia's AI Report Card:
Better Than Expected, Which Is to Say, Still Concerning

Every quarter, the Australian Government's National AI Centre publishes its AI Adoption Tracker — a survey of 400 small and medium businesses per month, designed to give policymakers, researchers, and the occasional curious small business owner a clear picture of how AI adoption is actually progressing across the country. The Q4 2024 results are now in, and the headline is broadly positive. AI adoption among Australian SMEs is growing. More businesses are engaging with the technology. Fewer businesses have no idea what it is.

This is, unambiguously, good news. It is the kind of good news, however, that becomes considerably less comfortable when you look at what is still not happening — and at the particular reasons why.

The Numbers, Starting With the Encouraging Ones

Forty percent of Australian SMEs were actively adopting AI in Q4 2024, up from 35% the previous quarter. That five-percentage-point increase over three months is, by the standards of meaningful structural change in a diverse economy of hundreds of thousands of businesses, genuinely impressive. Something is moving.

Source: National AI Centre AI Adoption Tracker, Q4 2024 — Department of Industry, Science and Resources

The gains were particularly notable among the very smallest businesses — those with up to four employees — where adoption jumped from 25% to 34% in a single quarter. This matters because the very smallest businesses are typically the last to move on anything that involves technology, investment, or change to established routines. If the one-person plumbing business and the two-person bookkeeping firm are starting to take AI seriously, the message is getting through.

Also encouraging: the proportion of businesses that are simply unaware of how to use AI fell by two percentage points to 21%. A quarter of the SME population being unaware of how to use a technology that has been on the front page of every newspaper for two years is still, if we are being honest, a sobering figure. But the direction of travel is correct.

Now, the Less Comfortable Part

Thirty-eight percent of Australian SMEs say they have no intention of implementing AI in the next twelve months. This figure has fallen by four points from the previous quarter, which is progress. It is also still 38% of the small business community looking at one of the most significant technological shifts in decades and concluding, collectively, that they'll pass.

Source: National AI Centre AI Adoption Tracker, Q4 2024

It is tempting to be uncharitable about this — to suggest that 38% of small business owners are burying their heads in sand of their own making. But the Tracker data is more generous, and more useful, than that. The three barriers most commonly cited by businesses not yet adopting AI are skills gaps, funding constraints, and the sheer speed of technological change. These are not excuses. They are real obstacles, and they deserve real responses.

What Australian Businesses Are Actually Using AI For

Among the businesses that are adopting AI, the Q4 data shows generative AI assistants moved to the top of the most-used applications list — tools like ChatGPT, Microsoft Copilot, and their various counterparts. This is consistent with what we see working with small businesses every day: the first AI tool most people actually use is a conversational assistant, because the barrier to entry is low and the immediate usefulness is obvious.

Retail, trade, and hospitality businesses led the way in marketing automation — using AI to handle repetitive marketing tasks that have traditionally consumed disproportionate time relative to their impact. This makes sense. A retailer who automates their review requests, appointment reminders, and social media posting is not doing anything exotic. They are simply reclaiming hours they were previously spending on things a well-configured system can handle more reliably than a tired human at the end of a long day.

That scepticism about revenue is worth examining. It may reflect genuine caution from businesses that have been burned by overpromised technology before — and there is no shortage of candidates for that particular hall of fame. But it may also reflect a failure to connect the dots between the operational improvements AI actually delivers and the revenue consequences that follow from them. A business that answers every call, follows up every lead, sends every invoice on time, and maintains a visible presence on social media is almost certainly generating more revenue than one that does not. The mechanism just involves several steps rather than one obvious lever.

The Responsible AI Problem Nobody Talks About at Dinner Parties

Alongside the Adoption Tracker, the National AI Centre's Responsible AI Index 2024 — produced by Fifth Quadrant — added a dimension to the picture that is perhaps the most striking finding of all.

That gap — between what 78% of businesses believe and what 29% of businesses are actually doing — is extraordinary. It is the kind of gap that emerges when people feel confident about something they have not examined closely. It is, if you will allow a gentle observation, the precise opposite of the way confidence is supposed to work.

The businesses that are leading on responsible AI adoption share one characteristic: they have business leaders — CEOs, board members, owners — who are actively driving the AI strategy. Not delegating it to the IT department and hoping for the best. Not approving a budget line and considering the matter closed. Actually understanding what is being adopted, why, and what guardrails are in place.

What the Agriculture Sector Would Like You to Know

The Tracker data breaks down AI adoption by industry, and the contrast between sectors is striking. Health, education, and manufacturing led uptake at 45% — all sectors where the productivity case for AI is clear, the regulatory environment is pushing for better data management, and the workforce is generally accustomed to adopting new tools.

Agriculture, by contrast, came in at 6%. Six percent. This is not because farming is immune to the benefits of better data, automation, or predictive analytics — the applications in crop management, supply chain optimisation, and equipment maintenance are well-documented and already deployed at scale in other countries. It is, almost certainly, because the agricultural sector is geographically dispersed, access to quality training and support is inconsistent, and the immediate daily pressures of running a farm leave limited bandwidth for technology adoption projects.

This is a services and support problem, not a technology problem. The tools exist. What is missing is accessible, practical help in using them — and that is a gap the market has every reason to fill.

The Reasonable Conclusion

The Q4 2024 AI Adoption Tracker tells the story of a country in the early stages of a genuine transition. The headline numbers are moving in the right direction. Adoption is up, awareness is up, and the government is investing in the infrastructure — including a network of AI Adopt centres specifically designed to support SMEs through the process.

What the data also shows, clearly and consistently, is that the main barriers are not technological. Australian small businesses are not failing to adopt AI because the tools are too complicated or too expensive or insufficiently capable. They are failing to adopt — or adopting poorly — because of skills gaps, because of the absence of accessible, practical guidance, and because the pace of change is genuinely difficult to keep up with when you are also trying to run a business.

The solution to a skills gap is, with magnificent simplicity, skills. The solution to a guidance gap is guidance. These are not technically challenging problems. They are organisational and human problems, and they have the great advantage of being entirely solvable — one business, one team, one practical training session at a time.