A new RMIT and Deloitte report puts a dollar figure on low AI literacy in Australia. We unpack the numbers that should have every SME operator paying attention.
What’s happening: A report by RMIT Online and Deloitte Access Economics has found that more than five million Australian workers remain beginner-level AI users, despite 84% of the workforce now using at least one AI tool at work.
Most small business owners using AI are doing the same thing: opening a tool, typing a question, and hoping for something useful. That is not necessarily wrong, but a new report suggests it is leaving a significant amount of value on the table.
Research published this month by RMIT Online and Deloitte Access Economics found that 84% of Australian workers now use at least one AI tool at work, but only 7% have reached an advanced level of AI literacy. More than half, 54%, remain at a beginner level. The report estimates that if half of all beginner-level workers improved to intermediate, the economic dividend would exceed $18.9 billion.
For workers, the personal return is concrete. Moving from beginner to intermediate AI literacy is associated with around $7,000 in additional annual wages for full-time workers, according to Deloitte Access Economics modelling. Moving from beginner to advanced reflects a benefit of almost $11,000. For a small business owner, those figures are less about salary and more about competitive capacity, the difference between someone on your team who gets surface-level results and someone who can use AI to actually change how work gets done.
Nic Cola, CEO of RMIT Online, was direct about the cause. “The research makes it clear that self-guided, ad-hoc experimentation is not enough to move the needle on national productivity,” he said. “We are seeing a landscape of ‘shadow AI’ where nearly half of workers are teaching themselves through trial and error, often building surface-level technical skills while critical judgement lags.”
For most small businesses, trial and error is the only programme on offer.
Saving time, not building skill
The report found that the average worker saves nine hours per week through AI use. That sounds like a significant gain. But nearly half of workers, 47%, are putting that time straight back into completing more tasks. Only one in four, 25%, use the time for further skill development.
In a small business context, that pattern is easy to understand. When you are short-staffed and busy, recovered time gets absorbed immediately. But the report’s findings suggest that the businesses reinvesting even a portion of that time into building sharper AI capability will compound their advantage over time, while those running faster on the same treadmill will not.
The generational dimension adds another layer relevant to business owners. Workers over 55, many of whom hold senior or decision-making roles in small businesses, are far more likely to remain at a beginner level. More than three-quarters, 76%, of Baby Boomers in the survey possess only beginner-level AI literacy, compared with 43% of millennials. The report notes that if Boomers were to close the literacy gap with millennials, it would unlock a collective $3.1 billion in economic value.
The judgement gap nobody’s talking about
There is a split in AI capability that the report highlights and that gets less attention than tool adoption: the difference between technical skills and judgement skills.
Australian workers are twice as likely to be advanced in technical AI skills, such as writing prompts, at 21%, than in critical judgement capabilities, at 11%. Judgement skills include ethical awareness, risk assessment, and the ability to critically evaluate AI outputs. They are slower to develop, require domain expertise, and cannot be picked up through self-experimentation alone.
This imbalance carries real risk for businesses. The report draws on prior research showing that 95% of companies have experienced an AI-related incident in the past two years, with an average financial loss of $800,000. Separately, 56% of workers have made mistakes due to AI use, and more than half admit to presenting AI-generated content as their own without disclosure.
For a small business where one person’s error can affect a client relationship, a contract, or a public-facing document, the judgement gap is not a corporate governance issue. It is a daily operational one.
Rhiannon Yetsenga, Associate Director at Deloitte Access Economics, framed the challenge plainly. “The productivity gains from AI are undeniable, but the focus now needs to shift from mere adoption to competency,” she said. “The real opportunity is equipping workers to use AI effectively, strategically and ethically.”
What SME owners can do now
The report stops short of prescribing a single solution, but its conclusions point in a clear direction. Structured training, not informal tinkering, is what separates businesses capturing AI’s value from those moving fast but going sideways. Only 48% of workers nationally receive any AI training from their employer. Just 11% receive structured, ongoing support.
For small business owners without a dedicated training function, the entry point does not need to be elaborate. The report recommends targeting critical evaluation and practical use cases first, establishing clear guidelines for how AI can and cannot be used in the business, and tailoring any training to different levels of experience and confidence across the team.
The federal government’s AI Adopt Program has established AI Adopt Centres open to eligible small and medium-sized enterprises, designed to provide guidance, specialist training and access to expertise for businesses looking to adopt AI responsibly. It is one practical starting point for owners who want structured support rather than another tab to figure out alone.
The nine hours saved each week is real. What happens with it is the question that will separate businesses in the years ahead. More information is in the full report on this link
Keep up to date with our stories on LinkedIn, Twitter, Facebook and Instagram.
