More than 150 of Australia’s innovation experts gathered in Canberra last week for the annual National Innovation Policy Forum. Leaders from business, research and boundary-spanning entities, policymakers and parliamentarians were there, looking at how best to address the significant challenges facing local R&D.
ARM Hub founder and CEO, Professor Cori Stewart, led one of the roundtable discussions. She said the numbers were stark: Australia spends just 1.68 per cent of GDP on R&D compared to the OECD average of 2.7 per cent.
While other countries climbed from 2.3 to 2.7 per cent since 2008, we dropped from 2.25 to 1.68 per cent. We rank 105 out of 135 countries for commercialising research.
“At this time, we have industries starting to understand their troubles, and we have AI that can actually support SMEs as much as global corporations,” Stewart said.
“If we’re smart, we can do it. But if we don’t, there’s no Plan B.”
The SME advantage
The good news is Australian manufacturers can adopt AI faster than large corporations. SMEs lack the legacy systems that slow bigger competitors. With rising costs in electricity, wages and insurance squeezing margins across sectors, AI offers practical efficiency gains that can improve profitability quickly.
The technology is accessible. The tools are available.
But the barrier isn’t capability: it’s approach.
Back to the future
AI adoption sounds complex, but it isn’t. Adoption simply requires matching technology to business problems, then involving the people around those processes.
This isn’t revolutionary thinking; it’s how Australian businesses successfully adopted computer-based systems in the 1980s and 90s.
When office suites transformed businesses, replacing typewriters and paper ledgers, the technology wasn’t the hard part. Getting people to change how they worked was the challenge.
Once the focus shifted from technology, people were free to actually connect the new tech to on-the-ground problems.
1989’s Technology Acceptance Model showed that 40-60 per cent of technology acceptance depends on user perceptions, while the People, Process, Technology framework formalised in the 1980s advised the same focus.
The results of this people-first approach were transformative across all businesses. By 1994, more than 60 per cent of Microsoft Word sales and 70 per cent of Excel sales were bundled, because businesses recognised the advantage. And companies that adopted early gained ground their competitors struggled to match.
The most recent research shows the same pattern applies to AI. Successful AI adoption requires 70 per cent focus on people and processes, 20 per cent on technology and only 10 per cent on the algorithms themselves.
Yet 74 per cent of companies struggle to scale AI value because they focus on algorithms instead of people.
People First
AI differs from office software in one crucial way: it may eliminate roles while creating others. This makes the people-focused approach even more critical. Workers need time to build new skills, so companies need clear retraining pathways.
That makes 70/20/10 essential, not optional.
“The good news is we already know how to do this,” Stewart said.
“The challenge isn’t the technology. It’s the same people and process work we mastered when we moved from paper-based systems to computers in the 1980s and 90s.
“When you focus on matching AI to the right business problems and getting your teams ready to use it, adoption becomes straightforward. We can do this. We do do this. We just need to apply what we already know.”
Proof in practice
McKinsey’s latest State of AI report found that high-performing organisations “treat AI as a catalyst to transform their organisations, redesigning workflows and accelerating innovation”.
ARM Hub, one of four government-backed AI Adopt Centres, is demonstrating this approach works:
- Melbourne’s Nexobot is transforming logistics through practical automation; and
- Brisbane-based Microbio is detecting sepsis in under three hours, saving lives globally.
Both companies succeeded by treating AI as a tool to solve specific problems, not as technology requiring fundamental transformation.
“We’re not helping companies become AI experts, we’re helping them identify high-impact problems, train their teams, and deploy solutions that prove ROI within months,” she said.
Through Data & AI-as-a-Service, scaleup manufacturing programs and accelerators like Propel-AIR, ARM Hub is building the pathways from research to market that Australia needs.
Australian manufacturers and SMEs can close the innovation gap. They don't need massive investment. They need smart adoption focused on people, processes and practical outcomes.
