AI in Australian Manufacturing: Closing the Readiness Gap

**This story originally appeared in the July 2026 edition of AMT Magazine**

The most successful uses of AI are unlikely to come from technology alone. They will depend on whether manufacturers have the right data, skills, governance, infrastructure and business discipline to apply AI where it solves real operational problems.

The opportunity is significant. The CSIRO’s Artificial Intelligence Roadmap estimates that digital technologies, including AI, could increase productivity by up to 40% and contribute $315 billion to Australia’s GDP by 2030. The Australian Government’s Department of Industry, Science and Resources also identifies AI as a critical technology with potential to transform existing industries and build new ones.

For manufacturing, however, the value will not come from AI as a standalone tool. It will come from combining AI with existing industrial strengths: precision engineering, automation, robotics, quality systems, industrial software, connected equipment, advanced materials and process control. AI’s role is not to replace manufacturing know-how, but to make that know-how more responsive, data-driven and scalable.

Investment is Growing

The investment signals are already visible. According to the Australian Bureau of Statistics, business expenditure on R&D grew 18% to $24.4 billion in 2023-24. AI was the fastest-growing area, with businesses investing $668.3 million in AI R&D, more than double the $276.3 million recorded in 2021-22. Manufacturing remained the second-largest industry for business R&D expenditure, accounting for $5.0 billion, or 21% of total business R&D.

But the same data also points to tension. ABS figures show R&D spending directed at manufacturing outcomes fell 3% to $4.6 billion, continuing a downward trend since 2019-20. This suggests Australian manufacturing is investing in a period of pressure, not abundance. The business case for AI therefore needs to be practical: improved uptime, better yield, less waste, stronger forecasting, safer work and better use of scarce labour.

Global manufacturing research reinforces that point. KPMG’s intelligent manufacturing research found that 93% of respondents believe organisations that embrace AI will develop a competitive edge over those that do not. It also found that 72% intend to use AI to improve efficiency, 77% to drive growth, 96% have experienced operational and efficiency improvements, and 62% have seen ROI greater than 10%.

Uneven Adoption

Yet adoption is still uneven. The National AI Centre’s AI Adoption Tracker found that 41% of Australian SMEs were adopting AI. Among adopters, 22% reported improvements in decision-making speed and 18% reported productivity optimisation. In tandem, the share of businesses unaware of how to use AI had fallen to 21%, suggesting awareness is improving but capability gaps remain.

This is where the Australian manufacturing context becomes important. Many manufacturers already hold valuable data in ERP systems, production records, maintenance logs, quality systems, machine controllers, inspection records and supplier data. But that data is often fragmented, inconsistent or difficult to access. For AI to be useful, manufacturers need to understand what data they have, who owns it, how reliable it is, and whether it can be connected across the business.

The Reserve Bank of Australia has observed that many firms have been investing in cloud computing and data infrastructure as foundational steps for modernisation and future AI adoption. It also notes that firms expect investment in AI, machine learning, robotics and automation to be much higher over the next three years than it has been previously.

However, the RBA also highlights an important caution: many companies are still in an adjustment phase. Technology investment often begins with risk management, system upgrades and operational resilience, rather than immediate productivity gains. It also notes that AI adoption has often been piecemeal and employee-led, with many firms still seeking high-impact use cases. For established manufacturing businesses, the path to productivity may involve a “J-curve”, where short-term disruption and process adjustment come before longer-term gains.

That is a useful warning for manufacturers. AI should not be treated as a quick software purchase. It requires changes to workflows, training, data management, accountability and decision-making. A manufacturer may need to modernise legacy systems, clean data, document processes, train supervisors and involve frontline workers before the technology can deliver value.

The Challenges of Skills and Governance

Skills are a major part of the challenge. The CSIRO estimates that Australia will require as many as 161,000 people with specialist AI skills by 2030. It also warned of sovereign risks if sensitive sectors rely too heavily on AI technologies developed and controlled offshore.

AI adoption will intensify those pressures because manufacturers will need people who understand both production and digital systems. The most valuable capability may not be pure data science, but hybrid skills: tradespeople, engineers, operators and managers who can work with data, automation and AI-enabled tools.

Governance is another readiness issue. The Responsible AI Index 2025, commissioned by the National AI Centre, found that 79% of Australian businesses believed they were implementing AI safely and responsibly, but only 29% were assessed as doing so. On average, organisations were adopting only 12 of 38 responsible AI practices across areas including accountability, safety, fairness, transparency and contestability.

For manufacturers, that gap is not theoretical. AI may touch customer IP, product quality, worker safety, cybersecurity, procurement, production data or regulated obligations. A model used to support inspection, quoting, scheduling or maintenance needs clear ownership, validation and review. Responsible AI should therefore sit alongside existing systems for quality, safety, risk, cybersecurity and continuous improvement.

Government Policy and Research Infrastructure

Government policy is increasingly focused on closing that gap. The Australian Government’s response includes the $22.7 billion Future Made in Australia Plan, a $1.7 billion Future Made in Australia Innovation Fund, the $15 billion National Reconstruction Fund, $523.2 million for the Battery Breakthrough initiative, and investment in skills, apprenticeships and Industry 4.0 capability.

The policy direction is not only about AI, but AI is clearly part of the industrial strategy. The government response also highlights the National Robotics Strategy, released to harness robotics and automation and help revive Australian manufacturing. The strategy itself sets a vision for Australian industries to responsibly develop and use robotics and automation technologies to strengthen competitiveness, boost productivity and support local communities.

There are also more targeted measures. The AI Adopt Program provides grants of $3 million to $5 million over four years to establish AI Adopt Centres that help SMEs adopt responsible AI-enabled services. The program is designed to create a “front door” for SMEs, support real-world applications, build workforce skills and improve productivity in National Reconstruction Fund priority sectors.

Research infrastructure is another part of the picture. In March 2025, the Australian Research Council launched the ARC Industrial Transformation Research Hub for Future Digital Manufacturing, led by Swinburne University. The $5 million hub will focus on AI and Internet of Things technologies, including digital twins, to support more competitive and resilient manufacturing. Swinburne estimates that Australian manufacturing productivity and resilience could rise by as much as 30% through the hub’s work.

The ARC has said the hub will support the transition to smarter, more connected systems, including AI-powered tools for texture-modified food and advanced structural health monitoring and predictive maintenance for aerospace. It also links the hub’s work to the Future Made in Australia agenda and the need to build skills for Industry 4.0.

AI Adoption and Execution

Australia’s AI in manufacturing landscape is developing quickly, but the next phase will depend on execution. Investment is rising, research capability is growing and policy support is strengthening. The real test will be whether manufacturers can connect those opportunities to the systems, people and processes that make advanced manufacturing work.

AI will not define the future of manufacturing on its own. The manufacturers that gain the most will be those that treat AI as part of a broader capability agenda: better data, stronger skills, trusted governance and practical problem-solving.