Busy corporate office silhouettes against a cityscape.

Australia is ageing: How AI will help fill the workforce gap 

Our Thinking | Insight

Published

Author

7 Minute Read

RELATED TOPICS

Share Insight

Idea In Brief

Ageing squeezes the workforce from both sides

More care and public services will be needed just as the labour that also builds, teaches, and runs the economy grows more slowly than the population.

AI is a workforce multiplier, not a cure for demographic change

By removing administrative work that doesn't need scarce human effort, it can free clinicians and carers for the presence, judgement and relationships that only people provide.

Australia's real advantage is time, but only if it uses it

Gains depend on spreading AI beyond large firms and investing in skills and work redesign before shortages force change under pressure.

Australia’s ageing challenge is that more people will require health and care services while relatively fewer workers will be available to provide those services and everything else the economy needs. 

Treasury projects a slower-growing and older population over the coming decades. Migration and increased workforce participation will remain important, but neither will remove the underlying shift. Australia will need to meet rising demand for care and for public infrastructure and services from a workforce growing more slowly than the population it supports. 

That makes AI relevant to ageing in a specific and practical way. AI can act as a workforce multiplier: removing work that does not require scarce human effort and augmenting the work that does, so workers can produce more without simply increasing workload. 

The objective is to preserve human effort for work requiring presence, judgement, creativity, relationships, and trust. And the biggest test will be the care economy, where demand will rise precisely as labour becomes scarcer. 

Ageing creates a workforce problem

As the share of Australians of traditional working age declines, workforce growth will slow relative to demand. Maintaining improvements in living standards will therefore depend increasingly on participation, longer and healthier working lives, capital investment, and productivity. 

Ageing affects both sides of this equation. It changes demand by increasing the need for health, aged care, accessible housing, transport and services that support independent living. At the same time, it constrains the supply of labour available to deliver those services. 

The problem extends well beyond the care sector. The workers needed in health and aged care come from the same labour force needed to build housing, teach children, maintain infrastructure, operate businesses, and deliver public services. As labour becomes relatively scarcer, employers will be less able to respond to every new demand by adding employees. 

This makes productivity concrete. It is the practical task of producing more value from each hour of scarce human effort while maintaining high standards of safety and dignity. 

Questions for economic and workforce policymakers:

  • Where will ageing create the largest gaps between service demand and available labour? 
  • Which forms of work most require human presence and judgement, and which supporting tasks could be redesigned or automated? 

The care economy is the hardest test 

Care brings the workforce problem into sharpest focus. As Australia ages, demand for health and care services will rise. These services are labour intensive, and many rely on personal presence and professional judgement within sustained relationships. At the same time, the workers who deliver care are drawn from the same constrained workforce as every other sector. This will require us to distinguish between work that inherently requires a person to perform it and administrative or supporting activities that don’t. 

Clinicians and carers shouldn’t spend scarce time repeatedly entering the same information, navigating disconnected systems, searching for records, preparing routine documentation, or resolving avoidable scheduling problems. Used well, AI could reduce these burdens, support continuity of care, identify risks earlier, and make specialist knowledge more accessible. 

That could allow a clinician to spend more time with patients or a care worker to focus on relationships and observation. I could allow service managers to direct limited staff towards people with the greatest need. 

Used poorly, AI could create additional alerts, opaque decisions, intrusive surveillance, and new administrative work. It could also weaken care if organisations treat technology as a substitute for adequate staffing or human contact. Human accountability and the quality of the care relationship must remain central. The objective is to stop consuming carers’ time on work that doesn’t require a carer. 

Questions for health and care leaders:

  • Which activities are drawing clinicians and carers away from work that genuinely requires their expertise and presence? 
  • Could those activities be removed or simplified, or supported by AI, without weakening accountability or the safety and quality of care?

AI can make scarce labour go further

The Productivity Commission argues that AI could lift productivity, particularly in services, but that gains will depend on complementary investment in skills and in management and business processes. That is the right frame. 

AI can automate routine cognitive and administrative work and augment workers with information and decision support, opening the way to different service and business models. The benefit will come from redesigning work around the software. 

In health and aged care, AI-supported clinical documentation, rostering, decision support and remote monitoring could reduce administrative load and support earlier intervention, giving clinicians and carers more time with people. 

In construction and infrastructure, generative design, automated quantity take-offs, digital twins, machine vision and predictive maintenance could help a constrained workforce deliver and maintain more assets. 

In education and training, adaptive learning and AI-supported course design could help workers acquire new skills over longer and less linear careers. 

In government, AI could reduce routine administrative work and help staff search legislation and records to identify cases requiring closer attention. This matters because the public sector will also face workforce constraints while responding to greater demand for health and care services. 

The same opportunity exists across smaller businesses and community organisations. AI-assisted bookkeeping, customer support, compliance and inventory management could make capabilities previously affordable only to larger organisations more widely available. 

These gains are not guaranteed. If AI simply produces more correspondence, reports, alerts or layers of checking, it may increase activity without increasing value. Organisations need to measure capacity released and the quality and outcomes of services, as well as whether an individual task has become faster. 

Diffusion will matter as much as invention 

AI will not relieve labour scarcity at scale if its benefits remain concentrated in large firms and technically advanced organisations. 

Smaller businesses, community organisations, regional service providers and many care providers have less capital and weaker data foundations, which limits their capacity to redesign work. Without deliberate support, the organisations under the greatest workforce pressure may be among the least able to adopt effectively. 

Older and experienced workers must also be included. They often hold deep occupational knowledge but may receive less access to training or be given tools designed without their needs in mind. Poorly designed implementation could therefore displace valuable experience that AI should augment. 

Australia should support implementation and diffusion alongside research and pilots. That includes investment in data foundations, management capability, workforce training and practical work redesign. 

Technology that is available but difficult to adopt will not increase productivity across the economy. Nor will technology that increases the burden of implementation faster than it releases capacity. 

Japan shows both the opportunity and the limitation 

Japan provides the clearest precursor for Australia. In 2023, 29 per cent of Japan’s population was aged 65 or over. IMF research finds that ageing is contributing to labour shortages and is negatively associated with labour-productivity growth. It also finds that Japanese workers have lower exposure to AI than workers in several other advanced economies, limiting the technology’s present capacity to relieve those shortages. 

Japan matters because persistent labour scarcity changes the purpose of innovation. An ageing society must start by asking which work we most need people to continue doing, and how technology can help them do more of it. 

Japan also demonstrates that technology alone is insufficient. Differences between the skills workers have and those required in emerging roles can prevent labour from moving into areas of need. Management capability, workforce mobility, training and organisational redesign must therefore develop alongside AI adoption. 

Australia is not yet as old as Japan and has greater capacity to use migration. That gives us more room to adapt, and the available lead time should be used. 

Questions for Australian decision-makers:

  • What can Australia learn from countries already experiencing persistent labour scarcity? 
  • Which investments in service management and redesign, and in skills, need to begin before shortages become acute? 

Australia should act before labour scarcity becomes acute 

Japan is not alone. South Korea faces exceptionally low fertility, while China is moving from a demographic dividend towards an older population and slower workforce growth. Across the region, AI-enabled productivity may offset part of the demographic pressure, but only where adoption is broad and complementary reforms bring skills and investment forward. 

Australia has more scope to use migration than many ageing economies. But migration cannot, by itself, prevent population ageing. Migration, workforce participation, healthy ageing, later retirement, family policy, automation and service redesign will need to work together. 

The important advantage Australia has is time. Infrastructure and workforce capability take years to build, and institutions take years to change. Organisations also need time to test new approaches and redesign work, and to build trust with workers and the public. 

Waiting until shortages become acute would force change under pressure. Acting earlier allows Australia to determine where AI can safely release capacity and where human work must be protected, along with the complementary investments required. 

A practical workforce agenda for Australia 

Australia should treat AI as one component of a broader demographic adaptation strategy. 

The first priority should be sectors where future demand and labour scarcity are likely to coincide, particularly health, aged care, disability support, education, construction, logistics and public administration. 

Within those sectors, governments and employers should identify the work that most requires human expertise and relationships built on presence. They should then redesign the surrounding administrative and supporting work to protect that scarce human capacity. 

Funding should extend beyond research and pilots to implementation and diffusion. Smaller organisations will need assistance to improve data, strengthen capability, redesign processes and train workers. 

Governments and organisations should measure the capacity released and the outcomes improved. A faster task is not enough if the time saved is absorbed by additional low-value activity. Measures should include workforce time returned to core work, service access, quality, safety and outcomes. 

Australia should also invest in labour mobility and lifelong learning. Tools and training must be designed from the outset to include experienced and older workers. 

Safeguards will be essential where AI affects eligibility, care, income, safety or access to essential services. People must retain access to explanation and correction through human review, as well as non-digital service pathways where required. 

Finally, governments need to plan the enabling systems early. Energy, compute, cyber resilience, telecommunications, trusted data and workforce pipelines will determine how widely AI can be adopted and whether its productivity benefits are shared. 

The workforce we have will need to do more

Australia can continue to become larger and more prosperous as it ages. But continued improvements in living standards will depend increasingly on how effectively we use a workforce that is growing more slowly than the demand placed upon it. 

AI will not reverse demographic change, and it should complement migration, participation, healthy ageing, later retirement and investment in people. 

Its contribution is more specific and more credible. AI can help Australia stretch scarce labour and expertise further, and protect human effort for work that genuinely requires it. 

This matters most in the care economy. Australia will need more care at the same time as workers become relatively scarcer across the economy. Every hour of a clinician’s or carer’s time consumed by unnecessary administration is an hour unavailable for work that requires human judgement and presence. 

Countries further along the ageing path show us what structural labour scarcity looks like. Australia should use that lead time to build the capability to redesign work and spread effective adoption before shortages make change unavoidable. 

The central question is whether AI can help the workforce we have provide the services, including care, and the prosperity an older Australia will need. 

Get in touch to discuss how AI can help your organisation redesign work and make scarce labour go further.

Connect with David Hallinan on LinkedIn.

AI tools may have been used to support aspects of this publication. The insights and judgement represented in this article are those of Nous, with all outputs reviewed by Nous consultants and editors to quality assure the analysis, references, and language used.