AI Won’t Make Australia Prosperous Without the Right Economic Policies

Governments are betting on artificial intelligence to revive productivity and lift living standards. But technology alone cannot deliver prosperity. Australia needs an economic strategy that turns AI efficiency into higher real incomes, competitive industries and better public services.

Borrowing Doctor Who’s TARDIS, I travelled back in time and put on my Investment Commissioner’s hat to examine a familiar economic promise in a new technological age: artificial intelligence will make Australia more productive and, therefore, more prosperous.

But that conclusion deserves closer scrutiny.

Productivity is about producing more with the resources we have. Prosperity is about Australians becoming better off as a result. The two are closely related, but they are not interchangeable.

AI may enable businesses to produce more with fewer employees, reduce operating costs and deliver services faster. Yet those gains could be absorbed by lower prices, captured by overseas technology suppliers or concentrated among a relatively small number of businesses and investors. Higher productivity may create the capacity for rising living standards without ensuring that the benefits reach Australian households.

History offers a cautionary lesson. Email, cloud computing and online commerce transformed business efficiency, but widespread adoption also intensified competition and made yesterday’s technological advantages tomorrow’s minimum operating standards.

For governments promoting AI as the next great productivity revolution, the policy challenge is therefore more demanding than accelerating adoption. It is to ensure that the wealth AI helps create translates into higher real incomes, stronger Australian industries and better public services.

AI can make Australia more productive. Whether it makes Australians more prosperous will depend on the economic policies that accompany it.

The productivity promise

Consider an Australian professional-services firm that uses AI to reduce the cost of delivering an assignment from $1,400 to $900.

Initially, the firm may retain the saving through higher margins. But as competitors adopt similar technology, prices may fall, transferring much of the benefit to customers.

Employees may receive higher wages or undertake more valuable work, while technology suppliers collect subscription and usage fees. Some of those suppliers may be based overseas.

The productivity improvement is genuine, but its economic benefits are distributed among businesses, workers, consumers and technology providers.

The same principle applies nationally. Higher productivity can support wages, profits, lower prices and stronger public finances, but the distribution depends on competition, investment, workforce capabilities and the structure of the economy.

This is why governments should distinguish between encouraging AI adoption and developing an AI prosperity strategy. The first focuses on how extensively technology is used. The second asks whether its use creates lasting economic value and improves Australians’ lives.

History offers a cautionary lesson

The relationship between technological progress, productivity and prosperity is not new. Previous waves of digital innovation have delivered substantial efficiency improvements without guaranteeing lasting competitive advantages for the businesses adopting them.

Email dramatically reduced communication costs, but faster responses soon became an ordinary customer expectation rather than a source of competitive advantage. Cloud computing gave smaller businesses access to sophisticated systems previously affordable only to large organisations, yet widespread adoption eroded the advantage of possessing them. Online commerce expanded market access while intensifying competition and price transparency.

Each technology improved economic efficiency, but none guaranteed higher profitability for individual businesses. AI may follow a similar pattern: substantial productivity gains without an automatic or evenly distributed prosperity dividend.

Importantly, this does not mean the earlier technologies failed to contribute to economic prosperity. Consumers benefited from greater convenience, improved services and access to more competitive markets, while businesses developed new products, services and operating models.

The lesson is that technological adoption alone does not determine economic outcomes. The greatest gains often arise when businesses reorganise their operations, develop new capabilities and create products and services that were previously impractical.

AI policy should therefore focus not only on accelerating adoption but also on strengthening the economic conditions that allow productivity improvements to translate into lasting value.

Competition determines who benefits

One of the most important influences on the distribution of AI’s economic gains will be market competition.

Where businesses compete effectively, productivity improvements can translate into lower prices, improved services and greater consumer purchasing power. These benefits represent genuine improvements in living standards, even when they do not produce higher business profits.

However, AI also introduces potential dependencies. Businesses may rely on a small number of providers for models, computing infrastructure, business software and access to customers.

If switching suppliers becomes difficult, technology providers may gain substantial pricing power and capture an increasing share of the value their customers create.

For Australia, this raises a strategic question. How much of the economic benefit generated by Australian businesses will be retained domestically, and how much will flow to international technology suppliers?

Imported technology can deliver enormous benefits, and Australia need not develop every AI system itself. Nevertheless, maintaining competitive markets, encouraging interoperability and reducing unnecessary supplier dependence will be important to ensuring businesses can retain the benefits of their investments.

The policy objective should be to encourage innovation while preserving competition and genuine choice.

From adopting AI to building Australian capabilities

Governments frequently measure technological progress through adoption rates, investment announcements and the number of organisations implementing new systems.

These measures are useful, but they do not establish whether an economy is developing sustainable competitive advantages.

An Australian manufacturer using AI to automate administration may become more efficient. But if international competitors can adopt the same technology, the improvement may simply become necessary to remain competitive.

A more significant opportunity arises when AI is combined with distinctive Australian capabilities.

A mining technology company might use specialised operational data to develop valuable predictive services. An agricultural business could combine AI with local expertise to improve production systems. An engineering consultancy might develop exportable advisory products based on decades of technical experience.

In these examples, AI contributes to productivity while enabling businesses to create differentiated products, intellectual property and new sources of revenue.

This distinction matters because sustainable national prosperity depends not only on producing existing goods and services more efficiently, but also on developing valuable capabilities that international customers are willing to purchase.

Government policy should therefore support research commercialisation, industry collaboration, business transformation and export development, rather than concentrating solely on the purchase of AI tools.

Workers must share in the gains

AI’s contribution to prosperity will also depend on its effect on employment and wages.

Greater productivity creates the economic capacity for higher real wages, but does not guarantee them. The outcome depends on labour-market conditions, bargaining arrangements, skills and how businesses reorganise work.

Some employees will become more valuable because AI enables them to apply their expertise across a wider range of activities. Others may find that routine tasks are automated or their occupations substantially changed.

The challenge for governments is not simply to encourage more AI training. It is to help workers move into activities where their skills remain valuable and productivity improvements support better employment opportunities.

That requires practical industry-based training, accessible mid-career education, recognition of transferable skills and targeted assistance for workers facing significant disruption.

Education and workforce policies should also recognise that AI changes the skills businesses require. Professional judgment, technical expertise, problem-solving and the ability to work effectively with intelligent systems may become increasingly important.

Without effective workforce adjustment, productivity gains could coexist with economic insecurity and widening differences in employment outcomes.

Government must demonstrate productivity itself

AI also offers opportunities to improve public-sector performance, particularly in administration, regulatory processing, information management and service delivery.

But government productivity should not be measured simply by counting automated tasks or employee hours saved.

An agency that processes applications faster has achieved little if its decisions become less accurate, transparent or fair. Similarly, reducing administrative expenditure creates limited public value if the savings are offset by errors, inadequate oversight or poorer services.

The relevant measure is whether AI improves the quality, accessibility and cost-effectiveness of government services.

Public procurement can also encourage competitive markets by supporting interoperable systems, transparent contracting and outcome-based evaluation.

Government should demonstrate the same commercial discipline it expects from businesses: establish a measurable objective, test whether the technology delivers value and expand investment only when the results justify it.

Five priorities for an AI prosperity strategy

Australia’s policy challenge is to convert technological efficiency into enduring economic value. Five priorities deserve particular attention.

• Promote business transformation, not simply AI adoption. Government programs should encourage SMEs to redesign services, enter new markets and develop new revenue streams rather than merely automate existing administration.
• Maintain competitive and accessible AI markets. Competition oversight, interoperability, data portability and responsible procurement can help prevent unnecessary dependence on a small number of technology suppliers.
• Invest in workforce adaptability. Practical training, career mobility and support for occupational transitions are essential if productivity gains are to translate into broader employment opportunities and higher real incomes.
• Build distinctive Australian economic capabilities. Research partnerships, commercialisation and export development should focus on industries where Australian expertise, data and innovation can generate internationally competitive products and services.
• Measure prosperity rather than technology adoption. Governments should assess AI’s contribution through real household incomes, business investment, productivity, employment outcomes, consumer benefits and the quality of public services.

These priorities require coordination across competition policy, industry development, education, infrastructure and public-sector reform. They also require governments to distinguish between investments that genuinely expand economic capacity and subsidies that simply reduce the private cost of adopting technology.

Getting the metrics right

Perhaps the most important reform is to change how success is measured.

A government could report that thousands of businesses have adopted AI, millions of working hours have been saved and substantial investment has occurred without demonstrating that Australians are materially better off.

A credible national assessment should examine whether productivity improvements are accompanied by rising real household incomes, stronger business investment, internationally competitive industries and better public services.

It should also identify who captures the economic benefits.

Lower consumer prices may improve living standards even when wages remain unchanged. Higher business profits may finance investment and future employment. Increased public-sector efficiency may create resources for improved services or lower fiscal pressures.

These outcomes should be evaluated together rather than assuming that productivity growth automatically produces a particular distribution of benefits.

The objective is not to prescribe precisely how every dollar of AI-generated value should be shared, but to establish whether the economy is becoming more capable, competitive and prosperous.

Technology is not an economic strategy

AI could make a substantial contribution to Australia’s future economic performance. It may help businesses overcome resource constraints, improve public services and develop products and business models that were previously uneconomic.

However, technological progress alone cannot guarantee sustained improvements in living standards.

The ultimate results will depend on competitive markets, workforce adaptability, investment, innovation and the ability of Australian businesses to capture value from the capabilities they develop.

For governments, the strategic question must therefore move beyond how rapidly AI can be deployed to how effectively it can strengthen the foundations of national prosperity.

Australia does not need an AI policy measured primarily by how much technology it adopts. It needs an economic strategy measured by how much better off Australians become.

The distinction is fundamental. AI may enable Australia to produce more with the resources it has, but prosperity will depend on whether those productivity gains become higher real incomes, competitive businesses and better public services.

AI can provide the productivity. Sound economic policy must help turn it into prosperity.