As generative artificial intelligence (GenAI) reshapes everything from healthcare to productivity, it’s triggering a global scramble – not just for data and chips, but for electricity. Can Australia’s renewable energy resources provide a solution?
In the US and Europe, data centre electricity demand is projected to grow by 400 terawatt-hours between 2024 and 2030 – almost as much as France uses in a year.
Big tech companies are striking multi-billion-dollar energy deals to meet gen-AI’s surging demand. Microsoft has even partnered with Constellation Energy to resurrect the once doomed nuclear reactor at Three Mile Island.
In Asia, land constraints are driving countries like Singapore to look abroad for clean energy – sparking ambitious initiatives like SunCable, which aims to export Australian solar power internationally.
Despite such efforts, many countries will be unable to provide the energy and digital infrastructure needed to meet the anticipated demands of the GenAI economic transition.
These challenges present Australia with a strategic opportunity: to turn our renewable energy abundance into a competitive edge in the AI-driven economy – positioning the nation as a key player in sustainable global digital infrastructure.
In a recent paper, we argue that Australia is uniquely positioned to lead in sustainable AI computing – by building renewable-powered data centres that don’t just support GenAI, but help stabilise our energy transmission grid, lower power prices, and drive regional economic development.
We call this a ‘renewable digital dividend’ – turning AI’s electricity demand into a net benefit for Australian households, industry and the energy grid.
This strategy offers a rare win-win, aligning with the Federal government’s Future Made in Australia agenda and addressing election-defining issues like cost of living, energy security and industrial revitalisation.
The challenge and the opportunity
Data centres already account for around 6 per cent of Australia’s electricity use, with that figure expected to double by 2030. Without a plan, this growth could strain our grid and drive prices higher.
We propose integrating gen-AI data centres into Australia’s energy system as flexible, grid-responsive infrastructure – leveraging renewable energy, demand management, and policy tools to support stability and lower costs:
- Infrastructure scale-up: AI data centres offer a new source of long-term, bankable demand to underwrite investment in renewable generation, transmission, and storage.
- Grid stability: These data centres can operate flexibly through strategies like batch processing, workload shifting, and dynamic compute scaling to absorb excess renewable generation, reducing curtailment and supporting grid balance.
- Lower prices: By stabilising supply and reducing volatility, data centre demand can help avoid costly interventions—contributing to lower wholesale prices.
- Digital exports: Australia can export renewable-powered compute as a value-added product—an alternative to exporting raw electricity.
Risks and their mischaracterisation
Data centre operators often disconnect from the grid during voltage fluctuations, citing infrastructure risks. This was demonstrated during the Data Centre Alley event in July 2024, when 60 data centres in Virginia simultaneously disconnected due to a failed surge protector.
Yet this risk framing overlooks that modern power conditioning technologies – such as dynamic voltage restorers and advanced UPS systems – can protect equipment without full disconnection.
What’s really at play is an economic calculation. Data centres seek the reliability and cost advantages of being connected to the grid while maintaining autonomy to disconnect when grid conditions reduce profitability.
This allows them to benefit from public infrastructure without fully contributing to its stability.
In Australia, such practices could worsen pressures on our energy system and increase costs – a classic Tragedy of the Commons, where individual actors draw excessively on a shared resource for short-term gain, ultimately harming the collective good.
A more equitable approach would align incentives and responsibilities, encouraging large energy users to contribute to collective resilience.
Why Australia can lead
Australia has what the world wants: space, sunshine, and wind. And with Future Made in Australia, it now has the policy framework to act.
This new framework is a reassertion of industrial policy in the national interest – recognising that building sovereign capability in clean energy and AI is essential for economic resilience and meeting the net zero emissions target.
We’re already seeing this thinking in action with initiatives like the Whyalla green steel package. Australia will need thousands of wind turbines and enhanced transmission infrastructure to meet its clean energy targets – and potentially far more to contribute to the export of renewable energy through gen-AI compute.
Recent additional funding for green metals projects from the Future Made in Australia Innovation Fund signals growing government recognition of this opportunity.
The prospect of a circular economy is within sight: green energy powering green industry and digital infrastructure, enabling exports and attracting foreign investment, which in turn strengthens green energy systems.
To achieve this outcome, we need action on several fronts:
- Planning reform: Fast-track approvals for integrated, grid-connected hubs that combine data centre infrastructure with renewable energy generation.
- Digital infrastructure: Expanded high-capacity fibre to regional zones with high renewable energy potential, underpinning the jobs and industries of the future.
- PPAs and regulation: Design of Power Purchase Agreements (PPAs) to unlock renewable investment by guaranteeing stable energy demand.
- Geopolitical urgency: First-mover advantage in a context where China is on track to control half of the world’s renewable electricity capacity by 2030.
As AI workloads grow globally, longstanding vulnerabilities in physical infrastructure—particularly subsea cables that transmit 95 per cent of international data – have gained renewed attention. Recent disruptions in the Red Sea and around Taiwan highlight these risks.
While some nations may try to localise AI infrastructure, most lack the renewable resources, land, or grid headroom to meet growing demand.
Australia demonstrates strategic value by offering the renewable energy, space and political stability needed to host large-scale AI infrastructure – supporting global inference capacity in a more geopolitically diversified way.
A political crossroads
With the federal election approaching, Australians are focused on rising energy costs, climate change, and economic resilience in a world becoming more dangerous and uncertain by the day.
Whatever the case for nuclear energy – hardly a compelling one – it remains a much longer-term proposition. Meanwhile, the GenAI revolution is already accelerating global demand for clean, reliable energy—and Australia’s abundant renewable resources are ready to meet that demand.
By integrating renewable energy investment with data centre infrastructure, Australia can lead the world in supporting the transition to a more productive and sustainable knowledge-driven economy.
In this context, we would be able to lower electricity prices, strengthen the grid and create an entirely new export sector.
But only if we act now.
Associate Professor Jo-An Occhipinti is an NHMRC Principal Research Fellow and Co-Director of the Mental Wealth Initiative at the University of Sydney’s Brain and Mind Centre. She is a systems scientist focused on aligning public policy with social, economic, and environmental resilience.
Dr Ante Prodan is a computer scientist and complex systems researcher with the School of Computer, Data and Mathematical Sciences at Western Sydney University. He serves as the Mental Wealth Initiative’s Chief Scientific Advisor on AI, where his research explores the societal impacts and opportunities of generative AI.
Emeritus Professor Roy Green is Special Innovation Adviser at University of Technology Sydney, and former Dean of the UTS Business School. He is also chair of the Advanced Robotics for Manufacturing (ARM) Hub.
Do you know more? Contact James Riley via Email.