Australia is deploying a new generation of digitally connected electricity infrastructure at extraordinary speed, while the cyber security needed to protect the grid is lagging a long way behind. Such is the pace of investment in batteries that the Department of Home Affairs warns that the Australian energy sector is becoming “an attractive target for malicious cyber actors”.
Australia is replacing a relatively small number of large, centrally controlled electricity generators with a large fleet of digitally controlled and remotely managed generation and battery storage units. The batteries, and many of their components, are connected to the internet, dramatically increasing the number of systems that can be attacked through cyber means. Add to this the impact of artificial intelligence, which can replace human hackers with machine counterparts able to operate at vastly greater speed and scale. In our race to decarbonise, we have been slow to address a rapidly growing national security risk—the capacity of a hostile power to hack our green grid and shut Australia down.
In June this year, the Minister for Cyber Security, Tony Burke, released a rather obscurely titled paper: The Australian Action Plan for Horizon 2 of the 2023 to 2030 Cybersecurity Strategy. It represented the latest official thinking on cyber security as it stood before the OpenAI “intrusion” into a public Medicare database. The report says that “malicious cyber actors … are exploiting insecure technologies, weak data practices and systemic dependencies. They are using artificial intelligence to increase the speed, scale and sophistication of cyber-attacks.”
But the Action Plan does not discuss the security of batteries or wind and solar installations or the electricity grid. The plan grandly claims that by 2029 “we will advance the global frontier of cyber security. We will lead the development of emerging cyber technologies to adapt to new risks and opportunities.” Aspiration is a fine thing but the gap between this objective and the current reality of Australia’s unpreparedness to deal with the cybersecurity implications of artificial intelligence is enormous.
The Albanese government’s current approach appears to be that we can regulate ourselves to greater cyber security. On September 24 the terms of reference were released for a “rapid review into Australian government arrangements for an AI-driven cyber incident”.
This followed advice from OpenAI to the government about an intrusion into a Medicare database which took place on June 18 and was undetected by Australian authorities. The terms of reference focus on ways to strengthen reporting requirements relating to AI-driven cyber incidents. They assess Commonwealth government and information-sharing arrangements, information-sharing obligations of AI firms, and will investigate the adequacy of existing legislative regulatory and enforcement frameworks.
A focus on reporting and regulation will do little to strengthen Australia’s cyber defences. The final term of reference for the rapid review is to look at mechanisms to strengthen networks and systems of government departments and agencies against AI vulnerabilities. That may create some scope to look at practicalities, but overwhelmingly the reality is that Australia is not well placed to handle the likely increase in malicious cyber activity.
The Action Plan noted that malicious cyber actors are increasingly using artificial intelligence. Attackers using artificial intelligence were responsible for 16 per cent of data breaches in 2024. What this means is that the government should not have been as surprised as it apparently was by the OpenAI cyber intrusion.
The obvious question to ask is why a government website containing Medicare information was so vulnerable. What protection mechanisms were in place, and why was the intrusion undetected? The failure to identify the OpenAI intrusion, which was essentially a training exercise rather than a malicious hacking event, suggests that there may be numerous AI-assisted cyber intrusions that have also been undetected. What steps are being taken to discover other potential hacking events?
Returning to the security of the electricity grid and the many thousands of batteries, solar and wind power systems now being connected to it, the urgent need is for government to set appropriate cyber security standards for these systems.
There is not yet a publicly documented case of an attacker remotely taking control of a grid-scale battery to disrupt an electricity system. But the Australian government’s own security assessment warns that attackers can remotely compromise and control batteries.
In December 2025, a Russian-linked cyber-attack targeted at least thirty wind and solar installations in Poland. The attack severed communications between the facilities and electricity distribution operators. Polish investigators said the attackers demonstrated the ability to interfere with the operational technology controlling renewable facilities, but they did not succeed in interrupting electricity production. Polish Prime Minister Donald Tusk said on January 15, 2026, that there were strong indications that a group connected to Russian intelligence services was behind a series of cyber-attacks targeting Poland’s energy infrastructure.
PV Magazine, quoting cyber security expert Roberto Setola, reported:
In the Poland attack, an AI-based system acted as an orchestrator to identify potential targets. Each target was then probed using a range of attack strategies. When a vulnerability was found, the attack disrupted the target’s communication capabilities; otherwise, the system moved on to the next target.
This incident points to a newly emerging risk: attacking numerous small power and battery sites simultaneously could damage wider power networks. A larger number of systems connected to the grid creates more cyber “attack surfaces”.
Panasonic said in March 2026 that it was developing cyber security monitoring systems for grid-scale battery energy storage systems. The company advises: “As grid-scale storage, renewable integration, and distributed energy resources continue to evolve, cyber security resilience will become a fundamental requirement for stable energy operations.”
A study released in early September modelled a co-ordinated cyber-attack on battery storage infrastructure in the United Kingdom. It assessed that attacking 29 per cent of national battery storage capacity—around 400 grid-scale batteries—could trigger nationwide electricity outages. The Centrii report Gridlock: What a Coordinated Battery Attack Would Cost the Grid found that:
a coordinated attack on battery storage doesn’t need to stop power generation to cause a blackout. It only needs to desynchronize the balancing layer, forcing batteries to charge or discharge in a coordinated, disruptive pattern.
The April 2025 blackout across Spain and Portugal shows how rapidly a relatively small number of digitally controlled inverter-based renewable assets can destabilise a grid. At that time, a small number of large solar and wind sites stopped producing at the peak of the day, together shedding 2.5 gigawatts of generation in less than twenty seconds—enough to collapse the Iberian grid. An investigation by the Spanish government found there was “no evidence of a cyber incident or cyber-attack as the cause of the energy crisis”. Instead, the blackout resulted from a “multifactorial surge”. However, the Centrii report points to the inherent vulnerability of renewable power generators and “fleets” of battery storage devices managed remotely through cloud-based control platforms across a national grid.
In Australia, grid-scale batteries are predicted to comprise 12.5 per cent of the National Energy Market’s roughly 72 gigawatts of installed generation and storage capacity by the end of 2027. This amounts to a seven-fold increase from 2024 levels. The cyber vulnerability arises because modern batteries are digitally controlled, networked systems that can be remotely monitored, operated, maintained and updated.
Chinese manufacturers dominate important parts of the battery supply chain on which Australia depends. The Australian market consultancy SunWiz estimates that Chinese manufacturers account for more than 70 per cent of Australia’s battery sector. At the utility scale, Chinese companies including CATL, Hithium and Trina Storage are now major battery-energy storage system suppliers, while the 850 megawatt Waratah Super Battery in New South Wales contains 3,598 Chinese-made battery units.
The exposure is not unique to Australia. The US Department of Energy found in 2024 that between 90 and 100 per cent of US battery energy storage systems had one or more critical components sourced from China. The important question is not whether an Australian battery carries a Chinese brand name. It is how much hardware, software, and control architecture in Australia’s electricity infrastructure depends on Chinese supply chains.
Operating documentation for major Chinese battery and inverter systems makes clear that remote access is an integral feature of their operation, maintenance and support. The global dominance of Chinese technology in renewable power generation creates new opportunities for intelligence gathering and foreign interference for Beijing. Less widely understood is that Beijing’s national security laws compel companies and individuals to help national security agencies when required.
The Department of Home Affairs’ Critical Infrastructure Security Centre set out the consequences in a remarkably blunt Factsheet, Risks to Battery Energy Storage Systems. Batteries “can be remotely compromised and controlled, potentially enabling espionage, sabotage, or foreign interference. Outsourcing, offshoring of data, access and control, and supply chain dependencies can exacerbate the risks.” In April 2024, the US FBI Director Christopher Wray described a “broad and unrelenting” Chinese cyber campaign against critical infrastructure to “induce panic and break America’s will to resist”.
The threat is not limited to China. As reported by the Australian Signals Directorate in 2021, the Queensland government-owned electricity generator CS Energy, which generates 10 per cent of the electricity for the national electricity market, was targeted by the Moscow-aligned Conti ransomware group. When CS Energy became aware of a ransomware incident affecting its network, it cut the external internet connection to its corporate network and initiated business continuity procedures. A ransomware threat was averted. But the number of digitally connected systems, and therefore potential “attack surfaces”, has multiplied substantially.
The government clearly recognises that Australia’s critical infrastructure security framework needs to adapt. Following an independent review of the Security of Critical Infrastructure Act 2018, the Albanese government announced in March that it would consult with industry on critical infrastructure reform. In July, officials from the Department of Home Affairs told an industry group that the government wants to amend the Security of Critical Infrastructure Act to explicitly capture battery energy storage systems, virtual power plants and other assets where significant aggregation and control can influence the electricity system.
Home Affairs guidance recommends that data and control remain in Australia and that businesses ensure battery suppliers are not obliged to comply with foreign government demands. The reality, though, is that many more gigawatts of grid-scale and household battery storage will be installed in Australia’s electricity system before our national security and critical infrastructure legislation catches up with the green energy roll-out.
The security problem we face is far broader and harder to manage than the 2018 decision to exclude Chinese companies from bidding into the 5G network. Then, for once, the government got the policy settings in place before the technology rollout. With 5G, the government could decide which companies could build the network. With batteries, governments, utilities and private investors are already deploying the technology across thousands of projects, often using globally integrated hardware and software supply chains.
Purchasers of large-scale battery storage used to stabilise the grid should understand that an amended Security of Critical Infrastructure Act will likely impose more onerous risk assessments, reporting and due diligence requirements. An interim option to strengthen battery cyber security would be to develop capability where Australian firms are putting their own control architecture and security assurance around foreign, mostly Chinese, hardware.
In April this year, the Albanese government put a microscopic $4.3 million into developing local battery manufacturing and “increased energy resilience”. Climate Change and Energy Minister Chris Bowen described this initiative as “critical to building a stronger, more secure energy system”. A more urgent effort is needed to secure our grid. The government should make it clear to state governments and other battery purchasers that tougher security standards will soon be applied to reduce the threat of malicious state and organised crime targeting electricity critical infrastructure.
We need to develop supply chains outside the People’s Republic of China. For essential industrial control components, we should encourage Australian industry to develop technology that prevents control of the grid from being taken offshore.
Australia cannot afford to build its future electricity system first and secure it later. The government should accelerate the development of Australian control and security capabilities around the foreign hardware already entering the grid, while diversifying the supply chains critical electricity infrastructure depends on.
