The fight over AI data centres is becoming Australia’s next Indigenous policy challenge

Jenna Imad Harb, Shanine Ryan, Australian National University

In July 2026, more than 200 community members gathered at Mandoon Bilya (Helena River) in Perth’s eastern suburbs. They planted more than 3,000 native flora as part of a wetland restoration project. The hopeful scene was a stark contrast to what the river might have become.

Until recently, Mandoon Bilya was the proposed site of a hyperscale artificial intelligence (AI) data centre. It was scheduled to be one of Australia’s biggest. Aboriginal groups, environmental activists and local community groups successfully opposed the development.

Instead of a 120-megawatt AI facility, the area is now focused on ecological restoration and protecting Country. Community groups are working to realise new visions of the river’s future.

Yet the Mandoon Bilya opposition is about more than one defeated development proposal. As AI drives demand for new facilities, data centres are expected to proliferate across Australia.

The campaign highlights an urgent policy challenge. Policymakers have focused on regulating AI technologies. Yet Aboriginal communities have had little say in AI infrastructure planning.

This includes participating in decisions about the resources and infrastructure on which AI technologies depend. Decisions made now could shape Indigenous relations with the government around AI for years to come.

The case of Mandoon Bilya

Restoration work at Mandoon Bilya follows a hard-fought campaign. It was led by the Bibbul Ngarma Aboriginal Association and supported by environmental and community groups.

Developers planned the project on land containing Aboriginal heritage sites. Opponents argued the proposed data centre threatened a river system with cultural connections for local Aboriginal groups. The river also supports Perth’s drinking water and local ecosystems during warmer months.

All this raised concerns about adequate free, prior and informed consent and impacts on Country.

In May 2026, the proposal was cancelled before being considered by Western Australia’s Development Assessment Panel. Campaigners widely praised the decision. They have since shifted their focus towards broader ecological restoration plans.

The need for Indigenous leadership

Mandoon Bilya isn’t an isolated case; similar debates are emerging elsewhere in Australia. In Victoria, opposition to a proposed AI Hub has raised concerns about cultural heritage and environmental impacts.

AI is often talked about as something virtual. Yet it depends heavily on physical resources: land, water and energy. Data centres use large amounts of all three, creating environmental pressures familiar to many Aboriginal communities.

Internationally, on Turtle Island (what some local First Nations peoples call North America), Indigenous communities are also challenging proposed data centres. They argue such infrastructures compromise Indigenous rights, culturally significant sites and water systems. Such challenges underscore principles echoed in the United Nations Declaration on the Rights of Indigenous Peoples.

More people now recognise the importance of Indigenous leadership in data and digital technologies. We see this, for example, in commitments to Indigenous data sovereignty. This means Indigenous people get to own and control access to their data.

Indigenous leadership should also be central to decisions about the physical systems that make AI possible.

Towards working together

Some projects have begun to incorporate Indigenous participation and ownership. Western Australia’s largest data centre is pursued through an Indigenous Land Use Agreement with the Karajarri People. They are also co-owners of the project.

In the Northern Territory, Energy North has stated it is engaging with Traditional Owners. This consultation is part of the assessment process for its proposed hyperscale data centre.

Yet such examples remain the exception. While Australia is regulating AI infrastructure, few policies require Aboriginal leadership in decision-making.

Kaley Nicholson, chief executive of the Federation of Victorian Traditional Owner Corporations, confirms this. She argues current government expectations for data centres represent “an investment failure”. Such efforts don’t reference collaboration with Traditional Owner groups, despite lasting impacts on Country.

This raises an ongoing and crucial question. Who gets to decide about digital systems that directly impact Indigenous lands, rights and livelihoods?

Not just AI

Conflicts over AI infrastructure echo longstanding debates about resource extraction in Australia. Aboriginal peoples have long been excluded from decisions about such developments on or near their Country. However, they often bear the environmental, cultural, social and personal impacts.

Our own research with Aboriginal communities examines their experiences of agreements with the resource industry. It suggests consultation processes don’t always translate into meaningful influence. Communities often shoulder the work of engagement, negotiation and compliance, while key decisions are made elsewhere.

Agreements don’t always lead to better outcomes given the costs and challenges of culturally sensitive, meaningful participation. Participating in agreements requires a lot of time and unpaid work. This includes coordinating community members to translate complex legal documents.

This work often falls on Elders and leaders, who bear logistical and administrative burdens. As AI systems expand, the challenge is ensuring that partnership isn’t burdensome and is more than box-ticking.

The time to act is now

Right now is the time to shape how Australia involves Indigenous peoples into AI development policies – before detrimental practices are locked in.

Consultation alone is not enough. Aboriginal people must have a seat at the table and be included before concept design to inform better outcomes for all. Aboriginal communities should have formal and resourced roles in decisions about land access, approvals and ownership.

It’s not too late to embed meaningful Indigenous leadership and governance in our AI development policies, but we must act now.

This article is republished from The Conversation under a Creative Commons license. Read the original article.