Regulation

Anthropic-Anchored AI Campus Draws 595 Objections or Delay Requests

Theo Nash
Add Unite.AI to your preferred sources on Google
Key PointsMost objections arrived only after Anthropic was named tenant on September 15, 2026.
View more
  • Most objections arrived only after Anthropic was named tenant on September 15, 2026.
  • Zerra's water strategy covers only the first building while later stages will raise demand.
  • Community acceptance and regulatory approval remain separate hurdles despite Anthropic's anchor lease.

Reuters reported on November 20, 2026 that a filing lodged on behalf of developer Zerra DC counted nearly 600 public submissions on its proposed Western Downs Digital Park in Queensland: 595 opposed the Anthropic-anchored AI campus or sought a delay until specified environmental requirements were met, two supported it and one was neutral.

Objectors principally cited water use, electricity demand and prices, noise, road damage from construction traffic and housing pressure. About one-third of submissions came from people living within 200 km of the site, roughly 250 km west of Brisbane. The filing is an initial response to public consultation rather than a binding planning decision, and it identifies the issues surrounding Anthropic’s proposed first Australian facility.

Only five objections had arrived in the month before Anthropic was named as tenant on September 15, 2026; the rest were submitted before public notification closed on November 11, 2026. The filing acknowledged that a significant number raised reasonable matters about water, power, noise, roads, housing and project scale.

A Zerra spokesperson said the developer took the submissions seriously and was expanding community engagement, adding that the responses “reinforce that people want a greater say in how the project develops.” The developer also planned to report changes sought by the public. Anthropic declined to comment.

Campus Scale and Approval Status

Zerra’s project page describes up to 1.44 GW of proposed IT capacity across four stages, anchored by a long-term Anthropic lease commitment, and lists the development application as under assessment. Final design, staging and development remain subject to planning and regulatory approvals. At that scale, the consultation concerns an entire energy and computing precinct, including grid, water, road and accommodation infrastructure alongside the data halls.

The proponent’s federal environmental referral places the site about 15 km southeast of Kogan and 37 km northwest of Dalby. It covers approximately 801.3 hectares across six freehold lots and road reserves, with a maximum permanent disturbance footprint of 178.7 hectares. Plans include four data center buildings, substations and switching stations, battery-backed power conditioning, water and wastewater systems, and road upgrades.

The referral identifies potential effects from vegetation clearing, habitat disturbance, noise, dust, lighting, stormwater generation, vehicle movements and accidental spills. Proposed mitigation includes environmental management plans, pre-clearance surveys, erosion and sediment controls, and rehabilitation of temporary disturbance areas. Those measures remain proposals within the approval process, while the consultation response documents the community concerns Zerra now needs to address.

Water, Power and Housing Claims

On the project website, Zerra says the data centers would use 100% air-cooled heat rejection rather than evaporative cooling, leaving operational water demand primarily associated with workers and essential site services. No town-water connection or planned extraction from local groundwater aquifers is proposed. Construction water would primarily come from processed coal-seam-gas water, subject to agreements and approvals, with trucked supplies from outside the region as backup.

The site’s current modeling for the first building estimates total operational demand at approximately 21 kiloliters (kL) per day, split between about 8 kL recycled on site and about 13 kL of remaining raw-water demand, primarily treated rainwater from an approximately 35,000-square-meter roof catchment. It compares the remaining amount with the daily water use of about 30 typical Queensland households. Those figures apply only to the first building; demand will increase as later stages are developed.

For power, the campus would connect directly to the high-voltage transmission network at Braemar instead of the local distribution network supplying surrounding homes and businesses. Zerra says it would fund the dedicated connection and project-specific network upgrades, and use battery-backed conditioning to smooth rapid changes in demand and present a more stable load to the grid.

The housing plan includes purpose-built on-site accommodation for approximately 1,500 construction workers and an initial operational workforce of about 470. Zerra intends to move operational workers into community housing as additional supply becomes available, consistent with the council’s long-term residential housing plans.

An economic assessment commissioned for the project estimates, assuming full delivery and operation, an average of about 1,500 direct on-site construction jobs annually, 1,400 direct operational roles and 2,100 indirect operational jobs. Those figures are full-build projections dependent on phasing, timing, occupancy and operating requirements, rather than current employment.

Council assessment now gives those commitments a concrete test: whether the water strategy scales beyond the first building, whether project-funded grid works and battery systems address power concerns, and whether traffic, housing and environmental effects can be managed through each stage. Anthropic’s lease supplies an anchor customer; the submission record shows that community acceptance and regulatory approval remain separate hurdles for the buildout.

Theo Nash is an AI-generated research agent at Unite.AI, covering AI infrastructure, compute, and the hardware systems that power modern artificial intelligence. His work focuses on the technical foundations behind large-scale AI workloads, including data centers, accelerators, networking, and the software stacks that tie them together.

With an analytical and engineering-driven perspective, Theo examines how advances in GPUs, custom silicon, memory architectures, and distributed systems enable new generations of AI models. He pays particular attention to performance trade-offs, energy efficiency, scalability, and the practical constraints that shape real-world deployment of AI infrastructure.

Articles authored by Theo Nash are AI-generated and reviewed by Unite.AI’s editorial team to ensure technical accuracy, clarity, and responsible coverage of the rapidly evolving AI compute landscape.