As Australia pursues investment in large data centers and frontier AI while weighing infrastructure costs, South Australian Liberal Sen. Andrew McLachlan urged protections for artists and biodiversity, The Guardian reported Nov. 15.
The Guardian reported that McLachlan wants compensation for artists whose work is used in AI training to be considered individually, saying an industry-wide arrangement could be difficult to make fair. He also warned that data-center development should not damage biodiversity, telling the newspaper, “I choose real forests over virtual ones.”
A September consultation from Australia’s Department of the Prime Minister and Cabinet proposes requirements for large data centers and conditions for frontier AI training. The proposed standards cover energy and water use, costs, community engagement, location and workforce impacts. The paper defines a large center as a facility designed for at least 30 megawatts (MW) and asks whether some obligations should vary by size. Submissions closed Oct. 9.
The consultation says wider AI use will require larger and more complex data-center infrastructure. The government says attracting investment in large data centers and frontier AI training could strengthen national security, productivity, science, innovation and resilience. It also aims to prevent demands for energy, water, land and infrastructure from shifting costs to households, businesses and communities.
Copyright remains part of the standards work. In a July speech, Prime Minister Anthony Albanese said writers, musicians, artists and journalists must retain ownership and control of their work. “Anything less, is theft,” he said. EconoTimes previously covered First Nations artist Sally Scales’ concerns about AI and artists’ control of cultural work.
In a separate step, the government announced Nov. 10 that it would legislate six safety and security duties for frontier AI companies operating in Australia. The proposed measures cover security controls, management of serious model risks, independent evaluations, rapid incident notifications to the Australian government, publication of safety and security frameworks, and responses to updated standards and rules. A proposed targeted regulator would enforce the duties, with further legislative details to follow.
Albanese said in July that the government aimed to bring AI standards legislation to Parliament early in 2027.
