Australia Creates Office of AI, Plans Mandatory Standards by Early 2027

The Australian government is establishing a dedicated Office of AI inside the Department of the Prime Minister and Cabinet and plans to legislate mandatory national AI standards by early 2027, Prime Minister Anthony Albanese announced in a speech at the University of Sydney on July 15, 2026^7,9,10.

The speech, titled "AI in Australia's Interests," marks a shift from the government's previous case-by-case approach to a centralized, interventionist AI policy framework¹². Albanese declared that AI "is a bigger challenge and a bigger opportunity than social media" and that Australia must move now to determine its "social licence".

The new Office of AI will coordinate policy across government rather than leaving individual departments to develop their own approaches. The office will also help design new national standards covering the use of AI, data centres, and intellectual property.

A central component of the framework targets data centre operators. The government plans to make it a legal requirement for large AI data centres to underwrite new energy generation, ensuring at least as much energy is put back into the grid as data centres consume. The framework will also force data centres to reduce power consumption during peak periods and mandate minimisation of industrial water usage. Albanese said the framework would ensure no costs are passed onto homes or businesses.

Before new data centres receive approval, operators would have to provide evidence of financial benefits to local communities through employment opportunities, investment, and local spending¹³. The government plans to take the AI framework to National Cabinet in August.

The policy extends beyond data centres. The use of AI in automated decision-making by government departments and agencies will be subject to new rules under a national plan covering consumer protections, workplace safety, and privacy, with rules prioritising fairness, accuracy, and transparency^1,5. Federal departments and agencies continue to rely on automated decision-making in some service delivery functions.

The Albanese government also rejected a copyright exemption that would have allowed AI companies to use protected content for free when developing AI models. Albanese said no company should use Australian books, music, art, or news to build or train AI without the artist's control, and promised laws to address how payments for creators' content reach actual creators.

Labor's proposed digital duty of care legislation will extend to AI firms, meaning private companies such as OpenAI and Anthropic may need to conduct risk assessments and publicly report on the consequences of their systems to comply with Australian law⁴. The Albanese government plans to pursue AI safety through every relevant agency and regulator, spanning consumer law, therapeutic goods, workplace health and safety, and online safety.

National security concerns were a driving force behind the policy pivot, according to government sources speaking on condition of anonymity. Assistant Technology Minister Andrew Charlton said the Artificial Intelligence Safety Institute has already begun testing models from AI developers.

The plan has drawn criticism. Deputy Liberal Leader Jane Hume warned that too much government intervention would be akin to handing unions a veto over AI projects. The AAP reported the plan was described as "just more bureaucracy" by critics.

ANALYSIS The creation of a centralised Office of AI within the prime minister's own department, combined with legislated standards and data centre energy mandates, represents a structurally different approach from the voluntary guidance models adopted by jurisdictions such as Japan. The framework's requirement that data centres underwrite their own energy supply could materially affect the cost calculus for AI infrastructure operators considering Australian deployments.