Prime Minister Anthony Albanese has unveiled Australia’s first national AI standards, but many have argued the government’s reforms are too little, too late.

During a speech at the University of Sydney on 15 July, Albanese announced the introduction of a new Office of AI within the Department of the Prime Minister and Cabinet, and a national set of AI standards.

“Today I announce that to seize and shape the generational opportunity that AI represents, our Government will establish a set of Australian Standards for AI,” Albanese said in his speech, “clear, consistent and mandatory.”

Framing the announcement as another opportunity for Australia to lead the world, Albanese pointed to the country's history of "world-first" reforms—including Medicare, compulsory superannuation, the secret ballot, women's right to stand for Parliament and the recent under-16 social media ban.

The proposed framework will introduce standards governing AI safety, copyright, government use of AI, and the energy and water demands of large-scale data centres.

The new rules come amid growing calls from leading economists, AI researchers and environmental experts for the Labor Government to introduce and enforce stricter regulation on AI.

A statement from the Stanford Digital Economy Lab said that, if left unregulated, AI “could drive unprecedented transformation of our economy, larger than the Industrial Revolution”.

Greenpeace Australia argued that, without appropriate guardrails, data centres in Australia could “derail the renewable energy transition, entrench gas and turbocharge climate pollution”.

Labor’s announcement represents a significant shift in their approach to AI. Before unveiling the new framework, the government’s AI strategy included a set of expectations, rather than AI-specific legislation.

The Greens have repeatedly called for stronger oversight of AI infrastructure, emphasising that rapid expansion of data centres risks increasing pressure on Australia's electricity grid, water resources and emissions if environmental safeguards are not strengthened.

Sarah Hanson-Young, chair of the Artificial Intelligence and Data Centres enquiry and Greens spokesperson, said that Australian’s AI regulations were “lagging behind”.

“Australia is sleepwalking into an AI catastrophe,” she said.

The Greens’ AI and digital rights policies focus on environmental protections, public accountability and limiting the influence of Big Tech in shaping AI policy.

They have pledged “transparency and accountability around how Big Tech uses AI”, following claims that the Albanese government has been “courting global AI companies” without consulting the public.

Pauline Hanson, leader of One Nation, capitalised on Labour’s weak policing of AI at her recent press club speech, where she campaigned for “enforceable safeguards” due to concern over the rise of AI and its impact on jobs.

Albanese will seek agreement on his approach from Premiers and Chief Ministers at the National Cabinet next month and aims to bring the legislation to Parliament early next year.

Hanson-Young said the government cannot allow a “free-for-all” in between announcing these regulations and actually enforcing them.

According to Lamestream, the new framework, which they said “doesn’t mean stronger regulation, just national consistency”, does not apply to the 90 data centres currently in construction around Australia.

On the same day that Albanese boasted of Australia’s ability to lead the race on major global issues instead of “beg[ging] and borrow[ing] from elsewhere”, New York imposed a one-year pause on new AI data centres.

The order affects data centres with an electricity capacity greater than 50 megawatts. For comparison, the West Footscray data centre has a capacity of 225 megawatts, while the proposed Mamre Road data centre in Sydney plans to have a capacity of 1.2 gigawatts.

The New York Governor and Democrat, Kathy Hochul said the pause exists while New York establishes their own AI frameworks and guardrails. She said they aim to reduce pressure on their energy grid and water supply, and minimise land disruption in the meantime.

US Senator Bernie Sanders has been one of the most outspoken critics of AI. In March, Sanders, alongside Democratic Representative Alexandria Ocasio-Cortez, introduced a new AI Data Centre Moratorium Act, legislation that would pause development of new AI data centres until safeguards are in place to ensure AI is developed in a way that prioritises “safety and humanity”.

Sanders said “maybe we better slow the process up in order to understand the extraordinary impact that AI is going to have” in basically all facets of life. 

While the moratorium has drawn criticism from the tech industry, co-chair of the House of Democratic Commission on AI, warned that a pause risks “surrendering the lead in innovation to China”.

Meanwhile, environmental experts have argued that a temporary pause is justified.

Biological and Environmental Engineering Professor Lindsay Anderson said that, while the moratorium will not resolve the world’s AI issue, it will create some breathing room for better planning.

“Imagine the difference it would have made if these limits had been put in place when the world first grasped the risks of these platforms,” Albanese said, talking about Australia’s implementation of the eventual under-16 social media ban.

The comparison raises the question: could earlier intervention have shaped a safer path for AI?

While New York introduced temporary regulations to allow time for more permanent solutions to be developed, Australia has chosen to move “first”—despite data centre construction already underway across the country.

The question remains: is a faster response necessarily the best response, or does the quality and effectiveness of regulation matter more than the speed at which it is introduced?

Image source: AAP / Dan Himbrechts