After more than a year of consultations, discussion papers and warnings from every corner of the economy, the federal government is finally moving to say how it intends to govern artificial intelligence. According to ABC News, the Prime Minister is preparing to outline the government’s plan for AI regulation, a decision that will shape how the technology is built, sold and used across the country for years to come.
The signal matters because Australia has spent an unusually long time circling this question. The technology raced ahead of the policy, and businesses, workers and the tech industry have all been left guessing about whether Canberra would write new laws, adapt the rules already on the books, or try to do both. That uncertainty has become a story in its own right, and it is one FluentSea has tracked repeatedly, from the government’s earlier attempts at a national framework to the fragmented state of AI training across schools, universities and small business.
How we got here
The groundwork stretches back to the previous term of government. In 2023 the then Industry Department ran a wide public consultation on safe and responsible AI, and in September 2024 it followed up with a proposals paper canvassing mandatory guardrails for AI used in high-risk settings, alongside a voluntary safety standard for developers. The idea was to treat AI the way regulators already treat other risky products: heavier obligations where the stakes are high, such as in recruitment, healthcare, policing or critical infrastructure, and a lighter touch everywhere else.
Since then the debate has broadened well beyond the industry portfolio. The Productivity Commission spent much of the past year examining how AI could lift a stubbornly flat productivity rate, and it has leaned towards caution on heavy-handed rules, arguing that Australia should avoid duplicating protections that existing law, in areas like privacy, consumer protection and anti-discrimination, already provides. That framing sits in tension with the safety-first approach many civil society groups and unions have pushed for, and the gap between those two camps is precisely what a prime ministerial plan will have to bridge.
Two views on what the plan should do
On one side sit business groups and much of the technology sector, who worry that a rush to legislate could saddle Australian firms with obligations that overseas rivals do not face. Their argument is that the country is a small market that mostly imports its AI from the United States and China, and that overly prescriptive local rules could slow adoption at the very moment productivity gains are within reach. Many in this camp would prefer Canberra to clarify how current laws apply, fund practical guidance, and reserve new legislation for genuinely high-risk uses.
On the other side are unions, creative industries and consumer advocates, who see light-touch regulation as an invitation to harm. They point to the real-world damage already surfacing in Australian communities, from deepfake images circulating in schools to AI-generated abuse material and the use of chat logs as evidence in criminal trials, all of which FluentSea has reported on this year. For these groups, voluntary standards are not enough, and the test of any plan will be whether it puts enforceable obligations on the companies deploying the technology rather than leaving individuals to seek redress after the fact.
Creators have their own stake. Writers, musicians and visual artists have spent the past year pressing for rules on how their work is scraped to train AI models, and for transparency about what goes into those systems. The music world offered a vivid example when the duo Peking Duk reworked an AI-assisted track after a backlash, a reminder that the cultural questions here are not abstract. Any regulatory plan that stays silent on copyright and training data will disappoint a large and vocal constituency.
What it means for Australia
The choice the government makes will ripple far beyond Canberra. For the enterprise sector, clear rules would remove a genuine handbrake on investment: banks, insurers and health providers have been reluctant to deploy AI at scale while the legal picture stays murky, and several of Australia’s biggest institutions have said as much. Commonwealth Bank research earlier this year found that customers are wary of blind trust in AI banking, which suggests the public appetite for guardrails is real and not just a Canberra preoccupation.
There is also a hard infrastructure dimension. Australia’s ambitions to host data centres and build sovereign AI capability are colliding with the country’s energy constraints, a tension that has flared in debates over data centre power rules, regional projects in places like Wagga Wagga, and gas developments in the Northern Territory. Regulation is not only about safety and ethics: it will help decide whether Australia is a place where AI is merely consumed or genuinely built, and whether the jobs, skills and economic value that come with that stay onshore.
For small business, the stakes are more immediate and more practical. Owners who are only now experimenting with AI tools need certainty about what is allowed, not a thicket of obligations they cannot afford to navigate. The success of any plan will partly rest on whether it comes with the support, hotlines, training and plain-English guidance that smaller operators have been asking for, rather than rules that only large firms with legal teams can comply with.
What happens next
The immediate question is how far the Prime Minister’s plan goes. Will it commit to legislation for high-risk AI, as the earlier proposals paper canvassed, or settle for a mix of updated existing laws and voluntary standards? Will it name a lead regulator, or spread responsibility across the agencies that already oversee privacy, competition and online safety? And will it tackle the thorny copyright and training-data issues that the creative sector is watching so closely?
Whatever the detail, the announcement marks a shift from consultation to commitment. After a period in which Australia risked looking like it was studying the problem rather than solving it, the government is now expected to say where it stands. The reaction from business, unions, technologists and the creative industries will tell us quickly whether the plan has struck a workable balance, or whether the country is in for another round of argument about how to govern the most consequential technology of the decade.
Sources: ABC News.
