The Federal Government has announced a new national framework that will shape how AI is regulated in Australia. Experts unpack what we know so far about the framework, how it could affect employers and the steps HR can take now to prepare.

AI regulation in Australia is entering a new phase. Last week, the Federal Government announced a National AI Framework aimed at supporting innovation while strengthening safeguards for workers, businesses and the wider community.

The new approach aims to “shape the future rather than letting the future of AI shape us”, said Prime Minister Anthony Albanese in a statement.

The framework will go before the National Cabinet in August, with legislation expected to be introduced to Parliament early next year.

While many of the details are still to come, the announcement offers a glimpse into how AI will be governed and regulated in Australia over the coming years.

“Unlike the European Union, for example, Australia hasn’t had a comprehensive AI regulatory framework – we have more of a mishmash of different regulations that all work together,” says Mitchell Wright, Partner at law firm Gadens.

“This denotes a structural shift by the Australian government to come up with a national framework that deals with some of the higher-level policy issues.”

Here’s what the announcement signals about where AI regulation is heading – and what employers should be preparing for now.

New Australian Standards for the use of AI

A central feature of the framework is the government’s commitment to develop a set of Australian Standards for AI. While these standards have not yet been published, the announcement outlines several priority areas they are expected to address.

Among them are improving AI safety, protecting intellectual property, supporting workers through technological change, strengthening Australia’s data centre infrastructure and ensuring AI contributes to economic growth without undermining jobs or workplace protections.

Responsibility for these standards will sit partly within a newly created Office of AI within the Department of Prime Minister and Cabinet. However, policymakers have yet to provide details on how compliance will be monitored or enforced.

“They’ve still not announced a standalone AI act to regulate AI use, which may indicate that they’ll continue to take the approach that they’ve taken so far, which is to implement different additional sections and provisions in existing legislation,” says Emma Lutwyche, Partner at Pinsent Masons.

While we don’t yet have details of the standards and how they apply to employers, Lutwyche says employers can already gain a sense of where the standards are heading by looking at recent AI-related reforms introduced across other areas of Australian law.

For example, one clear priority is worker safety. The establishment of the Australian AI Safety Institute in 2025 and the recent introduction of the NSW Digital Work Systems Bill, which requires NSW employers to manage risks arising from digital systems, signal an increasing expectation that organisations understand and actively manage the workplace impacts of AI.

Another emerging focus is transparency around automated decision making. Amendments to the Privacy Act due to commence in December this year will require organisations to disclose when AI or another computer program has made, or substantially assisted in making, decisions that affect individuals.

“[It’s about] ensuring that, where there is AI decision making that impacts employees, there is a human element to that decision making, and there is a really good understanding by humans in the business about how that decision has been reached,” says Lutwyche. 

“That’s one of the key things that employers are going to be regulated on: if you’re implementing technological changes that use AI, you need to understand how they work before you implement them.”

A “social licence” for AI

According to Andrew Charlton, Assistant Minister for Science, Technology and the Digital Economy, one of the key goals of the new framework is to establish a “clear and enforceable social licence for AI”. 

While the phrase hasn’t yet been clearly defined, Wright interprets it as signalling a stronger expectation that employers will involve workers in AI adoption rather than simply introducing new technology without discussion.

“It’s about bringing employees on the journey for the implementation of AI, and demonstrating fairness across the board,” says Wright. “Consultation processes and transparency in how AI is being used within a business are probably the key issues here.”

With this in mind, it’s likely that the new standards will involve expectations to consult employees earlier and more meaningfully when introducing AI that changes how work is performed or how employment decisions are made.

“We already have consultation obligations about major change under various awards and enterprise agreements, and a large percentage of the Australian workforce are covered by those,” says Amanda Junkeer, Partner at Gadens.

“‘Major change’ can include technology… but if there’s a change to expressly say that AI changes in the workplace must be consulted about, that will bring employees and unions to the table when the changes are first being conceptualised.”

This could prompt pushback from employers concerned that the regulations will hinder AI innovation and growth, she adds. Policymakers will therefore need to strike a careful balance between protecting workers and ensuring regulation doesn’t unnecessarily discourage innovation or delay productivity gains.

Given that the framework promises to ensure that Australians share in AI’s economic benefits, it’s likely that job security will be another key consideration in the construction of a ‘social licence’ for the use of AI.

“From a policy perspective, the government is really interested in the overarching impact on employment in the future, because labor market utilisation is a really important economic driver,” says Junkeer.

That could eventually translate into expectations for employers to invest in workforce capability to help employees develop the skills needed to work alongside AI, while encouraging organisations to use the technology to augment jobs rather than eliminate them.

Four steps HR can take now

As the framework takes shape in the coming months, there are several practical steps HR teams can take now to stay ahead of future regulation.

1. Audit where AI is already being used

As new AI standards are introduced, employers should be aware that they are likely to apply not only to future AI projects, but also to AI tools and systems already in use across the organisation.

“It’ll be really important for employers to have a really good understanding of what AI products they’re already using or what AI in their supply chains already might exist, and then making sure that anything they’ve got in operation complies with the new standards as they’re developed and implemented,” says Wright.

2. Review workforce impacts

With the new framework’s focus on job security in mind, employers would be well-placed to assess their future capability needs and identify roles that might be at risk.

“You need to really think about your short, immediate, and longer-term workforce impacts,” says Junkeer.

“AI technology is changing at a rapid pace, and it will continue to change at a rapid pace. New tools will be available. Jobs will change quite quickly. And if an employer is not prepared for that, they end up having to implement change at speed. And when you implement change at speed, you don’t bring people along with you as part of the consultation process – you get disengagement and you get disputation.”

3. Build consultation into AI projects

In anticipation of possible additional standards for consulting employees on AI adoption, Lutwyche recommends building consultation into AI projects from the planning stage.

“The best-practice approach is to consult about these things if there’s any risk that it’s going to have an impact on psychological safety,” she says. 

“For example, something like implementing Copilot into an enterprise email system in a professional workplace probably isn’t something you need to consult on, because it probably isn’t going to have implications on safety. But changing a system that deals with surveillance, time and attendance recording, rostering, performance management, or recruitment processes – those are things that likely need to be consulted on, because they may have an impact on psychological safety.”

4. Break down organisational silos

As regulation expands, it will become increasingly important to coordinate oversight of AI across different organisational functions.

“The executive team, the IT team, the safety team and the HR team can’t be siloed when it comes to this particular issue,” says Lutwyche. “Staying up to date and staying consistent within the organisation is really important.”

The bottom line for HR

Although the new framework doesn’t yet create significant new obligations for employers, it signals that expectations around transparency, consultation, workforce planning and responsible AI governance are only likely to increase. 

Organisations that begin strengthening these capabilities now will be better positioned as the framework evolves into detailed standards and legislation.

🧰 HR’s career resource toolkit

• Article: How can HR step out of the AI pressure cooker in 2026?
• Podcast: Stay across Australia’s evolving employment law landscape with AHRI’s ER/IR Unpacked podcast.
• Learning: Gain clarity and stay ahead of AI-driven change with AHRI’s AI & HR: What You Need to Know to Stay Compliant course.
• Article: How can HR keep people at the centre of AI transformation?