Australia does not have a national strategy for regulating AI in the workplace. That is not an opinion. That is the finding of a report released in early 2025 by the Australia Institute’s Centre for Future Work, which examined how AI is already affecting workers across industries and found a significant gap between how fast the technology is moving and how prepared the regulatory environment is to manage it. The report flagged surveillance risks, unsafe workloads, and rising job insecurity as three of the most pressing concerns. For business owners, that gap is not just a policy problem. It is a decision you are going to have to make before anyone tells you what the rules are.
What the Report Actually Found
The Centre for Future Work report documented several patterns that are already showing up in Australian workplaces. AI driven monitoring tools are being used to track productivity in ways that were not disclosed to workers at the time of hiring. Automated workload allocation systems are setting output expectations that humans cannot realistically sustain. And workers in roles most exposed to automation are reporting higher levels of job insecurity, not because they have been told their jobs are at risk, but because no one has told them anything at all.
That last point is the one that should sit with business owners longest. The insecurity is not always coming from what is happening. It is coming from the silence around it.
The report also noted that Australia lacks a coordinated framework for how AI should be introduced into workplaces, what disclosures employers are required to make, and what protections exist for workers whose roles are being reshaped by automation. Other countries are further along. The EU AI Act, for example, classifies AI systems used in employment and workforce management as high risk, meaning they come with mandatory transparency and human oversight requirements. Australia does not yet have an equivalent.
The Regulatory Gap Is Real and It Is Moving
The Australian Government has been consulting on AI regulation since at least 2023. The Department of Industry, Science and Resources released a voluntary AI Ethics Framework and has since put forward a proposal for mandatory guardrails for high risk AI applications. But as of mid 2026, mandatory regulation specifically covering AI in the workplace remains in development. That means businesses are operating in a space where the technology is available, the adoption is accelerating, and the rules that will eventually govern how you use it are still being written.
This is not unusual. Regulation almost always follows technology rather than leading it. Cloud computing, social media, gig economy platforms. The pattern is consistent. The tools arrive, businesses adopt them, problems emerge at scale, and then governments respond. The businesses that handled those transitions well were not the ones that waited for the rules. They were the ones that set their own standards before they were required to.
That is the position Australian business owners are in right now with AI in the workplace. You are not being told what to do. But what you do in this window will define how your team experiences the transition, and it will also define how exposed your business is when regulation does arrive.
The Three Risks Worth Taking Seriously
The report identified surveillance, workload, and job insecurity as its primary concerns. Each of these has a direct implication for how you are using or planning to use AI in your business.
Surveillance. AI powered monitoring tools are being sold to businesses as productivity solutions. Time tracking, screen monitoring, keystroke logging, communication sentiment analysis. Some of these tools have legitimate use cases. But deploying them without being transparent with your team about what is being tracked, why, and how it is being used is a significant risk. Not just ethically, but practically. When workers find out they are being monitored in ways they were not told about, trust breaks down fast. And trust, once broken in a team going through significant change, is very hard to rebuild.
Workload. AI can increase the volume of work a person handles without anyone consciously deciding that is what should happen. A customer service team using AI to draft responses might find they are now expected to handle 40% more tickets. A finance team using AI to generate reports might find the expectation shifts from one report per week to three. The tool increases capacity, and the organisation fills that capacity without asking whether the people doing the work are okay with the new baseline. This is how AI creates unsustainable workloads without anyone intending it.
Job insecurity. When businesses introduce AI into workflows without explaining why, what it means for the team, and what the path forward looks like for each person, workers fill the silence with their own interpretation. That interpretation is almost always more alarming than the reality. The fear of being replaced is a reasonable response to watching your organisation adopt tools that can do parts of your job. If you are not actively addressing that fear, you are letting it compound.
What Business Owners Should Do Before Regulation Arrives
The regulatory gap is not an invitation to move fast without thinking. It is an opportunity to build something your competitors probably are not building yet. A workplace AI culture that your team actually trusts.
Here is what that looks like in practice.
Tell your team before you deploy. Before any AI tool goes live in your business, have a conversation with the people it will affect. Not a policy memo. A real conversation. What is changing, why you are doing it, what it means for their role, and what support they are going to get. This is not complicated. It is just often skipped because it feels slower than buying the software and getting started.
Be transparent about monitoring. If you are using or considering any AI tools that track how your team works, be clear about it. What is being collected, how it is being used, and who sees it. If you would not be comfortable your team finding out after the fact, that is a sign you need to have the conversation before the fact.
Set a workload baseline before you automate. Before you introduce AI into a workflow, document what the current expectations are for the people in that role. Then, when AI increases capacity, you have a reference point for what a fair adjustment looks like. Productivity gains from AI should benefit the business and the people doing the work, not just add volume to the same workload.
Start building your internal AI policy now. You do not need to wait for the government to tell you what the rules are. Write your own. What AI tools are approved for use, how they are to be used, what data they can and cannot access, and how decisions made with AI assistance will be documented. A policy does not have to be perfect to be useful. It just has to exist, and it should be written with your team’s input, not handed down from above.
The Real Question Here Is Not About Compliance
When regulation does come to AI in Australian workplaces, and it will, the businesses best placed to adapt will not be the ones that scrambled to meet the minimum requirements. They will be the ones that had already built something worth keeping. A team that was brought along through the transition, that understood what was changing and why, and that trusts the business to keep making decisions with their interests in mind.
The report from the Centre for Future Work is a warning about what happens when that does not happen. Workers being surveilled without disclosure. Workloads set by algorithms without human oversight. People sitting in uncertainty because nobody told them what their role looks like in six months.
None of that has to happen in your business. The technology is not making those decisions. The people running the businesses are. And right now, before the rules are written, you still have the space to decide what kind of AI workplace you are building. That decision is a lot easier to get right at the start than to fix once trust has already been damaged.
Get in touch at www.xsiv.au/#form