When the Algorithm Joins the Team

By Dr Anna Kiaos  |  Founder, Mind Culture Life Australia PTY LTD

Safe Work Australia's guidance on AI and digital technologies makes one thing clear — the tools reshaping our workplaces are a work health and safety matter, and the duty to manage them already sits with every employer.

Artificial intelligence has quietly become a colleague. It screens résumés, routes customer inquiries, schedules shifts, monitors productivity and flags safety incidents before a person notices them. For most Australian organisations, the question is no longer whether to adopt AI and digital technologies, but how to do so responsibly. Safe Work Australia's guidance on AI and digital technologies offers a useful re-frame for that conversation: these tools are not simply an IT upgrade or a productivity play. They are a work health and safety (WHS) matter — and the responsibility for managing them already rests with the people running the business.¹

That framing matters because it changes who owns the decision. Adopting new technology is often treated as a question for IT and finance. Safe Work Australia's guidance moves it onto the same footing as any other workplace hazard, and squarely into the remit of leaders, managers and the people professionals who look after workforce wellbeing.¹

The duty you already have

Under Australia's model WHS laws, a person conducting a business or undertaking — a PCBU, in practice most employers — must eliminate risks to health and safety so far as is reasonably practicable, and where elimination isn't possible, minimise them. Safe Work Australia is explicit that AI and digital technologies are no exception. When you introduce a new system, or change an existing one, you must manage the associated WHS risks the same way you would manage any other hazard: identify it, assess it, and put control measures in place.¹

In other words, there is no separate, softer standard for technology. The same hierarchy of control that governs a piece of plant on the factory floor applies to the algorithm making decisions about workers. For leaders used to treating software procurement as a back-office matter, that is a meaningful shift in accountability.

Risk isn't a one-off event

One of the most practical insights in the guidance is that WHS risk from technology is not fixed at the moment of purchase. It moves. Routine changes to IT systems, and updates “pushed” or initiated by software providers, can introduce new risks or change existing ones — often without anyone in the organisation actively deciding to make a change.¹

This is a genuine blind spot. A vendor's update to a scheduling tool, a new feature in a monitoring platform, or a tweak to an AI model's behaviour can alter how work is done, how workers are assessed, or how much cultural pressure they are under — all before the next scheduled review. Managing technology risk therefore has to be ongoing, not a box ticked at implementation. It calls for the same vigilance organisations already apply to other evolving hazards.

The double edge

What makes AI genuinely challenging is that the same tool can protect and harm at once. Safe Work Australia notes that introducing a technology can control one hazard while creating or changing risks elsewhere.²

Automation is the clearest example. Handing routine tasks to software can reduce drudgery and error. However, if it is implemented poorly, the work that remains for people can become more complex and more concentrated — a phenomenon known as work intensification — which increases fatigue and, with it, the risk of error and harm.¹ The technology “worked” on its own terms while quietly degrading the conditions of the humans around it.

This is why AI cannot be evaluated on efficiency alone. A change that looks like a clear win on a productivity dashboard may be shifting load, cultural pressure and risk onto workers in ways that only surface later as burnout, turnover or incidents.

Where the human risks hide

For many organisations, the sharpest risks from AI and digital technologies are psychosocial rather than physical. Safe Work Australia's own catalogue of examples spans workplace monitoring, automated decision-making, changes in exposure to distressing or traumatic material, customer and client interactions, harmful behaviours and harassment, and remote or isolated work.²

Each of these touches on how people experience their work. Intensive digital monitoring can erode trust and autonomy. Automated decisions about rosters, performance or pay can feel opaque and unfair, and can carry bias. Some roles now expose workers to traumatic material at scale. These are not fringe scenarios — they are becoming ordinary features of digitally mediated work, and they sit squarely within an employer's duty to manage psychological as well as physical safety.

Getting this right depends less on the sophistication of the technology than on the quality of the thinking around it: whether workers were consulted, whether the human impact was weighed alongside the operational case, and whether there is a way for people to question a decision a machine has made.

The opportunity side of the ledger

None of this is an argument against adoption. The same guidance is clear that AI and digital technologies can also improve safety — surfacing hazards that people miss, using wearable devices to monitor for risk, removing workers from dangerous tasks through autonomous plant and robotics, and enriching training.² Used deliberately, these tools can lift the standard of safety in ways that were not previously possible.

The difference between the opportunity and the risk is rarely the technology itself. It is the intent and the process behind its introduction.

Leading well

For employers and WHS and people professionals, the guidance points toward a familiar discipline applied to a new domain. Treat the introduction or change of AI and digital technologies as a change process with WHS at its centre. Consult the workers who will actually use — or be assessed by — the system, because they see the risks first. Look past the immediate function to the effect on workload, autonomy, fairness and wellbeing. In addition, build in review, so that the “set and forget” model gives way to ongoing oversight that keeps pace with the updates arriving from vendors.

The organisations that will handle this era well are not those that adopt AI fastest, nor those that avoid it out of caution. They are the ones that hold two ideas at once: that these tools can make work safer and better, and that realising that promise depends on treating the people around the technology as the point of it. That is a leadership question as much as a compliance one — and it is exactly the conversation worth having before the next system goes live.

About the Author — Dr Anna Kiaos

Dr Anna Kiaos is the Founder of Mind Culture Life Australia, a specialist research and consulting firm working at the intersection of organisational culture and workplace mental health. She is both a trained organisational culture expert and therapist — a rare combination that sits at the core of her work helping leaders identify the cultural conditions that drive psychological injury before they escalate into claims, attrition or crises.

Dr Kiaos completed her PhD in Management at Macquarie University Business School under Professor Gideon Kunda, a protégé of Emeritus Professor Edgar Schein at MIT. Her peer-reviewed research on organisational subcultures and structural reform has been published in the Australian Journal of Public Administration, and she is a researcher at the University of New South Wales. She is the creator of the Culture Pressure Map™, a proprietary diagnostic framework that makes cultural and psychosocial risk visible before it surfaces as harm.

References

1. Safe Work Australia, Artificial intelligence (AI) and digital technologies. Available at: https://www.safeworkaustralia.gov.au/safety-topic/hazards/digital-technologies-ai

2. Safe Work Australia, Artificial intelligence (AI) and digital technologies — Example opportunities and risks. Available at: https://www.safeworkaustralia.gov.au/safety-topic/hazards/digital-technologies-ai/example-opportunities-and-risks

Learn more at mindculturelife.com.au.

This article draws on Safe Work Australia's guidance on artificial intelligence (AI) and digital technologies. It is general information for Australian workplaces and is not legal advice; organisations should consider their own circumstances and obligations under the applicable WHS laws.