WHS Risks in Using AI and Digital Technologies in the Workplace
Artificial Intelligence (AI) and digital technologies are increasingly becoming a part of everyday business operations. According to the latest data published by the Australian Bureau of Statistics, almost one in two businesses (46%) have adopted new technologies, including AI during the last financial year.
From AI-powered decision-making and workflow automations to advancements in wearable devices, organisations are increasingly embracing technology to improve their efficiency and productivity. Technology has evolved in such a way that it's no longer just a tool that supports a task. In many cases, it now influences how work is done, allocated, monitored and even assists with decision-making.
While these technologies offer significant benefits, they have also created new work health and safety (WHS) risks. In fact, according to the new guidance published by SafeWork Australia, an employer's WHS obligations must also extend to managing these risks associated with the integration of AI and digital technologies in the workplace.
Understanding these emerging risks is the first step towards managing them effectively. In this article, let's take a closer look at what some of these risks are.
Understanding the Risks
According to SafeWork Australia, risks that can arise as a result of adopting AI and digital technologies can be both physical as well as psychosocial.
Physical Risks:
While technology has the potential to improve the safety of workers by removing them from hazardous tasks and environments, it can also introduce new physical risks if it is not properly managed. For example:
Musculoskeletal Injuries:
These can develop in instances where technology creates prolonged sedentary work or tasks that involve repetitive moments. Certain wearable devices might also create ergonomic risks if they are poorly designed, fitted incorrectly or used for extended periods.
Physical Injuries:
These can arise among workers interacting directly with systems, robotics or machinery in cases of equipment malfunction or failure of safety controls. In some cases, technology-driven productivity expectations may also contribute to unsafe work practices and increase the likelihood of injuries.
Source: SafeWork Australia
Psychosocial Risks:
While physical hazards remain important, many of the emerging risks linked to AI and digital technologies are psychosocial in nature. These include:
High Job Demands & Low Job Control:
Technology can increase the pace and intensity of work. Systems that monitor worker productivity or algorithm-driven task allocation might pressure workers to complete tasks faster. At the same time, they may have less influence on how work is done, leading to stress and frustration.
Reduced Autonomy & Low Job Demands:
Interestingly, these technologies can also create the opposite problem. As routine and complex tasks become automated, workers may find themselves doing less stimulating work with fewer opportunities to apply their skills, solve problems or make decisions, leading to reduced job satisfaction.
Poor Support & Lack of Role Clarity:
When new technologies are introduced without adequate training and support, it can create confusion and frustration among workers. They may end up being uncertain about their responsibilities and how technology can support their roles. In addition, frequent software updates and changing processes can create uncertainty and stress.
Poor Organisational Change Management:
Successful implementation of a technology relies on more than simply purchasing it. Without effective planning, consultation and communication, it can disrupt workflows, increase workloads and create unnecessary stress for workers.
Poor Organisational Justice:
AI assisted decision-making can raise concerns around fairness and transparency. Workers may feel uncomfortable when decisions affecting their work are influenced by algorithms that they do not understand or have limited ability to challenge, especially when human oversight is limited.
Recognition, Feedback and Harmful Behaviours:
As workplaces become increasingly digitised, opportunities for meaningful human interaction can decrease. Greater reliance on online communications may also expose them to online abuse, harassment and other harmful behaviours.
Fatigue Risks:
The ability to work from anywhere is often viewed as one of technology's greatest advantages. But the downside? Constant connectivity can create new fatigue risks. These can occur when workers feel pressured to remain constantly connected, algorithms allocate work without considering reasonable time for breaks and technology is relied upon with limited human oversight.
Remote or Isolated Work:
Remote work offers flexibility, but it can also reduce opportunities for social interactions, informal collaboration and access to support. Over time this can lead workers to feelings of isolation and disconnection.
Source: SafeWork Australia
While these risks may seem varied, they can be effectively managed through proactive planning, consultation and ongoing monitoring. SafeWork Australia provides additional examples and guidance in its AI and digital technologies factsheet.
Managing the Risks
Like any other WHS risk, PCBUs have a duty to eliminate or minimise the risks associated with AI and digital technology adoption. So, what can you do? First and foremost, the introduction of new technologies should never be treated as a standalone IT project. It should always include meaningful consultation with workers, considerations of potential WHS impacts and ongoing monitoring to ensure risks are managed effectively.
Worker consultation plays an important role in this. They are often best placed to identify how a certain technology may affect their workloads, job demands and practices. Maintaining human oversight is another important aspect, especially in areas such as work allocation, performance monitoring and decision-making.
Learn more about actions PCBUs can take to control risks here.
Maintain and Review
Managing any type of WHS risk is not a once-off activity. PCBUs must regularly review control measures to ensure they are working as intended.
Learn more about how and when control measures should be reviewed here.
Key Takeaway
AI and digital technologies cannot replace human judgement. While it can analyse large amounts of data, identify patterns and automate routine administrative tasks, it cannot fully understand the workplace context, individual circumstances of workers or the broader context in which decisions are made.
For example, an algorithm may determine the most efficient way to allocate work, but it may not recognise factors such as workload pressures, worker fatigue or competing priorities. Similarly, automated decisions can be difficult for workers to understand, question or challenge.
This is why human oversight remains essential. Technology should support decision-making, not replace critical thinking when health, safety and worker well-being are concerned.
As organisations continue to adopt AI and digital technologies, striking the right balance between innovation and human oversight will be critical. When implemented thoughtfully, technology can enhance workplace safety and productivity. However, organisations must remain vigilant to ensure new risks are identified, managed and reviewed over time.
Visit the Safe Work Australia website for more information on the published guidance.