Trusting artificial intelligence at work

Project status: Complete

Do you like receiving advice from a computer?

This project examined human-machine interactions to enhance our understanding of arising risks as ‘thinking machines’ continue to be introduced in our workplaces. The use of AI-powered systems is growing across a range of industries, as they can offer significant innovation at work, and positive economic return.

AI is already known to greatly affect the worker’s environment, e.g., role design, task allocation, time management but what about the potential work health and safety (WHS) harm it can generate? In particular, the extent to which working alongside an AI-powered system that a worker does not fully trust can harm its health and safety.

Working with Community to guide better outcomes

In partnership with Monaghan Dreaming and the Behavioural Insights team, we worked with an advisory group of Aboriginal representatives to unpack current and emerging issues facing Aboriginal and Torres Strait Islander workers and businesses.

We found that a significant amount of the population could be classified as vulnerable to risk of harm at work.

This helped to guide recommendations for future areas of research, including:

  • Mechanisms for reporting and handling racism and bullying in the workplace
  • The individual, organisational, and community effects of lateral violence
  • The importance and effects of community obligations
  • The Aboriginal and Torres Strait Islander experience of work-related stress and burnout

With a view to inform the Department of Customer Service’s Better Regulation Division Aboriginal Engagement and Education Strategy, these recommendations will help build a body of evidence to create safer working environments for Aboriginal and Torres Strait Islander People.

Want to know more?

To work with the Centre, or stay up to date with our research, head to our Engage with us page.