UQ researchers are investigating how AI and automation can support better decision making while helping people maintain the skills, judgement and expertise needed to work effectively, safely and fairly in a digital age.
What's the problem?
AI-powered decision support tools are increasingly being used across workplaces and industries. While these technologies can improve efficiency and performance, there are growing concerns that overreliance on AI and automation may erode important human skills and expertise over time.
Research has shown that when decision-making is increasingly delegated to automated systems, workers can lose opportunities to practise, maintain and develop critical capabilities. Importantly, the skills that erode are often the very same skills needed to evaluate outputs, identify errors or respond effectively when automated systems fail.
As AI becomes more deeply embedded in workplaces, organisations face an important challenge: how can people benefit from AI assistance while continuing to develop and maintain their own expertise?
What's UQ's innovation?
Led by Professor Tim Miller from UQ's School of Electrical Engineering and Computer Science, with Associate Professor Ida Someh, Dr Tapani Rinta-Kahila and Dr Mint Tanprasert, this research investigates how AI-driven decision support tools can be designed to enhance human capability rather than diminish it, drawing on expertise in human-centred AI, explainable AI and organisational transformation.
Rather than assuming skill erosion is an inevitable consequence of AI adoption, the researchers are exploring how interaction design, explainable AI, human-AI collaboration, and organisational work design can help people remain engaged in decision-making processes while continuing to build and apply their expertise.
Drawing on systems thinking, case studies and action design research, the team is investigating how AI shapes workers' skills over time and developing new ways to support high-quality decision making while maintaining essential human capabilities.
The work examines how AI systems can provide effective support while encouraging users to retain critical knowledge, exercise judgement and maintain the skills needed to perform complex tasks independently.
What's the impact?
The work aims to contribute to a new systems-level understanding of skill erosion and inform evidence-based frameworks for managing AI in ways that protect and strengthen human expertise.
The findings could help organisations strike a better balance between automation and human capability, enabling them to realise the benefits of AI while supporting worker wellbeing, maintaining critical skills and maximising human potential in the age of AI.
Did you know?
This research challenges the common assumption that AI will inevitably reduce human skills. Instead, it explores how AI systems can be designed to help people make better decisions while continuing to develop and maintain their expertise.
Led by
Key contributors
| AI Research Strength | Human-Centred AI AI Strategy & Governance |
|---|---|
| Industry Portfolio | Advancing AI |
| Key Partners | ARC Centre of Excellence for Quality Work in a Digital Age (QWiDA) |
| Key Publications | Gaube, Susanne and Langer, Markus and Miller, Prof. Tim and Baum, Kevin and Dachselt, Raimund and Feit, Anna Maria and Gadiraju, Ujwal and Kaur, Harmanpreet and Keane, Mark T. and Landers, Richard N. and Laux, Johann and Liao, Q. Vera and Lim, Brian and Onnasch, Linda and Schrills, Tim and Sonenberg, Liz and Tan, Chenhao and Tintarev, Nava and Xiao, Ziang and Zhang, Hanwei, Keeping an Eye on AI: A Framework for Effective Human Oversight of AI Systems (March 31, 2026). Available at http://dx.doi.org/10.2139/ssrn.6576159 |
| Related Project | Rinta-Kahila, Tapani, Penttinen, Esko, Salovaara, Antti, Soliman, Wael, and Ruissalo, Joona (2023). The vicious circles of skill erosion: a case study of cognitive automation. Journal of the Association for Information Systems 24 (5) 2 1378-1412. https://doi.org/10.17705/1jais.00829 |
Published 15 September 2026