The pervasiveness of Artificial Intelligence (AI) technologies is increasingly evident in all facets of lives. Particularly in the contemporary business world, AI is being extensively used to support business managers in decision-making either in a primary or secondary role.
The autonomy of AI systems ranges from semi-autonomous to fully autonomous. As the pace of AI autonomy advances using vast amounts of data and feature engineering, their appearance is getting opaquer and more incomprehensible for humans. This phenomenon generates an imminent need to develop mechanisms to enhance the explainability of AI, especially for users and those likely to get affected.
This PhD will develop a framework to identify the analytical rationale behind AI systems. Through an extant review of academic and practitioners’ literature, the first deliverable of the PhD will be a holistic summary of the current issues around the explainability of AI and its implications in autonomous decision-making.
The second deliverable is expected to be a transdisciplinary experiment-based investigation where guidance from various disciplines such as data science, social science, computer science, and organisational behaviour is synergised to understand the explainability of AI and its impact on autonomous decision-making in an organisational setting.
The expected outcome will be a framework for autonomous decision-making through explainable AI. Lastly, the developed framework will be tested through various case studies, conducted together with a PhD Internship at Infonyx Pty Ltd, a data solutions company who was recently recognised in the Financial Times as one of the high-growth companies in the Asia-Pacific region.
