Why we have formed
The AI Forum has formed a working group to draft a white paper on Explainable AI (XAI). The working group comprises subject matter experts from across Aotearoa and aims to provide an examination of the application areas, model types, challenges, and governance considerations associated with XAI.
Explainable AI (XAI) also known as Interpretable AI, or Explainable Machine Learning (XML), refers to the field of artificial intelligence that aims to make AI models, systems, and their decision-making processes more understandable, interpretable, and transparent to humans. XAI hopes to help users of AI-powered systems perform more effectively by improving their understanding of how those systems reason.
Traditional AI models, especially deep learning systems, are often considered “black boxes” because their inner workings and decision-making processes are hard to understand even for experts. This lack of interpretability can lead to issues related to trust, accountability, and adoption of AI systems.
The working group will produce a white paper which will explore a range of questions including:
- What model types can be replaced by Interpretable models?
- What problems could be modelled using interpretable or explainable models?
- How far can explainable techniques go to explain our most complex models including Deep Learning, Generative models, Large Language Models.
The AI Forum New Zealand facilitates AI advancement for Aotearoa New Zealand across a multi-stakeholder environment. Members are encouraged to consider different points of view when formulating projects, to provoke conversation within the working groups and cultivate open critique and reflection.
Overall, the working group aims to provide insights and guidance to organisations and individuals considering or already using XAI in New Zealand to ensure they do so ethically, responsibly, and sustainably.
The Forum’s intent for the working group is to use the Forum’s power to convene and bring together knowledge from across the AI ecosystem to collate and bring to life AI best practices and guidelines for its use in Aotearoa, New Zealand.
Purpose
The group has been formulated to address the following:
- To provide expert information to organisations in Aotearoa New Zealand on XAI and its application including what problems could be modelled using interpretable or explainable models.
- To provide an analysis of whether Te Ao Māori may be embedded in XAI adopted in New Zealand.
- To provide guidance on the responsible governance of XAI.
Deliverables
White Paper on XAI for New Zealand organisations
- Objectives: The objective of the white paper is to provide a comprehensive overview of opportunities, benefits, challenges, and governance considerations associated with XAI in New Zealand. This will help organisations and individuals make informed decisions about how to use XAI responsibly and ethically.
- Audiences: The white paper will be delivered to the NZ AI Forum and other relevant industry groups, as well as to academic institutions and government agencies. It will also be made available to the broader public interested in understanding the impact of XAI in New Zealand.
Get involved
The Explainable Working Group meets regularly and works together to produce advice on practical tools for New Zealand organisations and their leaders. If you are interested in getting involved and leading some of that work, please get in touch here.