This unit will be a core elective in the Master of Artificial Intelligence to be introduced in 2020.

Human-centred AI represents a merger of two dominant trends in modern technology evolution: The rise of both (a) more human-centered systems, in terms of processing and interpreting the meaning of natural human communication, behavior, and bio-signal patterns; and (b) systems supported by powerful artificial intelligence tools and methods (e.g., machine learning, planning, symbolic reasoning). Rather than focusing exclusively on AI algorithms and techniques, or the development of fully autonomous systems, human-centred AI aims to harness AI technologies in support of human goals, activities and values. For example, such technology might support an end-user's goal of conserving finances when shopping, of engaging in activities to learn more effectively, or observing their values regarding preserving personal privacy. Human-centred AI also aims to develop synergistic human-machine systems in which users exercise some control in deploying the AI system, and they understand how the system works (i.e., via system transparency and explainability) and when it is not working properly (e.g., an error is occurring).

12/04/2019 - Adding Assessment breakdown.

On successful completion of this unit students should be able to:

1. critique AI only systems, transform them into human-centred AI systems and identify the motivation and benefits for doing the transformation.
2. propose and complete a project that represents a human-centred multidisciplinary AI model.
3. utilise tools for signal processing and interpretation required for behavioural analytics (e.g. for predicting users' social/emotional, cognitive, or health/mental health status).
4. critically evaluate empirical findings on the positive versus negative impact of AI technologies on human users and society, and reuse this knowledge to hypothesise the impact of future AI technologies.
5. recognise ethical, legal and regulatory challenges associated with developing different types of AI applications.

020119 Artificial Intelligence

This unit will explain how AI technologies are enabling more deeply human-centred design, including walking through illustrations of implementing predictive behavioral analytics and adaptive interface design in application domains like medicine and education. It will summarize the major design and development themes associated with implementing human-centred AI systems on current platforms, such as robotics, automotive, smartphones and wearables. Students will learn the philosophy, foundations, models, rationale and multidisciplinary origins of human-centred AI. Emphasis will be placed on students' critical analysis of AI technologies, based on an examination of their positive vs. negative impact on users and society. Students will hear from experts on the current ethical, legal, and regulatory challenges involved in developing prosocial AI systems, and guidelines for avoiding major pitfalls in these areas.

On-campus

Examination (2 hours): 40%, In-semester assessment: 60%

6

Minimum total expected workload equals 12 hours per week comprising:

• Two hours/week lectures
• Two hours/week laboratories

A minimum of 8 hours per week of personal study for completing lab/tutorial activities, assignments, private study and revision.

FIT5047, FIT5197