Senior Lecturer Dr. Jason Zagami

Griffith University, Australia

Bio: Dr Jason Zagami is a Senior Lecturer in the School of Education and Professional Studies at Griffith University, Australia. His teaching and research focus on artificial intelligence in education, digital technologies curriculum, virtual and extended-reality learning environments, educational gaming, and teacher education. His work examines how emerging technologies can be used thoughtfully to strengthen learning, creativity, inclusion, and teacher capability while addressing the ethical and practical challenges they introduce. Dr Zagami leads the research focus of the Australian Technologies Teacher Educators Network (ATTEN), the Griffith Research on Artificial Intelligence in Education (GRAIL), and the Education, Change and Sociotechnical Futures Research (ECSFR) groups. Bringing together research, policy, and educational practice, he works to develop rigorous, responsible, and human-centred approaches to digital transformations in education and society.

Talk Title：International Trends in Digital, Data, and AI Literacy through Initial Teacher Education
Abstract：Initial Teacher Education (ITE) provides a valuable lens through which to examine global trends in digital, data, and artificial intelligence literacy. Unlike many areas of educational practice, ITE is generally supported by an examinable documentary architecture comprising teacher standards, accreditation requirements, programme specifications, course documentation, and practicum assessment materials. These sources do not fully represent enacted practice, but they make it possible to compare what education systems formally expect future teachers to learn, demonstrate, and apply.
Drawing on a large cross-national documentary analysis using the Global ITE Digital, Computing, and AI Preparation Scale (G-IDCAPS), this presentation examines how digital, data, and AI-related capabilities are being incorporated into professional preparation. The evidence suggests that foundational digital literacy and the pedagogical use of digital technologies are increasingly visible across jurisdictions. Preparation in computational thinking, digital assessment, data interpretation, and AI literacy remains less consistent and is rarely connected to assessed performance during professional experience. Ethical principles, safety, and governance are becoming more prominent, but often remain general expectations rather than demonstrated professional capabilities.
ITE therefore offers more than a view of teacher preparation: it provides an institutional vantage point from which to identify wider international trajectories in educational capability, regulation, and professional practice.

Dr. Kazem Banihashem

Maastricht University, Netherlands

Bio: Dr. Kazem Banihashem is an Assistant Professor at the Department of Educational Development and Research, School of Health Professions, Maastricht University, the Netherlands. He is an internationally recognized scholar and active contributor to the field of AI in Education, with a research program centred on the intersection of AI and feedback, human–AI collaboration, hybrid intelligent feedback, and the responsible use of AI in feedback practices. His broader research portfolio spans learning analytics, assessment, self-regulated learning, learning theory, and learning design. Dr. Banihashem serves as an Associate Editor of the Australasian Journal of Educational Technology (AJET) and has served as Guest Editor for several international journals, including the International Journal of Educational Technology in Higher Education and the Journal of Computer Assisted Learning. He has delivered invited keynote and panel presentations on human–AI collaboration and the future of feedback practices. His work has also attracted competitive research funding from national funding bodies, including the Dutch National Organization for Education Research (NRO), for projects advancing responsible and meaningful uses of AI in education.

Talk Title：Human-AI Collaboration for Meaningful Feedback Practices
Abstract：Generative AI has increasingly been recognized as a powerful source of feedback in education. However, much of the existing literature has focused on whether AI-generated feedback is comparable or superior to human-generated feedback. This emphasis risks reducing feedback to the mere provision of information, overlooking the processes through which students engage with, interpret, and act upon feedback. In this presentation, I will critically examine emerging approaches to human–AI collaboration in feedback practices, highlighting their potential and limitations. Building on this discussion, I will introduce a human–AI collaborative feedback process model that moves beyond feedback provision to consider how humans and AI can collaborate throughout the feedback process. Finally, I will discuss the roles of AI literacy and feedback literacy in shaping effective, meaningful, and responsible human–AI feedback collaboration.

Prof. Mingming Zhou

University of Macau, Macau, China

Bio: Dr. Mingming Zhou is now Professor in Faculty of Education at University of Macau. She currently serves as Associate Dean (Research and Graduate Studies) in the faculty. Her research combines educational psychology and educational technology to investigate how university students self-regulate their learning both in online and offline settings. Her current research examines how students interact with AI to support their self-regulation.

Talk Title：From Self-Regulation to Hybrid Regulation: Conceptualizing and Operationalizing Human-AI Shared Control in Learning
Abstract：Self-regulated learning (SRL) has long positioned the learner as the sole agent responsible for planning, monitoring, and reflecting on their learning. The integration of artificial intelligence into educational environments, however, is fundamentally destabilizing this assumption. Adaptive systems increasingly perform regulatory functions by diagnosing knowledge gaps, generating feedback, and recommending strategies, which were once the exclusive domain of human learners. This raises a critical question: what happens to SRL when regulation becomes a shared endeavor between human and artificial agents?
In this talk, I will introduce an integrated framework for AI-mediated self-regulated learning as a conceptual and operational framework, by integrating models of Molenaar (2022), Hadwin and Oshige (2011) and Zimmerman (2002) to (a) distinguish different modes of learner-AI interaction in regulation, (b) map these onto specific SRL phases, and (c) account for varying degrees of learner agency. Regulation in this model is not treated as a fixed human capacity, but as a dynamic system of shared control that can fluidly shift between learners and AI across the SRL cycle. The framework specifies four distinct regulatory modes: AI-dominated regulation, learner-AI co-regulation, learner-AI shared regulation, and AI-assisted regulation. It articulates the conditions under which control is productively transferred between AI and the learner. It also identifies the critical challenge of metacognitive offloading, wherein the role of AI in the development of self-regulatory skills becomes more complicated than expected.
Operationalizing different modes of learner-AI interaction in regulation requires moving beyond self-report measures to trace-based instrumentation capable of detecting real-time SRL processes and triggering adaptive scaffolding. I will use the FLoRA Engine as an example to demonstrate how learning analytics and GenAI can be integrated to detect when learners disengage from regulation and to intervene with personalized, theory-grounded support. I will conclude by outlining the empirical, pedagogical, and ethical questions that remain unresolved, most notably, how to design hybrid systems that know when to assist, when to step back, and how to ensure that the regulatory handover genuinely serves learner development rather than algorithmic convenience.