Conference Agenda

Higher education institutions (HEIs) face increasing pressure to respond rapidly to changing learner needs, technological disruption, and evolving labour market expectations. Artificial intelligence (AI) and Agile methodologies have emerged as promising approaches to enhance responsiveness, yet they are typically adopted in isolation. This paper introduces the AI Enabled Agile University (AIAU) Model, a ‘four layer’ conceptual architecture that integrates institutional value streams, Agile practices, AI capabilities, and enabling conditions to support continuous improvement across teaching, learning, assessment, and student support. The model is derived from a synthesis of literature on Agile in education, AI enabled learning systems, and digital transformation frameworks. The model is supported by a clear rationale for the four-layer architecture and the interactions between value streams, Agile practices, AI capabilities, and enabling conditions. Two illustrative use cases demonstrate how the model can support curriculum iteration and student success analytics. This paper contributes a theoretically grounded conceptual model that unifies AI and Agile practices at an institutional level. The AIAU Model provides a structured foundation for future empirical research and offers HE leaders a coherent framework for AI driven organisational agility.

Artificial intelligence increasingly shapes higher education, yet most evaluations focus on average learning outcomes and overlook differences across student groups such as gender.
This paper presents a secondary analysis of a controlled experimental study comparing a scaffolded, institutionally embedded tutoring system (IRIS), an unrestricted conversational AI tool (ChatGPT), and a no-AI condition. The original study found that AI support improved task performance but did not consistently improve learning outcomes. This analysis extends the model by testing whether effects differ by gender.

An ANCOVA on posttest scores controlling for pretest revealed a small but significant Condition $times$ Gender interaction (F(2, 383) = 3.55, p = .030; f² = .019), driven by the ChatGPT condition: female students showed higher adjusted posttest scores than the no-AI control (d ≈ +0.28), whereas no comparable effect emerged for male students (d ≈ -0.19) or in the IRIS condition. The pattern held across six sensitivity analyses. No comparable interaction emerged for exercise performance, intrinsic motivation, or cognitive load.

These findings suggest that AI-supported learning environments are not uniform in their effects across student groups. The two AI tools differ along several dimensions simultaneously, so the present design cannot isolate which dimension drives the moderation; this analysis adopts 'how AI structures access to help' as an interpretive framing motivated by prior work on stereotype threat and ambient belonging rather than as an identified mechanism.

Generative artificial intelligence is rapidly becoming a pervasive cognitive infrastructure in education, reshaping how problems are framed, explored, and evaluated across disciplines.
While these developments create new learning opportunities, they also raise concerns about epistemic automation and cognitive offloading in AI-mediated environments. Current educational responses often focus on tool adoption or prompt-engineering skills, which remain insufficient to address the deeper regulatory challenges introduced by large language models.
This paper reconceptualizes Natural Language Programming as cognitive and metacognitive programming: the intentional design and regulation of reasoning processes through language-mediated interaction with generative systems. To operationalize this perspective, we propose the Minimal Dialogic Cognitive Programming Architecture, a domain-independent framework composed of five dialogic cognitive primitives (Framing, Decomposition, Constraint and Criteria Control, Dialogic Validation, and Reflective Regulation) organized as a recursive regulatory cycle. Prompt-Based Learning is presented as the pedagogical instantiation of this architecture, reframing prompts as programs of thought rather than commands for output optimization.
Preliminary observations from a Master’s course suggest that students frequently engage in co-reasoning with generative systems, highlighting the need for assessment approaches capable of capturing AI-mediated reasoning processes.

This study examines the cognitive factors shaping student trust in ChatGPT-generated mathematical word problem solutions. A within-subject design was employed to assess how perceived reliability, task characteristics, and information quality influence trust across three scenarios: ChatGPT, disclosed Human–AI, and non-disclosed Human–AI. Results indicate that trust is context-dependent and varies across scenarios, with different information quality dimensions influencing trust depending on transparency. Reliability and correct solutions were the strongest predictors of trust in the ChatGPT scenario, while conceptual errors dominated trust assessment when ChatGPT use was undisclosed. Math anxiety was negatively correlated with accurate error classification, raising concerns about students' ability to critically evaluate AI-generated content. These findings suggest that cognitive trust in AI is situational and that disclosure of AI use meaningfully shifts how students attribute and assess information quality.