Image credit: Microsoft PowerPoint Stock Images
Higher education is experiencing another shift in the conversation about artificial intelligence (AI) as it exposes the limitations of integrity approaches focused primarily on risk, compliance and detection. While these concerns remain important, AI has given us the opportunity to rethink integrity, assessment, and learning in a world where technology can perform tasks once considered evidence of human achievement.
Higher education institutions have a responsibility to develop ‘AI-ready’ graduates and maintain the integrity of learning and assessment. As attention moves beyond risk and compliance, universities are increasingly focused on preparing students to use AI responsibly in their studies and future professional lives. Increasingly, educators and students must make context-dependent decisions about how, when, and why AI should (or not) be used.
Recent scholarship is placing increasing emphasis on human capabilities in responsible AI use. Whether described as human capabilities (OECD, 2026), transferable adaptive capabilities (Lodge et al., 2026), or durable capabilities (Bertram Gallant & Rettinger, 2026), there is growing recognition that graduates require more than disciplinary expertise when working with AI.
Yet, capabilities alone will not guarantee responsible behaviour. Individuals may possess sophisticated capabilities, such as thinking critically, managing complexity or leading collaborations, but still make poor decisions about how those capabilities are applied. Knowing how to use AI well is not the same as knowing when it should be used, why it should be used, and whether its use is appropriate in a particular context.
This suggests that two important elements are being overlooked, namely, human values and sound judgement. Developing responsible AI users requires recognising the interconnected roles of values, capabilities, judgement, and behaviour as outlined in the four-step model in Figure 1.
Figure 1: A Four-Step Model for Responsible AI Use
Capabilities (Step 2) do not exist in isolation but are guided by the values that individuals hold. Human values (Step 1) such as truthfulness, integrity, courage, excellence, respect, empathy, and accountability influence our understanding of the responsibilities and the choices we make.
Yet even values and capabilities are not enough because judgement (Step 3) is also needed to evaluate a situation or context and to weigh up differing opinions or options before determining an appropriate course of action. Judgement helps navigate ambiguity, balance competing priorities, and make decisions that align with both personal values and broader ethical responsibilities. The question is no longer ‘Can I use AI?’ but ‘Should I use AI here, and why?’ These are questions of ethics and judgement not just acting on a capability.
The fourth and final step is the practice of responsible use, which is the visible outcome of how values, capabilities, and judgement interact in a situation. Too often, responsible AI use is treated as an end goal that can be achieved through policies, guidelines or capability development alone. However, responsible behaviour does not emerge in isolation. It is the culmination of a series of interconnected steps, beginning with values, strengthened through capabilities, and mediated by judgement.
As universities continue to develop frameworks for responsible and ethical use of AI, it may be timely to look beyond capability and pay greater attention to the two missing links that connect capability with responsible behaviour: values and judgement. While capability frameworks help define what students need to know and be able to do, they do not fully explain how responsible decisions are made. We must also focus on cultivating the values and judgement that guide the application of capabilities. Only then can we move beyond compliance and towards the development of trustworthy, responsible AI users.
References
Bertram Gallant, T. & Rettinger, D. (2026). Going forward with integrity, staying human. International Center for Academic Integrity Summer Intensive Series (Season 2) Webinar 5. Summer Intensive Series (Season 2) Biweekly webinar series May-July 2026
Lodge, J.M. & Associates. (2026). Assuring quality learning in a gen AI-integrated future: The role of adaptive capabilities. Tertiary Education Quality and Standards Agency, Australian Government. Assuring quality learning in a gen AI-integrated future: The role of adaptive capabilities
Organisation for Economic Co-operation and Development. (2026). OECD Digital Education Outlook 2026: Exploring Effective Uses of Generative AI in Education. OECD Publishing. OECD Digital Education Outlook 2026 | OECD
Dr Christine Slade is Associate Professor in Higher Education and Academic Lead: Assessment and Academic Integrity in the Institute for Teaching and Learning Innovation (ITaLI) at the University of Queensland, Australia.
The author’s views and conceptual framework are her own. Microsoft Copilot was used as a writing support tool, with all suggestions subject to the author’s expertise, judgement, and final approval.
Thank you for being a member of ICAI. Not a member of ICAI yet? Check out the benefits of membership and consider joining us by visiting our membership page. Be part of something great!