Scaling AI Literacy: A Design Framework for University Assessment Alignment

The article begins by explaining that the public release of large language models in late 2022 accelerated the use of generative artificial intelligence in higher education. Universities responded by creating policies, training programs, and new pedagogical practices to address both the opportunities and risks of GenAI.

The authors emphasize that institutional guidelines generally describe GenAI as both a learning opportunity and a challenge to academic integrity. At the same time, universities continue to face unresolved issues related to equity, data privacy, and staff capability.

AI literacy is presented as a graduate capability rather than a specialist technical topic. The article argues that many institutional responses remain fragmented because they rely on workshops, tool tips, prohibitions, or detection-based policies without embedding AI literacy into the curriculum.

The article highlights UNESCO’s 2024 AI competency frameworks for students and teachers as important references for universities. These frameworks emphasize human-centered values, ethics, foundational AI knowledge, pedagogical integration, and professional learning.

The study positions itself as an effort to translate global AI competency frameworks into discipline-appropriate learning outcomes and proficiency levels. It also examines how learning experiences and assessments can support integrity and authentic learning in AI-infused environments.

AI literacy is defined as the knowledge, skills, and dispositions needed to understand AI systems, use them appropriately, evaluate their outputs critically, and engage with them ethically and responsibly. The article notes that later reviews expanded this construct to include recognizing AI in context, applying tools, evaluating outputs, creating with AI, and addressing ethical issues.

The authors argue that GenAI literacy requires students to understand probabilistic text generation, hallucination risks, data provenance, socio-technical impacts, prompt workflows, verification practices, and ethical judgment. For universities, the central challenge is scaling these competencies across diverse disciplines and student populations.

The introduction concludes by framing academic integrity as a design problem rather than merely a detection problem. Since GenAI can generate plausible outputs that obscure misunderstanding, assessment should be redesigned to elicit authentic evidence of student thinking, process, and judgment.

This study uses a qualitative research design based on integrative synthesis and document analysis. The approach is suitable because it allows the researchers to examine diverse texts, interpret their purposes and contexts, and develop recommended guidelines for aligning curriculum with generative AI governance. The qualitative design supports contextual interpretation and the identification of emerging themes when translating complex competency frameworks into practical institutional models.

The data sources consist of secondary qualitative data, especially UNESCO’s student and teacher AI competency frameworks and recent sector guidance on GenAI. The researchers extracted competency dimensions and descriptors from these documents, mapped them to curriculum design principles involving learning outcomes, activities, and assessment, and organized the results into a functional alignment matrix. Trustworthiness was supported through triangulation across multiple sources, while ethical considerations were addressed by relying on publicly available secondary data.

The results present the Elevate Assessment Integrity Design framework and a rubric toolkit for authentic assessment in AI-infused learning environments. The framework connects learning outcomes, AI-use boundary conditions, and verification moments to help universities redesign assessment in ways that support both learning and integrity.

The Elevate-AI Alignment Model is built around five core principles for integrating AI competencies into higher education. These principles are human agency and responsibility; ethics, equity, and safety by design; progressive mastery; constructive alignment; and educator capability.

The first principle, human agency and responsibility, emphasizes that human decision-making and accountability must remain central when students use AI. This principle is operationalized through critical evaluation, reflective justification, and evidence that students can explain and defend their choices.

The second principle, ethics, equity, and safety by design, requires ethical reasoning to be embedded directly into learning tasks. Instead of treating ethics as an isolated topic, the model integrates bias awareness, privacy protection, data handling, and harm mitigation into assessment criteria.

The third principle, progressive mastery, organizes AI literacy development into staged proficiency levels: foundation, intermediate, and advanced. This allows universities to build AI literacy across a program rather than relying on isolated modules or optional training sessions.

The fourth principle, constructive alignment, requires learning outcomes, learning activities, and assessment evidence to be coherently connected. In this model, assessment should reward process transparency, situated judgment, and the student’s ability to explain how AI was used.

The fifth principle, educator capability, recognizes that scalable AI literacy depends not only on students but also on teachers. Staff development must be aligned with teaching roles and supported through reusable learning objects, assessment redesign resources, and professional learning opportunities.

The article provides an example alignment matrix linking UNESCO competency dimensions to curriculum-embedded learning outcomes and assessment evidence. The dimensions include human-centred mindset, ethics of AI, AI foundations, AI pedagogy or learning design, and AI for professional learning.

The model also proposes rubric-ready learning outcomes and proficiency levels. Foundation-level outcomes focus on recognizing errors, safe use, basic ethical issues, and basic disclosure of AI use. Intermediate outcomes emphasize structured verification, discipline-specific application, ethical frameworks, audit trails, and distinguishing AI contributions from student work.

Advanced outcomes involve designing evaluation protocols, auditing bias and robustness, leading ethical review, aligning practice with policy and regulation, and creating reproducible workflows or team documentation standards. These levels make AI literacy assessable across different years of study and institutional contexts.

The discussion argues that scalable AI literacy requires curriculum-embedded learning outcomes and teacher capability building. Universities that rely only on optional training may increase inequities because students with prior exposure to AI tools will benefit more than those without such experience.

The article concludes the discussion by emphasizing assessment redesign and governance as central levers for institutional change. By rewarding process transparency and verification, universities can encourage students to use AI as a learning tool rather than a shortcut, while governance must address disclosure norms, tool selection, data handling, intellectual property, and privacy.

This study established the Elevate-AI Alignment Model, a structured framework designed to integrate global AI competency standards into higher education curricula through five core principles: human agency, embedded ethics, progressive mastery, constructive alignment, and educator capability. By utilizing an integrative synthesis of UNESCO frameworks and current generative AI (GenAI) governance, the research translated abstract competencies into a practical alignment matrix and rubric-ready learning outcomes. The findings emphasize that AI literacy is not a standalone skill but a developmental process that requires transparency in assessment and a move toward evaluating the "process" of AI interaction rather than mere "output" detection.

The primary contribution of this research lies in its methodological translation of high-level policy guidance into actionable pedagogical tools, filling the gap between theoretical AI governance and classroom implementation. By providing a scalable model of proficiency levels—ranging from foundational safety to advanced workflow critique—the study offers a blueprint for institutions to mitigate inequities caused by ad hoc AI adoption. Furthermore, it advances the field of assessment design by promoting "situated judgment" and "process transparency" as robust alternatives to detection-based integrity strategies, aligning with broader shifts toward comprehensive AI assessment literacy.

Future research should focus on longitudinal pilot programs to validate the impact of these proficiency levels on diverse student cohorts and to refine the workload assumptions for educators implementing these frameworks. There is a critical need to explore how discipline-specific requirements—such as those in the creative arts versus STEM—might require specialized adaptations of the Elevate-AI alignment matrix. Additionally, future inquiries should investigate the intersection of AI literacy with evolving data rights and intellectual property regulations to ensure that institutional governance remains rights-respecting and ethically resilient as GenAI technologies continue to proliferate.

Allen, L. K., & Kendeou, P. (2023). ED-AI Lit: An Interdisciplinary Framework for AI Literacy in Education. Journal of Educational Computing Research. https://doi.org/10.48550/arxiv.2305.10688

Annapureddy, R., Fornaroli, A., & Gatica-Perez, D. (2024). Generative AI Literacy: Twelve Defining Competencies. Digital Government: Research and Practice, 5(2), 1–18. https://doi.org/10.1145/3685680

Azungah, T. (2018). Qualitative research: deductive and inductive approaches to data analysis. Qualitative Research Journal, 18(4). https://doi.org/10.1108/qrj-d-18-00035

Chan, C. K. Y. (2023). A comprehensive AI policy education framework for university teaching and learning. International Journal of Educational Technology in Higher Education, 20(1), 38. https://doi.org/10.1186/s41239-023-00408-3

Cheong, H., Lyons, A., Houghton, R., & Majumdar, A. (2023). Secondary Qualitative Research Methodology Using Online Data within the Context of Social Sciences. International Journal of Qualitative Methods. https://doi.org/10.1177/16094069231180160

Chiu, T. K. F., Ahmad, Z., Ismailov, M., & Sanusi, I. T. (2024). What are artificial intelligence literacy and competency? A comprehensive framework to support them. Computers and Education: Artificial Intelligence, 6, 100214. https://doi.org/10.1016/j.caeai.2024.100214

De Silva, D., Jayatilleke, S., El-Ayoubi, M., Issadeen, Z., Moraliyage, H., & Mills, N. (2024). The Human-Centred Design of a Universal Module for Artificial Intelligence Literacy in Tertiary Education Institutions. Education and Information Technologies. https://doi.org/10.1007/s10639-024-12628-w

Elbardan, H., & Kholeif, A. (2017). An Interpretive Approach for Data Collection and Analysis. In Model-Based Systems Engineering. https://doi.org/10.1007/978-3-319-54990-3_5

Farrelly, T., & Baker, N. (2023). Generative Artificial Intelligence: Implications and Considerations for Higher Education Practice. Journal of University Teaching & Learning Practice, 20(2). https://doi.org/10.53761/1.20.2.1

Francis, N. J., Jones, S., & Smith, D. P. (2025). Generative AI in Higher Education: Balancing Innovation and Integrity. International Journal of Educational Integrity. https://doi.org/10.1186/s40979-025-00124-x

Gephart, R., & Saylors, R. G. (2020). Qualitative Designs and Methodologies for Business, Management, and Organizational Research. Oxford Research Encyclopedia of Business and Management. https://doi.org/10.1093/acrefore/9780190224851.013.230

Kazanskaia, A. (2025). Data Analysis and Interpretation in Qualitative Research. NEYA Global Journal of Non-Profit Studies. https://doi.org/10.64357/neya-gjnps-ql-rchfr-mth-08

Moorhouse, B. L., Yeo, M. A., & Wan, Y. (2023). Generative AI tools and assessment: Guidelines of the world's top-ranking universities. Computers and Education Open, 5, 100151. https://doi.org/10.1016/j.caeo.2023.100151

Morgan, H. (2022). Conducting a Qualitative Document Analysis. The Qualitative Report, 27(1). https://doi.org/10.46743/2160-3715/2022.5044

O’Dea, X., Ng, D. T. K., O’Dea, M., & Shkuratskyy, V. (2024). Factors affecting university students’ generative AI literacy: Evidence and evaluation in the UK and Hong Kong contexts. British Journal of Educational Technology. https://doi.org/10.1111/bjet.13454

Okada, A., Sherborne, T., Panselinas, G. E., & Kolionis, G. (2025). Fostering Transversal Skills Through Open Schooling Supported by the CARE-KNOW-DO Pedagogical Model and the UNESCO AI Competencies Framework. International Journal of Artificial Intelligence in Education. https://doi.org/10.1007/s40593-025-00458-w

Ruggiano, N., & Perry, T. E. (2017). Conducting secondary analysis of qualitative data: Should we, can we, and how? Qualitative Social Work, 18(1). https://doi.org/10.1177/1473325017700701

Sanusi, I. T., Ahmad, Z., Ismailov, M., & Chiu, T. K. F. (2024). UNESCO’s AI competency framework for teachers: A normative reference for educator preparation. Journal of Digital Learning in Teacher Education. (Note: Data disesuaikan dari entitas UNESCO/Chiu et al pada CSV).

Xia, Q., Weng, X., Fan, O., Lin, T. J., & Chiu, T. K. F. (2024). A scoping review on how generative artificial intelligence transforms assessment in higher education. International Journal of Educational Technology in Higher Education, 21(1). https://doi.org/10.1186/s41239-024-00451-w