Teaching beyond AI: Gamified problem-based formative assessment aligned with UNESCO AI competency frameworks
Year: 2026
Type: article
Abstract: The arrival of capable generative AI (GenAI) tools has posed a genuine and largely unresolved challenge for summative assessment in higher education. When AI-generated submissions evade institutional detection 94% of the time and examination scores rise sharply without any change in teaching, the evidentiary basis of take-home and invigilated assessment is genuinely in question. This paper draws on two lines of work to argue that classroom-based gamified problem-based formative assessment (GPBFA) offers a more durable response than detection-focused approaches. The first line is an empirical study of saval.app, a GPBFA platform evaluated through Keller’s ARCS motivational model across a full university semester (pre n = 41, post n = 44). Results across all four ARCS dimensions were encouraging (all p < .0001; Cronbach’s alpha > .93; composite 81% positive response rate), with near-zero motivational decline across 12 weeks. The second line is a review of UNESCO’s 2024 AI Competency Frameworks for Teachers and Students, which we argue are philosophically and pedagogically aligned with classroom GPBFA in ways the existing literature has not yet made explicit. Taken together, we suggest that moving assessment into the classroom — where it is live, observed, and problem-centred — may address both the integrity challenge and the AI literacy imperative that UNESCO frameworks define.
Source: ASCILITE Publications
Institutions: Higher Colleges of Technology, Ajman University
Language: English
Subfield: Developmental and Educational Psychology
Field: Psychology
Domain: Social Sciences
SDG: Quality education
Open Access status: diamond
OpenAlex ID: W7233776422
