Designing Critical Human–AI Learning in Higher Education: The SC-AIM–FAVIC Framework for Analytical and Logical Reasoning

Generative artificial intelligence (AI) is now routine in university study, yet unstructured use risks cognitive offloading, surface learning, and uncritical acceptance of fluent but unreliable output. This article aimed to develop and specify the Symbiotic Critical Human–AI Learning Model (SC-AIM), operationalised through the five-stage FAVIC framework (Formulation, Augmentation, Verification, Integration, Creation), and to set out the assessment instruments and evaluation design that accompany it. Following the Analysis and Design phases of the ADDIE research-and-development approach, we synthesised recent experimental, meta-analytic, and survey evidence to derive design requirements, translated them into a stage-by-stage instructional specification, built a process-sensitive assessment rubric, and defined a quasi-experimental and qualitative validation plan for Indonesian higher education. The results are the model and its components: an evidence-to-design mapping, the FAVIC stage specification, a three-level process-sensitive rubric, and an evaluation design with hypotheses. Conclusion: SC-AIM offers lecturers a concrete, theory-grounded sequence that places human reasoning before and after AI assistance and makes verification an assessed activity, giving higher education a ready-to-use pedagogical model for developing analytical and logical reasoning in the age of generative AI.