As generative AI tools become embedded in the everyday routines of students and educators alike, higher education faces a defining assessment question: Is what we are measuring actually learning? The reflexive answer from many institutions has been to pivot quickly to "authentic assessment" as the silver bullet — but the research increasingly shows that authenticity alone is not enough, and that poorly designed authentic tasks can still be completed by AI with surprising fluency.
This discussion session challenges practitioners to examine a more uncomfortable truth: that the real value of authentic assessment is not found in circumventing AI, but in forcing a reckoning with what genuine competency demonstration looks like — and in demanding that students actually own that demonstration. Drawing on findings from a qualitative, phenomenological research study of 12 competency-based education (CBE) program administrators across the United States, this session presents practitioner-validated evidence that the most effective programs have long embedded authentic, mastery-based assessment at their core — not as an AI countermeasure, but as a foundational pedagogical and organizational principle aligned to workforce expectations and student success.
Participants will explore the critical difference between tasks that merely look authentic and those assessments that require authentic performance, examine how CBE program leaders have designed and sustained mastery-based assessment systems, and grapple with the real design challenges: faculty workload, grading consistency, rubric development, equity/access, and continuous improvement. The session invites honest conversation about shortcuts — the ones students are taking with AI, and the ones faculty and institutions are tempted to take with assessment design. Participants will leave with a framework for rethinking their own assessment approaches, grounded in both applied research and the real-world experiences of CBE program leaders navigating these same tensions. "AI hasn't created the assessment problem — it has exposed it."
