Artificial intelligence has intensified a question many educators were already beginning to ask: What evidence truly demonstrates learning? For decades, higher education has relied on products—papers, exams, presentations, and projects—as proxies for student understanding. Generative AI challenges this assumption by making it increasingly difficult to determine whether a finished product reflects a student's own thinking. While many institutional responses have focused on detection, surveillance, or restrictions, this workshop explores a different possibility: What if AI is not simply creating an assessment problem, but revealing one that has existed all along? Drawing on experiences redesigning undergraduate and graduate courses through ungrading, authentic assessment, and AI-integrated pedagogy, participants will examine how process-centered approaches can make learning more visible while preserving student agency. Through examples of self-assessment, reflective learning narratives, revision cycles, and AI transparency practices, attendees will explore ways to shift attention from evaluating products to cultivating evidence of growth, judgment, and engagement. Participants are encouraged to bring an existing assignment or assessment practice to the workshop. Using a guided redesign framework, they will identify opportunities to incorporate reflection, process documentation, feedback loops, and authentic engagement in ways that center human learning in AI-rich environments. Situated within the liberal arts tradition, this workshop embraces the paradox that AI simultaneously challenges and clarifies the value of education. As machines become increasingly capable of generating polished outputs, the most meaningful aspects of learning may be those that are difficult to automate: curiosity, ethical reasoning, self-awareness, intellectual risk-taking, and the ability to make meaning from experience.
