What if, instead of trying to prevent students from using AI, we designed assessments that require collaboration with it? At a small liberal arts college, mathematics assessment for non-majors was reimagined through alternative grading and structured revision cycles. When assessment shifts from point accumulation to iterative feedback and proficiency-based revision, students begin to use AI not to produce answers, but to ask better questions, test ideas, and clarify their thinking. This session explores the evolving partnerships between students, instructors, and AI in redefined learning spaces, striving to increase a sense of belonging. Let’s explore redesigning assessment to sustain rigor, preserve human connection, and invite authentic mathematical curiosity in the age of generative AI.