Beyond the Right Answer: A model for integrating AI and cognitive bias into problem-based learning

Students are already using AI. Recent surveys show 95% use it in their studies, and about half of student AI conversations ask for the finished answer rather than help with the thinking. Meanwhile, the profession is moving the other direction: as AI absorbs routine work, the premium on judgment keeps rising. This session presents a practical response. We reframe accounting as a judgment profession and teach it that way, using problem-based learning cases that build the critical thinking and domain knowledge students need to direct AI rather than defer to it. Along the way, we use the classroom itself to make cognitive biases visible and teachable. You will experience the method firsthand through a live case, then leave with a simple session structure and materials you can drop into your own course.