Let's be honest: Trying to make assignments “AI-proof” is like trying to write a “calculator-proof” math problem.

With that in mind, we explore how to design AI-ready assessments that reward genuine understanding and insight when answers are cheap and instant.
Alex and Dr. Aliza unpack a college course that embraces AI rather than hides from it. Tulane University associate professor Nick Mattei walks us through a term project where his students prompt multiple AI models, compare outputs, and critique errors before writing drafts and transforming their essay into another medium. The plot twist: Nick's assignment requires students to defend their choices in-person! That one change re-frames the assignment so students don't try to conceal AI use, and instead spend more time learning the material well enough to explain it to others.
From there, we sit down with Shantanu Sinha, founding president of Khan Academy and now VP and GM of Google for Education. Shantanu argues that AI shifts the spotlight from product to process. For instance, he draws a sharp line for high-stakes essays: AI can suggest structure, but only the student can supply authentic voice. The goal isn’t to ban technology, it’s to cultivate critical thinking, problem-solving, creativity, communication, and teamwork — skills that will outlast any AI or edtech tool.
If you’re a teacher, parent, or curious learner, this episode will leave you with concrete strategies to re-think homework, re-wire assessments, and turn AI from a crutch into a scaffold.

aiEDU: The AI Education Project

• aiEDU.org
• linkedin.com/company/aiedu/