The Syllabus Is Not the Course: AI, Course Design, and the Junk Playground

Generative AI can produce a "complete-looking" university course in minutes: learning outcomes, weekly topics, readings, assignments, rubrics. For anyone who designs courses, that is both an extraordinary capability and a quiet trap. In Episode 107 of Modem Futura, Sean Leahy and Andrew Maynard reflect on more than a year of using AI as a private instructional designer, and on why the most important part of a course may be the part these systems are least able to build.

From sounding board to finished product

Used as a thinking partner, AI can be remarkably helpful. It can test whether a sequence holds together, surface readings no single person would find, and catch overlap between weeks. It is also good at turning existing material into worksheets or self-contained interactive pages that work around the limits of a learning management system. The trouble starts when the tools move from responding to finishing. Ask for help thinking through a course and, increasingly, they return a complete one: plausible, polished, and conventional. These systems are very good at reflecting what has worked before and much weaker at imagining what hasn't been tried (which is no surprise, but increasingly frustrating).

Content is abundant; journeys are designed

The distinction the hosts keep returning to is between content and journey. What remains scarce is the intentional design of experience, the sequence of challenges and moments of realization that carries a learner somewhere new. It echoes John Dewey's argument that education should build on learners' natural impulses toward inquiry, communication, construction, and expression rather than fill them with facts. A course that a single AI prompt could reproduce may be telling its designer something.

A junk playground for the AI age

The episode's most generative idea comes from 1943, when a playground designed by Danish landscape architect Carl Theodor Sørensen opened in Emdrup, Copenhagen. Sørensen had noticed that children were more drawn to piles of scrap than to the playgrounds he designed, so he gave them a space full of loose materials and a play leader to support rather than direct them. He later described it as both the ugliest and the best of his works. Andrew is now exploring what a digital junk playground might look like: a curated space where students build with whatever AI tools they choose and learn through the making.

For educators, that points to a different role, less authority over learning and more what Sean calls an elevated co-participant in it. When nobody yet knows the rules for learning alongside AI, play, experimentation, and care may be among the most rigorous approaches available.

Listen to the full conversation on Modem Futura.

Subscribe and Connect!

Subscribe to Modem Futura wherever you get your podcasts and connect with us on LinkedIn. Drop a comment, pose a question, or challenge an idea—because the future isn’t something we watch happen, it’s something we build together. The medium may still be the massage, but we all have a hand in shaping how it touches tomorrow.

🎧 Apple Podcast: https://apple.co/4yZg4X0

🎧 Spotify: https://open.spotify.com/episode/55Z17htCK9S3HKDEGDQd7w?si=D99hOJiUQdOK7jaKn5kHJA

📺 YouTube: https://youtu.be/Ob6PjC4tGVg

🌐 Website: https://www.modemfutura.com/