Teaching Computer Science When AI Can Write the Code

John Boyland

What should students learn, and how should instructors assess that learning, when AI can produce much of the work students were once expected to do themselves?

In our November Active Teaching Lab, John Boyland will share how AI has led him to rethink how he teaches computer science. Traditionally, programming assignments have served as both practice and evidence that students can design, write, test, and debug code. But when AI can generate a working solution, a finished program may tell us little about what a student understands or can do.

John will discuss his efforts to refocus homework on practice while assessing foundational skills more directly through paper-based coding, explanation quizzes, and debugging reports. He’ll also share what he has observed since making these changes to his course.

Finally, we’ll consider what his experience suggests for computer science education, and for other disciplines where AI can now produce work that once served as evidence of learning.

Location

This session will be held in-person and virtually via Zoom. Registration is not required but helps us prepare.

Join via this Link!

Engelmann Hall, RM B73
2033 E Hartford Ave,
Milwaukee, WI 53211

Facilitators