Artificial Intelligence and Challenge Based Learning: Sustaining Human Capability in an Age of Intelligent Systems

The Challenge Institute just released a white paper on AI and Challenge Based Learning that includes revisiting Steve Jobʼs Bicycle of the Mind metaphor.

Generative artificial intelligence enables learners to produce sophisticated knowledge artifacts without necessarily engaging in the cognitive processes through which understanding develops. This paper explores this reality by bringing together research on cognitive offloading, extended cognition theory, and emerging causal evidence on AI-mediated learning. The central argument is that the critical educational distinction is not whether learners use AI, but whether AI amplifies or bypasses learner cognition. Amplification extends what learners can do while preserving their responsibility for interpretation, synthesis, judgment, action, and reflection. Bypass removes the cognitive events that learning is meant to cultivate. To clarify this distinction, the paper uses two diagnostic concepts: metacognitive displacement, in which learners lose opportunities to assess what they know and can do independently, and unearned fluency, in which AI-generated outputs create the appearance of understanding without the formation of durable capability. Challenge Based Learning (CBL) is positioned as a framework for sustaining cognitive engagement in AI-mediated environments, not because it was designed in response to AI, but because it is grounded in the conditions through which human capability develops: authentic challenge, sustained inquiry, purposeful action, and reflection. The educational imperative is to design learning environments in which intelligent systems extend human capability without displacing the cognitive work that makes learning possible.