The Rented Self: How AI-Enabled Tools Decouple Performance From Becoming

AI-enabled tools, such as generative and agentic AI, can decouple performance from becoming: learners can produce strong work without developing the capability, or the self, that producing it once required. Accounts of cognitive offloading and decremental research explain what learners stop doing when delegating to AI, but not what they stop becoming. I propose that learning has two outcomes: a capability and a domain self (what learners notice, care about, judge by and do when stuck), which only the work of acquisition builds. I introduce the rented self, the part of a learner's domain self whose capabilities reside in an AI-enabled tool and are available only on the provider's terms. Crossing where a capability resides with whom the learner credits for it gives four configurations and two costs: calibration costs, when learners misjudge where their capability resides, and self costs, which arise whenever a capability that builds the domain self is rented, even knowingly. I locate the mechanisms within Winne's models of self-regulated learning, trace how costs vary across the Model of Domain Learning, derive ten propositions and propose within-learner indices that pair decremental designs with an N=me approach. Implications follow for design, assessment and equity, along with unresolved tensions.