Answers are grounded in this paper and cite their sources.
Generative AI can improve immediate task performance without strengthening the user's ability to perform or assess the task independently. This preprint develops an integrative framework of intellectual sovereignty for learning and knowledge work, examining who sets task boundaries and evidentiary standards, how cognitive delegation is limited or withdrawn, and how affected learners and workers can challenge decisions. It distinguishes individual evaluative capacity (I-Cap), institutional conditions (E-Cond), time-indexed actions and resolved outcomes (Action_t1, Answer_t1, Outcome_t2), blinded justification quality (C4-Qual), and unaided retention and transfer (Out-Ret), translating these distinctions into a Three-Track Assessment System (35% / 35% / 30%), an eight-stage workflow governance model, and five falsifiable hypotheses (H1–H5) linked to four proposed prospective studies (Studies 1–4). The framework and measures have not been empirically validated, and none of the proposed studies has been preregistered. This manuscript serves as the academic theoretical foundation accompanying the author's forthcoming Chinese monograph Intellectual Sovereignty: Brain Functional Reconfiguration in the Age of AI (《智力主权:AI时代的大脑功能重置》).Chinese translation: https://doi.org/10.5281/zenodo.24288230.
No takes yet. Share an insight, caveat, or question.
TENGJIAO LIU (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: