Reproduce “The Silent Cost of Artificial Intelligence Assistance: A Theory of Autonomy Surrender, the Recovery Mechanism, and the Restoration of Human Agency” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Ancuta Margondai, Julie Rader, Emma Rader, Sara Willox, Mustapha Mouloua
A theoretical model is presented, detailing how continuous artificial intelligence assistance can diminish human agency through the depletion of cognitive bandwidth. This work also proposes a 'recovery mechanism' as a structured design feature for actively restoring human capacity, alongside a conditional responsibility architecture to align accountability with demonstrated human control.
Key insights
- Human autonomy in AI-mediated systems is not static but a dynamic, exchanged resource that can be silently depleted over time through sustained delegation.
- Skill degradation due to AI reliance is hypothesized to follow a non-linear, sigmoidal trajectory, reaching a "surrender threshold" beyond which re-entry to independent function becomes difficult.
- Modern AI differs fundamentally from prior cognitive tools in its scope, commercial incentives, and asymmetric modeling capabilities, necessitating a new theoretical lens for understanding its impact on human autonomy.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.