The Profile You Didn't Write: Digital Identity, Ocular Biometrics and Semantic Authority in Predictive Algorithmic Systems

This paper examines the architecture of contemporary algorithmic surveillance from the perspective of the individual agent operating in high-visibility digital environments. It argues that an individual's digital identity is not solely the result of explicit declarations — it is fundamentally a construction inferred by predictive systems from behavioral traces that were not deliberately authorized.

Documented evidence shows that fitness platforms, professional networks, and health applications generate, aggregate, and commercialize individual behavioral signals with third parties, and that this data is used to build psychographic profiles with measurable professional, financial, and insurance consequences.

The paper introduces an emergent dimension underrepresented in the literature: ocular biometrics — iris recognition, scleral pattern recognition, and gaze tracking — as a new frontier of identity capture with unique characteristics of permanence and irrevocability. An iris cannot be changed. A scleral vein pattern cannot be erased. This transforms the nature of the problem.

The Dynamic Coherence Model (DCM) is proposed as a structural response to this information asymmetry: the capacity to deliberately model how artificial intelligence systems process an identity, and to design that identity so that the classificatory output is the one the agent decided — not the one the algorithm inferred by default.