Discusión: Algorithmic Transparency and Informational Self-Determination
The structural tension between automated decision-making and constitutional habeas data protections highlights a fundamental deficit in algorithmic accountability across the digital financial sector. In the Colombian regulatory landscape, balancing artificial intelligence applications with individual consumer safeguards requires substantive interpretability rather than superficial disclosures of data aggregation practices (Balancing AI Data and Consumer Rights: The Colombian Context, 2026). When lending entities deploy machine learning models trained on vast, multidimensional datasets, the complex statistical interactions create opaque underwriting decisions that obscure the precise logic behind an adverse rating (Hurley & Adebayo, 2018). Consequently, traditional statutory habeas data guarantees—which historically evolved to rectify explicit factual errors in centralized financial databases—fail to provide effective remedies against automated inferences and latent algorithmic profiling. This systemic opacity weakens informational self-determination, as borrowers cannot meaningfully exercise their rights to update, correct, or challenge evaluations generated by non-transparent scoring pipelines. As automated scoring systems increasingly incorporate non-traditional behavioural indicators, supervisory bodies must establish clear technical parameters to translate constitutional protections into enforceable computational standards. Reconciling Colombian legal mandates with automated credit scoring therefore necessitates institutional frameworks that enforce algorithmic auditability, counter structural asymmetries, and guarantee actionable explanations for adverse automated outcomes.