Work

Algorithmic Malpractice

Year: 2026
Type: preprint
Abstract: Algorithmic malpractice represents a contemporary form of clinical negligencestemming from errors or omissions in the use of artificial intelligence systems.Since algorithms lack the legal personality required to be held accountable forfailed diagnoses or treatments, this technology does not eliminate human liability;rather, it redistributes it among healthcare professionals, institutions, developers,and regulatory bodies. This scenario necessitates rigorous frameworks foroversight, traceability, and accountability.At the same time, algorithmic bias poses a critical risk by replicating historicalinequalities—linked to sociodemographic factors—present in training data.Consequently, discriminatory decisions are legitimized under a false veneer ofobjective neutrality, further compounded by the use of flawed proxy variables—such as prior healthcare spending—which mask structural barriers to accessinstead of reflecting actual clinical needs.These shortcomings are particularly grave in high-stakes areas, such asintensive care or the allocation of organs for transplantation, where a biasedprediction directly jeopardizes human life. Therefore, it is imperative toensure explainability, independent auditing, and continuous clinical review.Artificial intelligence must function strictly as a diagnostic support tool,operating under an ethical governance framework that safeguards humandignity and the right to health
Language: English
Field: Medicine
Open Access status: green
OpenAlex ID: W7225869412
Cites:
Cited by:
Related to: 10