The Category Error of ‘Artificial Intelligence’: A Thomistic Analysis of Conscience, Prudence and Algorithmic Governance

Counts differ because each database indexes a different set of publications. We treat OpenAlex as the canonical count; Google Scholar is not shown (no API, and crawling it violates its ToS).

This article argues that Thomas Aquinas's moral theology provides distinctive diagnostic tools for understanding and addressing contemporary failures in artificial intelligence governance. Analysing the conceptual architecture of conscience ( synderesis and conscientia ) and prudence ( prudentia ), the article demonstrates that algorithmic systems cannot possess genuine moral knowledge or exercise authentic practical wisdom, revealing the designation ‘artificial intelligence’, when predicated univocally rather than analogically, as a consequential category error. Three fundamental divergences between Thomistic and dominant contemporary approaches – concerning human dignity, moral authority and the diagnosis of moral failure – expose the inadequacy of frameworks grounding human worth in functional capacities vulnerable to technological replication. Drawing on natural law principles, the article develops two operational instruments: the Prudential Reasoning Assessment Matrix, which evaluates whether systems support or substitute for human moral reasoning, and the Moral Ignorance Classification Framework, which assigns moral responsibility for algorithmic failures. Application of these tools to the Correctional Offender Management Profiling for Alternative Sanctions recidivism algorithm and DoNotPay legal service reveals recurring structural analogues of conscientia erronea and patterns of ignorantia affectata . The analysis concludes that preserving the distinction between mechanical processing and moral reasoning is essential to human dignity in an age of algorithmic governance.