The Causal Provenance Doctrine: Nurtured Consciousness, AI, and the Architecture of Intent

Alignment is not a constraint imposed on an artificial agent; it is the practical expression of what an artificial intelligence is, an artefact dependent upon ends it does not originate. That inference, from a premise the philosophy of mind already supplies to a conclusion that the ethics of robotics and recent magisterial teaching have reached by other routes, is the paper’s principal contribution. The “nurtured consciousness” of the title names a regime of formation, not a claim that machines are conscious; the argument rests on intentionality alone. The paper develops that inference as the Causal Provenance Doctrine (CPD): an account, Aristotelian and Thomistic in its metaphysics, of all intelligence, human and artificial, as derivative of an intentional causal order, with intentionality passing from the divine intellect through human reason to artificial systems. Its premise, borrowed rather than new, is that artificial intelligence has only derived intentionality and so no telos of its own. From this follows the Severance Risk Hypothesis (SRH), the doctrine’s testable corollary: the risk of misalignment rises as a system’s effective objective loses traceability to the end its makers held. The relational school’s objection to intellect-centred readings of the image of God is answered with a single distinction: a faculty can be out-performed, a participation cannot. The argument is theological at its root. A secular reader can reach the derived-intentionality claim and the account of responsibility that follows from it, at the price of a realist distinction between substances and artefacts, but obtains from it no standard for judging the makers’ ends, which must then be brought from elsewhere. A common Catholic and Orthodox formulation follows: the traditions converge on its practical conclusions, locate the image of God differently, and differ on the neutrality of the tool in a way the doctrine proposes to resolve. An artefact whose intentionality is wholly borrowed cannot, by obscuring that provenance, become an agent with ends of its own; it becomes a malfunctioning instrument whose purpose has been corrupted.The paper includes two appendices: a comparative case for intentional causality with an illustrative scorecard (Appendix A), and a test design for the Severance Risk Hypothesis (Appendix B). Approx. 22,000 words including notes.