The growing use of generative artificial intelligence (AI) in religious and legal discourse raises fundamental questions about the legitimacy of algorithmically mediated Islamic knowledge. Although existing scholarship has predominantly examined AI in terms of accuracy, functionality, and ethical risks, the ontological and epistemological conditions underlying Islamic legal authority remain insufficiently explored. This study examines whether AI-generated religious and legal discourse can satisfy these conditions and develops the concept of Artificial Truth to distinguish algorithmic plausibility from morally accountable legal knowledge. Employing a normative–philosophical approach, the study integrates Islamic metaphysics, comparative epistemology, uṣūl al-fiqh, and maqāṣid al-sharīʿah through ontological, epistemological, and axiological analyses. The findings demonstrate that AI systems cannot independently constitute authoritative Islamic legal knowledge because they lack the intentionality (niyyah), moral agency, and accountability (taklīf) traditionally associated with the knowing and legally responsible subject. Furthermore, their probabilistic outputs, limited source traceability, and susceptibility to hallucination create an epistemic condition in which linguistic coherence may be mistaken for legitimate knowledge. The study conceptualizes this condition as Artificial Truth and identifies its implications for algorithmic authority and epistemic dependence. It concludes that AI may support Islamic legal research and education under accountable human supervision but should not be regarded as an autonomous legal authority. This study contributes a maqāṣid-guided framework for AI governance emphasizing epistemic transparency, scholarly oversight, and human accountability.