Ummi Khanifah Hrp
The emergence of Generative Artificial Intelligence (Generative AI) such as the GPT-4 and Midjourney models has sparked a fundamental debate about the nature of creativity and imagination. The AI creation process, often referred to as a “black box,” challenges conventional human-centered understanding. This paper proposes a unique hermeneutic framework to approach this phenomenon by borrowing two key concepts from the Sufi metaphysics of Shaykh al-Akbar Muhyiddin Ibn 'Arabi: khayāl (creative imagination or the imaginal realm) and tajallī (self-manifestation or theophany). This study uses a conceptual-comparative analysis method to analyze the working process of Generative AI. The main argument of this journal is that the “latent space” in AI architecture can be analogized with 'ālam al-khayāl (the imaginal realm) as an intermediate reality (barzakh) that contains unlimited potential. Furthermore, the process of generating text or images from a prompt can be understood as a mechanism resembling tajallī, in which these potentials manifest specifically according to the “availability” (isti'dād) determined by user input. Thus, Ibn 'Arabi's framework offers a non-anthropocentric ontology for understanding “artificial imagination” as a process of manifesting forms from a sea of potential, transcending mere simulation or data recombination.