AI-Assisted Visuals in Indonesian Muslim Digital Religious Communication

Generative artificial intelligence (AI) is entering religious communication not only through chatbots and generated text, but also through images and reels on social media. Yet research in this field has paid limited attention to what images contribute beyond their accompanying words. This article examines how AI-assisted visuals are incorporated into Indonesian Muslim Instagram accounts linked to Nahdlatul Ulama (NU) and what communicative work they perform. Informed by the Religious Social Shaping of Technology framework and visual-culture scholarship, the study combines qualitative digital observation with content and multimodal analysis of eleven posts from three accounts and sampled comments. The visuals were adopted selectively within established organisational, humour-oriented, and preacher-led modes of address. AI functioned through amplification: synthetic construction gave bodily and spatial form to themes of national belonging, moral criticism, santri conduct, and interreligious engagement, while enabling encounters that could not readily be photographed. These moral positions became legible through familiar signs, but authority remained with the organisation, collective voice, or public persona through which the images were framed and circulated. AI-assisted imagery is therefore best understood as situated visual mediation: it expands what religious communicators can picture without independently determining the meaning or authority of what is pictured.