# Mimesis, Poiesis, and Imagination: Exploring Text-to-Image Generation of Biblical Narratives

arXiv: https://arxiv.org/abs/2507.02973 · HTML: https://mlsanity.com/paper/2507.02973/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 4 of its 34 references that are in the map.
- First posted on arXiv 26 Jun 2025.

## Builds on (most cited first, 4 of 4)

- [Design Guidelines for Prompt Engineering Text-to-Image Generative Models](https://mlsanity.com/paper/2109.06977/) (2021)
- [Can Computers Create Art?](https://mlsanity.com/paper/1801.04486/) (2018)
- [Paint it Black: Generating paintings from text descriptions](https://mlsanity.com/paper/2302.08808/) (2023)
- [Seeing The Words: Evaluating AI-generated Biblical Art](https://mlsanity.com/paper/2504.16974/) (2025)

## Abstract (arXiv)

This study explores the intersection of artificial intelligence and the visualization of Biblical narratives by analyzing AI-generated images of Exodus 2:5-9 (Moses found in River Nile) using MidJourney. Drawing on the classical concepts of mimesis (imitation) and poiesis (creative generation), the authors investigate how text-to-image (T2I) models reproduce or reimagine sacred narratives. Through comparative visual analysis, including Google image results and classical paintings, the research evaluates the stylistic, theological, and cultural dimensions of AI-generated depictions. Findings show that while AI excels in producing aesthetically rich and imaginative visuals, it also reflects the biases and limitations of its training data. The study highlights AI's potential to augment human imagination but questions its capacity for genuine creativity, authorial intent, and theological depth. It concludes by suggesting that AI can serve as a creative partner in reinterpreting biblical texts, though its role in sacred art remains complex and contested.