ARTIFICIAL INTELLIGENCE AND REPRESENTATION OF CULTURE: A DUAL ANALYSIS ON THE DEPICTION OF TURKISH CULTURE IN MIDJOURNEY VERSIONS
Öz
This study examines the representation of Turkish culture in Midjourney, one of the most widely used text-to-image AI applications worldwide. Using a standardized prompt protocol, it analyzes visual and textual outputs across versions from V1.0 to the current V8.0 through qualitative and quantitative methods. The minimal prompt "Turkish culture" was used to generate 12 images per version, yielding a dataset of 96 images. In the first stage, the images underwent interpretive qualitative analysis focusing on visual aesthetics, figurative representation, clothing, cultural symbols, objects, and spatial and architectural elements. In the second stage, the same images were converted into text using Midjourney's "describe" tool, and the resulting outputs were examined through quantitative word-frequency analysis to validate the visual findings. The findings show that Midjourney's visual-aesthetic capacity improved markedly across versions, moving from a painterly aesthetic and formal ambiguity toward compositionally more coherent, anatomically accurate, and photorealistic scenes. This progress, however, is not matched in the repertoire of cultural representation: although emphases shift between versions, outputs consistently converge, at the algorithmic scale, on national symbols and highly recognizable elements of historical heritage. Visual and textual findings reveal a stable pattern in which Turkish culture is represented largely through the flag, a red-and-white color code, the star-and-crescent form, traditional clothing, folk dances, bazaar scenes, and religious architecture. Despite the fluidity of culture and its multilayered, rich structure that also encompasses the present, this clustered pattern of representation is discussed in relation to the concept of the "frozen heritage effect."
Anahtar Kelimeler
Turkish culture, representation of culture, generative artificial intelligence, Midjourney
Kaynakça
- Almarwani, N., Aloufi, S., Alkhereyf, S., Alhassoun, M., Almutery, M., Alshalawi, N., & Al-Thubaity, A. (2025). KingdomGlimpses: Evaluating Saudi Cultural Representation Through Text-to-Image Models. IEEE Access, 13, 177822–177845. https://doi.org/10.1109/ACCESS.2025.3619432
- Bianchi, F., Kalluri, P., Durmus, E., Ladhak, F., Cheng, M., Nozza, D., Hashimoto, T., Jurafsky, D., Zou, J., & Caliskan, A. (2023). Easily Accessible Text-to-Image Generation Amplifies Demographic Stereotypes at Large Scale. 2023 ACM Conference on Fairness, Accountability, and Transparency, 1493–1504. https://doi.org/10.1145/3593013.3594095
- Darius. (2026, 01 24). Midjourney Statistics and User Demographics 2026. Retrieved 06 01, 2026, from Companies History: https://www.companieshistory.com/midjourney-statistics/
- Elsharif, W., Alzubaidi, M., & Agus, M. (2025). Cultural Bias in Text-to-Image Models: A Systematic Review of Bias Identification, Evaluation, and Mitigation Strategies. IEEE, 13, 122636 - 122659.
- Hall, S. (1997). Representations: Cultural Representations and Signifiying Practices. London - U.K.: Sage Publications.
- Hobsbawm, E., & Langer, T. (1983). The Invention of Tradition. Cambridge, U.K.: Cambridge University Press.
- Jha, A., Prabhakaran, V., Denton, R., Laszlo, S., Dave, S., Qadri, R., Reddy, C., & Dev, S. (2024). ViSAGe: A Global-Scale Analysis of Visual Stereotypes in Text-to-Image Generation. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), 12333–12347. https://doi.org/10.18653/v1/2024.acl-long.667
- Kannen, N., Ahmad, A., Andreetto, M., Prabhakaran, V., Prabhu, U., Dieng, A., Bhattacharyya, P., & Dave, S. (2024). Beyond Aesthetics: Cultural Competence in Text-to-Image Models. Advances in Neural Information Processing Systems 37, 13716–13747. https://doi.org/10.52202/079017-0439
- Kotliar, D. M. (2020). Data orientalism: on the algorithmic construction of the non-Western other. Theory and Society, 49(5–6), 919–939. https://doi.org/10.1007/s11186-020-09404-2
- Liu, Z., Janowicz, K., & Karimi, M. (2026). Assessing the Geographic Diversity of AI’s Platial Representations in Image Generation. (https://doi.org/10.48550/arXiv.2606.05188 - Arxiv preprint)
