What does it mean to “look like yourself” online? I did not want to create a perfect digital image of my face when making an AI selfie. I wanted to create an image that reflects how identity is shaped and distorted online. I did that by employing DALL·E and text prompts as opposed to a photograph. This option was important to me because, as Gibson (2008) points out when he is discussing the concept proposed by McLuhan, that the medium is the message, the format influences the creation of meaning.My self-representation is not a direct record of my appearance.
Figure 1 DALL•E Software Which I Used For Selfie Generation
Journey of Creation
I used trial and error as my process of creativity. My first search criteria were the following: realistic, soft light, and digital overlays. Initially, the results looked like generic stock portraits.Then I refined too many physical qualities, and I concentrated on the abstract terms of fragmentation, identity, digital mediation, and elements of interfaces. The impact of change like this was even greater. The image no longer tries to imitate my appearance. Instead, it begins to express what it feels like to exist within digital spaces.
Figure 2 Different Prompt Variations I Experimented With
Self-Representation and Identity
This made me realize that a selfie is not a neutral or accurate reflection of the self. Gorichanaz (2019) says that a self-portrait is a representation of the self, as well as its construction. The final image reflects my online identity, but it does not represent who I am in real life. Rather, it presents a mediated form of selfhood. It reflects a form of identity shaped by algorithmic aesthetics, visual codes, and the demands of digital culture.
Figure 3 My Final AI-Selfie
This image is important because it shows how my identity appears fragmented and shaped by digital systems. This also made me think about identity as performance and how it is perceived by audiences. We have surfaces on which the person is perceived to be online based on profile picture, posts, and visual style. Depending on the expectations of digital self-presentation, my AI selfie would be perceived as futuristic, distant, or even unstable by a viewer. The digital platforms do not passively consider identity on the margins as suggested by Guraya et al. (2025). They significantly shape identity formation, its presentation and perception, in other words, the self is being re-digitized. This idea is reflected in my image. As opposed to identifying identity as definite/ full-fledged, the image establishes it as a reconstruction that is continuously recreated, grounded on platform visibility and interface logic.
Ethical and Cultural Aspects
The AI-generated selfie also raises important ethical and cultural issues. The AI system technology does not come out of experience. They are developed through the support of training data, which presupposes that they can recreate prevailing norms and subliminal bias. Hofmann (2025) points out that bias in AI can be done in the forms of input bias, system bias, and application bias. It made me reflect on why my early results were drawn toward more conventional standards of beauty and familiar digital aesthetics. I also attempted to escape stereotypes, yet I was automatically driven to the visual default that could be culturally interpreted by the system. This raises important questions about representation, embedded values, and user control in AI systems.
Transliteracy Reflection
When I changed my academic analysis into a blog post, it altered my way of communicating my message. With this format, I needed to be more direct, less wordy, and clear. That is when the role of transliteracy obtains its meaning, not only the transfer of the ideas through the media, but also their re-education to a new audience. Blog form promotes reflection, readability, and multimodality. In support of this post, I presented the final AI selfie, the screenshots of my versions of prompts. This made the process more visible and supported my argument, that my AI selfie does not only show who I am. This made me realize that my identity online is not something I simply express, but something shaped by the systems I use.
References
Gibson, T. (2008). Introduction: Marshall McLuhan’s “medium is the message”: Information literacy in a multimedia age. MediaTropes, 1. ISSN 1913-6005
Gorichanaz, T. (2019). Self-portrait, selfie, self: Notes on identity and documentation in the digital age. Information, 10(10), 297. https://doi.org/10.3390/info10100297
Guraya, S. S., Ennab, F., & Guraya, S. Y. (2025). When the self “logs in”-a critical narrative review of digital identity in health professions education. Frontiers in Medicine, 12, 1715752. https://doi.org/10.3389/fmed.2025.1715752
Hofmann, B. (2025). Biases in AI: acknowledging and addressing the inevitable ethical issues. Frontiers in Digital Health, 7, 1614105. https://doi.org/10.3389/fdgth.2025.1614105
Hi! I found this post really meaningful because it challenges the usual idea that a selfie needs to look exactly like the person in order to represent them. What stood out to me was that the author did not only try to create a realistic copy of their physical appearance. Instead, they used the AI selfie to explore how identity feels when it exists inside digital spaces. I think this was a very interesting and creative choice, because sometimes the most accurate version of ourselves is not the one that looks the most physically similar to us, but the one that captures our inner condition or emotional experience.
ReplyDeleteThe part about fragmentation and digital mediation was especially powerful for me. It made me think about how online identity is often separated into pieces: profile pictures, captions, usernames, aesthetics, comments, and the way platforms organize all of this for others to see. In that sense, the AI selfie becomes more than an image. It becomes almost like a visual explanation of how the digital world changes the self.
I also liked the use of McLuhan’s idea that “the medium is the message,” because the tool used to create the image clearly affects the meaning of the final result. A camera selfie and an AI selfie are not the same thing. One captures a moment, while the other interprets a prompt.
At the same time, I wonder if this kind of fragmented digital identity gives people more freedom, or if it makes us even more dependent on how platforms want us to be seen. Overall, this post helped me understand that AI selfies do not simply show identity. They participate in creating it.