In March 2024, OpenAI released an upgraded image generation feature for GPT-4o. With enhanced quality and style, faster processing speed, and fewer limitations, the tool has become easier and more accessible for users. In particular, its ability to generate Ghibli-style images gained explosive popularity worldwide.

While many people are captivated by this technology, it has also sparked renewed debate-particularly around ethical concerns. Since there is currently no regulation requiring AI-generated works to disclose their training sources, proper attribution is often lacking. Moreover, as the quality of AI-generated content improves, it becomes increasingly difficult to distinguish it from human-made creations. As a result, more AI-generated content is spreading across the internet. Because it is still imperfect, this often leads to confusion and inconvenience for users.

Another concern is that if AI continues to train on unreliable AI-generated content in a repetitive cycle-known as AI autophagy-the overall credibility and accuracy of information may significantly decline.

Additionally, OpenAI's surge in popularity through Ghibli-style images has fueled debates over whether licenses were obtained for training data, and whether it is ethically acceptable to freely generate and distribute images in the Ghibli style. Since artistic styles are not clearly protected under current copyright laws, and reproducing a style without directly copying original works is generally not considered infringement, Studio Ghibli may find it difficult to assert its copyright claims in this context.

Despite these ongoing issues, many major tech companies continue to advocate for looser regulations in order to support AI development. Given that the emergence of ethical concerns alongside technological progress is inevitable, it is essential to establish clear ethical guidelines that align with the speed and direction of innovation.

하승연 강남포스트 학생기자  webmaster@ignnews.kr

<저작권자 © 강남포스트, 무단 전재 및 재배포 금지>