“It’s super fun seeing people love images in chatgpt, but our GPUs are melting” tweeted Open AI CEO Sam Altman as Ghibili studio trend hit the internet after the release of GPT-4o. Chat GPT released it’s 4o image generation tools on 25th March 2025. While it was portrayed as the AI model that delivers “photorealistic outputs”, the netizens unexpectedly loved a stranger version of its capabilities. Suddenly, the users were Ghiblizing their pictures and posting online. Elon Musk tweeted a Ghibili imagination of him holding the DOGE mascot in what reminds users of Lion King pose. Ghibili version of iconic Bollywood movies went viral. Cricket fans converted cricketing images while cricket great Sachin Tendulkar posted his world cup victory lap in Ghibli. Politicians too weren’t far behind as Maharashtra Chief Minister posted an image of PM Narendra Modi in Ghibili studio form. Sam Altman himself changed his profile pic with a Ghibli version. While the amusement with image generation refuses to die, AI has on a serious note solidified it’s image as a threat to artists and content creators worldwide.
Studio Ghibli is a renowned Japanese animation studio, founded in 1985 by Hayao Miyazaki, Isao Takahata, and Toshio Suzuki, known for its classic anime. While the ghibli style is beloved art work for decades now, the adoption of Ghibili by AI Chatbots raises important questions concerning copyright violations. In essence, the 4o image generator isn’t copying the artists because they are merely replicating the style and creating new images. In practice, the easy creation of impressive lookalikes photos in Ghibili style spells doom for artists worldwide. Unsurprisingly, the creator of Japanese animation Studio Ghibli has long criticized AI on several occasions. An old video of Studio Ghibli co-founder Hayao Miyazaki resurfaced where he called AI art “an insult to life itself”. As the Ghibili tsunami captured the imagination of internet, extraordinary artists and creators are staring at challenge to their livelihood and business models itself. While art should be a result of human experience, emotion and struggle, it is now a byproduct of a Graphic processing unit (GPU) in the AI world.
Image generating AI models
Ghibili style films are mostly hand drawn using rich watercolor and acrylic paints. Compared to traditional methods where every frame is drawn and colored by hand and computer animation is sparingly used, AI image generators don’t need colors or human skills. There are several AI image generators that can emulate the hand of the artists instantly. At the heart of it is a technique used in generative AI called diffusion. It begins with a dataset of existing images. Distortions are added to the dataset so that AI model is trained to separate the noise from the real image. Once the AI model is trained, it can generate entirely new images by identifying the noise and bringing an image to it’s improved version. The result is an entirely new image that corresponds to the trained dataset without copying it outright. AI image generators allow users to create image from text descriptions/prompts or by changing existing images/uploads. Whether it’s text to image or image to image, AI relies on pattern detections and iterations to modify the images. Approximately 34 million new images are generated daily. With the competition among AI image generators heating up, the number is set to skyrocket. The subtlety and creative process of an artists hand is accurately copied by AI as the tsunami of Ghibili studio images in the internet suggests. As per statista, 39 percent of American marketers use AI to create social media images. In the future, full-fledged animation movies maybe created with little to no assistance from a human artists. In the AI world, it’s tech savviness and astute prompts that determine the quality of art and not perseverance and practice of an artist.
Conclusion
The extraordinary reception of Chat GPT’s AI image generators solidifies the reputation of AI as an all-encompassing technology that will disrupt almost every sector. The content creator sector may get hit the hardest while jobs like coders, teachers and mathematicians are all under threat. The GPU’s of Open AI maybe melting but the potential of AI is solidifying due to the mass appeal. Apart from Open AI, there are image generators of MidJourney, Canva, Google Gemini and Ideogram that is servicing this demand for instant art. The rise of AI image generators brings in fresh set of challenges around copyrights and issues with enforceability. While The New York Times has filed a lawsuit against Open AI and Microsoft over use of their copyrighted news articles for training AI without permission, it’s ironical that NYT headline hailed the Ghibli style animation as “almost too good.” Everyone loves a good AI model that serves output instantly but hate it when AI models train themselves by feeding on copyrighted materials.
