OpenAI’s Latest AI Image Generator Sparks Studio Ghibli Style Memes
OpenAI recently launched a new AI image generator, and within just one day, social media platforms are overflowing with AI-generated memes mimicking the iconic style of Studio Ghibli, the beloved Japanese animation studio known for classics like “My Neighbor Totoro” and “Spirited Away.”
In the past 24 hours, netizens have shared AI-generated images depicting Studio Ghibli interpretations of various figures such as Elon Musk, characters from “The Lord of the Rings,” and even President Donald Trump. OpenAI CEO Sam Altman joined the trend by adopting a Studio Ghibli-style image as his new profile picture, presumably created using the GPT-4o’s in-built image generator. Users are leveraging ChatGPT to transform existing images into Studio Ghibli-esque artworks.
OpenAI’s move follows Google’s introduction of a similar AI image functionality in its Gemini Flash model. Earlier in March, this feature gained popularity when individuals utilized it to erase watermarks from images.
The introduction of OpenAI’s and Google’s new tools has made it simpler to replicate copyrighted styles by merely inputting a text prompt. However, concerns have arisen regarding the legality of training AI models on copyrighted content. Intellectual property attorney Evan Brown from Neal & McDevitt notes that while style isn’t explicitly protected by copyright, the method used by companies like OpenAI to achieve resemblance to Studio Ghibli’s works could potentially involve training on millions of frames from the studio’s films. The legality of this practice is currently being deliberated in courts.
Several lawsuits have been filed against OpenAI, including claims from The New York Times and other publishers alleging unauthorized use of copyrighted material in training their AI models. Similar allegations have been made against Meta and AI image-generation startup Midjourney.
In response to these accusations, an OpenAI spokesperson clarified that while ChatGPT doesn’t replicate individual artists’ styles, it can emulate broader studio aesthetics, such as those of Studio Ghibli. However, this distinction raises questions about the legal implications of replicating copyrighted styles through AI technology.
Users have explored GPT-4o’s image-generation feature by recreating styles from various studios and artists. Notably, someone recreated a portrait of Marc Andreessen in the style of Dr. Seuss, while a couple reimagined their wedding photos in the Pixar animation style.
In a comparative analysis of popular AI image generators, OpenAI’s tool emerged as the most adept at replicating Studio Ghibli’s distinct style.
The surge in demand for OpenAI’s new image tool prompted the company to postpone its release to free-tier users due to overwhelming interest. While these advancements in AI capabilities are driving increased usage, the legality of replicating copyrighted styles through AI models remains a subject of legal scrutiny. As we await court rulings on this matter, the future of AI-generated content creation hangs in the balance.