Just a day after its launch, ChatGPT’s latest AI image generator has ignited social media, with users rapidly sharing artificial intelligence-generated memes inspired by Studio Ghibli. The celebrated Japanese animation studio, known for iconic films like “My Neighbor Totoro” and “Spirited Away,” has become a prominent style benchmark for this new technology.
Ghibli-Style Images Flood the Internet
Within 24 hours, the internet has been populated with AI-generated visuals depicting Studio Ghibli interpretations of figures such as Elon Musk, scenes from “The Lord of the Rings,” and portraits of former President Donald Trump. OpenAI CEO Sam Altman also appears to have adopted a Studio Ghibli-esque image, likely created with GPT-4o’s integrated image generator, as his new profile picture. It appears users are experimenting by uploading existing photos into ChatGPT and prompting the chatbot to reimagine them in diverse artistic styles.
This update from OpenAI follows Google’s recent introduction of a comparable AI image feature in its Gemini Flash model. Google’s tool similarly went viral in March when it was utilized to remove watermarks from images.
Copyright Concerns Rise with New AI Image Tools
The accessibility of OpenAI’s and Google’s newest tools simplifies the process of replicating copyrighted artistic styles using simple text prompts. These advancements in AI image generation have reignited fundamental questions surrounding ongoing legal battles against generative AI model developers: Does training these models on copyrighted material constitute a breach of copyright law?
Legal Perspectives on AI Style Replication
Evan Brown, an intellectual property lawyer at Neal & McDevitt, suggests that tools like GPT-4o’s image generator currently operate within a legally ambiguous zone. According to Brown, artistic style, in itself, is not explicitly protected by copyright. This implies that OpenAI may not be violating laws merely by producing images in the manner of Studio Ghibli films.
However, Brown raises the point that OpenAI could have achieved this stylistic similarity by training their model on vast quantities of frames from Ghibli’s productions. Even if this is the case, numerous courts are still deliberating whether employing copyrighted material to train AI models falls under fair use protections.
“I believe this brings up the same core question that has been present for a couple of years,” Brown stated in an interview. “What are the copyright infringement implications of broadly collecting data from the web and incorporating it into these databases?”
Ongoing Lawsuits Against AI Companies
The New York Times, along with several publishing houses, has active lawsuits against OpenAI. These claims assert that the company trained its AI models on copyrighted content without proper credit or compensation. Similar allegations have been made in lawsuits against other leading AI companies, including Meta and AI image-generation startup Midjourney.
In a statement provided to TechCrunch, an OpenAI spokesperson indicated that while ChatGPT is designed to avoid replicating “the style of individual living artists,” it is permitted to emulate “broader studio styles.” It is important to note that living artists, such as Studio Ghibli co-founder Hayao Miyazaki, are indeed recognized for pioneering their studio’s distinctive styles.
User Exploration of AI Style Generation
Users have demonstrably utilized GPT-4o’s image generation capability to reproduce styles from various studios and artists. Examples include a Marc Andreessen portrait in the style of Dr. Seuss, and a couple who recreated their wedding photographs in the style of Pixar animation.
Testing AI Image Generators for Style Accuracy
We conducted tests across several popular AI image generators—including those within Google’s Gemini, xAI’s Grok, and Playground.ai—evaluating their ability to accurately mimic the Studio Ghibli aesthetic. Our findings indicated that OpenAI’s new image generator produced the most faithful representation of the renowned animation studio’s style.
Demand and Legal Future of AI Image Generation
Currently, the new image features from OpenAI and Google represent a significant advancement in AI-generated content, apparently driving a substantial increase in user engagement. OpenAI postponed the rollout of its new image tool to free-tier users on Wednesday due to overwhelming demand. While user demand may be a primary focus for these companies presently, the legal ramifications and future of these technologies will ultimately be determined by court decisions.