ChatGPT maker OpenAI has introduced a new feature for its GPT-4o model this week: Native image generation. Announced on Tuesday, this update allows GPT-4o to create a wide array of images—ranging from infographics and comic strips to memes, street signs, and user interfaces—solely based on text prompts. The model also supports refining and editing generated images with follow-up instructions, opening up new creative possibilities for users.
Among the standout features catching the internet’s eye is the ability to transform everyday photos into the whimsical, hand-drawn aesthetic of Studio Ghibli, the renowned Japanese animation studio celebrated for its lush backgrounds, soft color palettes, and nostalgic storytelling. Known for films like Spirited Away and My Neighbor Totoro, Studio Ghibli has long captivated a global fanbase. With GPT-4o’s new capabilities, users can now upload any image and request a Ghibli-style makeover, fueling a viral trend online.
Internet goes all Ghibli Crazy
The feature, dubbed “Images in ChatGPT,” has sparked widespread excitement. Netizens have flooded social media with their Ghibli-fied creations, turning ordinary snapshots into dreamy scenes reminiscent of the studio’s imaginative worlds. The Ghibli-style trend has struck a particular chord with Studio Ghibli’s dedicated fans, who see GPT-4o as a tool to reimagine their world through the lens of the studio’s iconic artistry. From lush landscapes to charming character designs, the internet is abuzz with creations that blend modern technology with timeless animation magic.
ChatGPT maker OpenAI CEO Sam Altman joins the fun
Even OpenAI CEO Sam Altman couldn’t resist joining the fun. On Twitter, Altman swapped his profile picture for a Ghibli-style rendition of himself, sharing a humorous take on the craze: “>be me >grind for a decade trying to help make superintelligence to cure cancer or whatever >mostly no one cares for first 7.5 years, then for 2.5 years everyone hates you for everything >wake up one day to hundreds of messages: ‘look i made you into a twink ghibli style haha.’” His lighthearted post underscores the feature’s playful impact, even among the company’s top brass.
Unlike previous image generation efforts that leaned on external models like OpenAI’s DALL-E, GPT-4o’s native capabilities allow it to craft images using its own inherent knowledge. “Creating and customizing images is as simple as chatting using GPT-4o – just describe what you need, including any specifics like aspect ratio, exact colors using hex codes, or a transparent background,” OpenAI stated in its announcement. For those who prefer it, DALL-E remains available as a standalone option.