ChatGPT’s new Ghibli-style image feature sparks social media trend and controversy

ChatGPT’s latest update has triggered a wave of Studio Ghibli-style image generation across social media, creating both fascination and controversy in the AI art community. The new native image generation capabilities in GPT-4o allow users to transform photos into anime-inspired artwork without connecting to external models, representing a significant evolution in multimodal AI that processes and creates across text, image, and other media formats.

The big picture: OpenAI’s GPT-4o model now generates images directly within ChatGPT for paid subscribers, sparking a trend of transforming famous photos into Studio Ghibli-style animations across social media platforms.

• The feature works natively within the model rather than routing requests to DALL-E 3 as previous versions did.
• Users have applied the style to everything from political figures like Trump to iconic historical images like the Tiananmen Square “Tank Man” and pop culture references.

How it works: ChatGPT Plus, Pro, and Team subscribers can transform any uploaded image into various artistic styles through a simple conversational prompt.

1. Log into ChatGPT with a paid subscription account
2. Ensure GPT-4o is selected as your model
3. Upload an image using the “+” button
4. Ask the AI to render your image in your desired style (like “make this in the style of Studio Ghibli”)

What they’re saying: Studio Ghibli creator Hayao Miyazaki has previously expressed strong opposition to AI-generated animation.

• Miyazaki appeared in a 2016 documentary where he reacted with “overwhelming disgust” to an early example of AI-powered animation by an OpenAI model.

Behind the numbers: The feature has proven unexpectedly popular, causing OpenAI to delay its planned rollout to free users.

• CEO Sam Altman noted on X: “images in chatgpt are wayyyy more popular than we expected (and we had pretty high expectations).”

The controversy: While imitating an artistic style isn’t generally considered copyright infringement in the US, questions remain about OpenAI’s training data.

• The model is believed to have been trained on copyrighted material, raising concerns among fans of the original Studio Ghibli animations.
• The ability to generate convincing character and scene styles mimicking famous studios has sparked debates about AI’s impact on artistic originality.