Join our daily and weekly newsletters for the latest updates and the exclusive content on AI coverage. Learn more
If you have been on the internet – or, at least, on the social network X – in the last day, you have probably encountered colorful and smooth images of style of famous photographs made in the style of the Japanese studio, Ghibli studio (the one that made Princess Mononoke,, The boy and the craneAnd My neighbor TotoroAmong many other classic animated films).
In fact, some users complain because their flows seem to be filled with almost exclusively these types of images.
Whether Current President TrumpThe emblematic image of the “Tank Man” during the 1989 Square Tiananmen demonstrations, Osama bin LadenJeffrey Epstein, or even other moments of pop culture and characters like Sam Rockwell emblematic camera The white lotus And many popular memes of the pastPeople made and shared these images to a quick clip.
Powered by the native image generation of the new GPT-4O model
Much of this is thanks to the new OpenAi update on the GPT-4O model behind Chatgpt for pro, more, and the team subscription levels, which dates back to the “generation of native images”.
While Chatgpt previously allowed users to create images from text prompts, he did it by transporting them to another, a separate Openai model, Dall-E 3.
But Openai’s GPT-4O Model is so named with an “o” because it is “omni” model-the company trained it not only on lyrics and code, but also on imagery and preumably, video and audio as well, allowing it to be able to understand all these forms of media and their Differences, Conceive of ideas across them (an “apple” is not just a word, but also something that can be drawn as a red or yllow or green fruit), and produce with precision media media without Connect to all external models.
Consequently, as the recent update of Rival Google AI Studio to include an experimental image creation model Gemini 2.0 Flash, the new OPENAI GPT-4O can also accept image downloads from any pre-existing image in your camera roller or that you have signed or saved on the web.
How to use chatgpt to create Ghibli-Studio style images (and change or transfer any image in any style!)
First of all, access to chat.com or chatgpt.com and make sure that you are connected with your chatgpt plus, pro or team account and that the AI model selector (located in the left corner of the session window) displays “GPT-4O” as a chosen model (you can click on it to deposit and select the appropriate model between the available options).
Once you’ve done this, you can download an image on Chatgpt using the “+” button in the lower left corner of the prompt entry text area, you can now ask the new GPT-4O model with image creation to make your image preexisting in a new style.
If you wish, you can try it by downloading a photo of yourself and friends and typing “Make all these people in the style of a Ghibli studio animation”. And after a few seconds, this will do it with fairly convincing and fun results. It even supports the connection of several images and combining them in a single room.
Chatgpt use of the free delayed level
OPENAI initially declared that it would also allow this feature for free (non-paying users of chatgpt), but unfortunately for them, the co-founder and CEO SAM Altman published today that the functionality will be delayed due to the overwhelming demand for existing payment subscribers to Chatgpt Plus, Pro and Team Third Party.
Like he Written on x::
“”Chatgpt images are much more popular than what we expected (and we had fairly high expectations).
Unfortunately, the deployment of our free level will be delayed for a while.“”
Meanwhile, those who have access will probably continue to browse the image changes in this recognizable or new styles.
Of course, not everyone is a fan of the work of Openai here. Actually, Studio creator Ghibli Hayao Miyazaki himself appeared in a documentary go back up 2016 – and one of the most memorable moments of this one still referenced to date is that it reacts with an overwhelming disgust and repulsion to an early example of animation and physics powered by AI, you guessed it, an Openai model.
As with many AI generative products and services, OPENAI training data for this new image generation capacity remain under WRAPS, but is largely speculated to contain material protected by copyright – and although Imitation of a style is generally not considered as a copyright violation In the United States, it rubs certain fans of Original animation in the wrong direction.
For the moment, these brands and companies that seek to play with this style should do so with caution and after serious consideration, given the possible negative return among certain users. But for those who are unwavered pro-Aui tools or with more indulgent and fun bases, it is clear that Optai has another blow on the hands.
