Studio Ghibli's enchanting aesthetic has taken the internet by storm, and now it's making waves in the digital world. The rise of AI-generated images in this beloved style using ChatGPT has surged in popularity, reportedly straining the technology to the point of overheating GPUs.
This week, ChatGPT received a major artistic upgrade. On 26 March, its creator, OpenAI, revealed the integration of AI image generation into its main chatbot. The feature has been a massive hit with users, with CEO Sam Altman even joking on Twitter that the company's 'GPUs are melting.'
ChatGPT Gets An Artistic Boost
It's unlikely that OpenAI's GPUs and servers are literally melting, but with increased demand, they can consume more power (and generate more heat) and make responses slower. Altman stated that to handle the surge in requests better, OpenAI is 'instituting temporary rate limits while we work on making it more efficient.'
CNET highlights that this kind of situation isn't unusual for creating images with AI; usually, moving up to subscription options can provide quicker and more reliable generation. It's similar to how purchasing a priority pass at Disney allows you to bypass lengthy queues for attractions.
When ChatGPT's image generator was first released on 25 March, there weren't any of those different access tiers. However, in the short period that this new image creation feature has been available, a clear trend has developed: Users frequently generate pictures that look like Studio Ghibli's artwork.
For those who don't know, Ghibli films such as 'My Neighbor Totoro' and 'Ponyo' possess a distinctive visual style. Staying with this popular trend, Altman even changed his X profile picture to an image of himself in the Ghibli style. However, problems continue between artists and AI developers.
>be me
— Sam Altman (@sama) March 26, 2025
>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"
OpenAI's decision to highlight this style is quite notable, particularly since, as 'Blood in the Machine' tech writer Brian Merchant mentions, Studio Ghibli's co-founder, Hayao Miyazaki, views AI tools as 'an insult to life itself.'
Image AI: ChatGPT vs. Dall-E
Previously, if people wanted to generate images with ChatGPT, they needed to use Dall-E. Dall-E is a distinct tool that users can find and access separately. According to CNET, it is a leading AI image generator, in part due to its interactive conversational method that makes designing and changing images straightforward.
The original Dall-E will still be accessible, and fortunately, the popular 'chat to edit' feature is also part of ChatGPT's new image capability. Whether you have a free or paid account, both Dall-E and ChatGPT are options for generating images. ChatGPT can be used without charge, with more comprehensive monthly plans available from £15.45 ($20).
Limitations Of ChatGPT's Image Generation
When creating visuals with ChatGPT, the limitations of your existing plan will be in effect. For instance, those on the free tier might face restrictions on using the 4o model for messaging, file uploads, and data analysis. Altman announced via Twitter that free users would soon have a daily cap of 3 image generations, but the specific date for this change wasn't shared.
Image creation within ChatGPT 4o will focus more on producing professional visuals, such as infographics and diagrams. OpenAI states that it has enhanced text rendering to facilitate this—a crucial improvement given that AI often makes errors and garbles words in images. Additionally, you can upload your own pictures and modify them using AI.
i like the studio ghibli memes as much as anybody, but you guys realize we're getting farmed, right? this is a mass onboarding event for paid subscriptions to chatGPT. everyone finding the ghibli style images so pleasing to look at is why it works. bravo, openAI, well played. https://t.co/tvMSAbtm2v pic.twitter.com/tdvlBrPOnb
— doomer (@uncledoomer) March 26, 2025
ChatGPT's image generation has some significant drawbacks. You might be unable to edit specific image regions accurately, which is important as AI can create mistakes like too many fingers. Face edits on your uploads could also disappear. Expect problems with cropping, data visualization, and multiple languages. The company states they are improving these and hope for fixes next week.
Generating Pictures With ChatGPT
Bringing your artistic ideas to life with AI is now simpler than ever. If you'd like to experiment with ChatGPT's new image generator, here are some ways to access this feature.
- Tap the three dots beside Search and Deep research on the main screen. Then, select Create Image.
- To make changes, just type what you want in the chat box and send it.
- When your image is ready, save your file by tapping the downward-pointing arrow in the top right corner.
The surge in Studio Ghibli-inspired AI art within ChatGPT highlights this new feature's immense popularity and potential strain. As OpenAI refines its image-generation capabilities and addresses current limitations, the future of AI-powered creativity looks both exciting and demanding.
Originally published on IBTimes UK
© Copyright IBTimes 2025. All rights reserved.
