GenAI Rewriting Interfaces Is Already Happening: The Ghibli GUI

Mar 28, 2025

GenAI Rewriting Interfaces Is Already Happening: The Ghibli GUI

There are many images in this entry, so if you are reading this text only in email you may not get as much out of it. If you don’t mind, I would suggest reading it online or allowing images in your email reader.

I recently wrote about how it is theoretically possible for AI to completely rewrite GUIs on the fly, how we could consume data visually in almost any way we want, and how we could potentially completely rewrite the interface for any application. Want Facebook to look like a Canon camera interface? No problem.

Could AI Could Just Rewrite Every App for You on the Fly?

Curtis Collicutt

Feb 24

Could AI Could Just Rewrite Every App for You on the Fly?

The future is hard to predict, and that is probably a good thing, although it does make stock picking a bit difficult. Also, strangely enough, a lot of things we didn’t predict seem so obvious in hindsight. How is that possible? Well, there’s no answer to that, because we’re not going to be able to predict the future in any real meaningful way, and yet …

Read full story

Of course, this is not feasible today, but it is not completely out of the realm of possibility for some time in the (near?) future. What’s more, as the imaging capabilities of models improve, more and more will become possible.

OpenAI 4o Image Update

The hubub: OpenAI has just released a new version of its GPT-4o model with new image generation capabilities, and it has taken over vaiblification” itself, but if we can set aside some of the ethical and artistic concerns, i.e. AI style appropriation and stylistic mimicry, it is interesting that we can adapt the interfaces at all. Just imaging that instead of copying Ghibli films, we did something else, something else more interesting.

Ghibli GUI would be great vibe for note-taking pic.twitter.com/ZaRMlJTMmb
— Wim Cools (@wcools) March 26, 2025

Rewriting Interfaces

Aside from making Ghiblified holiday pictures, there is a subset of people who are reimagining various graphical interfaces, Linkedin, ect, and while they use a similar Ghiblified style, if only for the next few days while the meme power dies down.

I think what is interesting is that it is possible at all.

So this morning I rolled out of bed and saw this post from Guillermo.
My first thought? “Damn, that looks sick!” And the best part? It’s actually super easy to recreate with @shadcn.
Greetings to the guy who claims every shadcn/ui project looks the same – catch this!
Watch… https://t.co/ll05ADFYmx pic.twitter.com/r6M9RjCKKg
— Matt Wierzbicki (@matsugfx) March 27, 2025

Example: Watercolors

Example: Salesforce GUI.

Example: The shadcn GUI library.

Example: An interaction on X FKA Twitter.

Example: A TODO list app.

Example: A trading platform.

Example: A trading platform.

Example: A SaaS platform.

Example: an IDE.

Ghiblification: Stylistic Mimicry

“I am utterly disgusted,” [Hayao Miyazaki] says. “If you really want to make creepy stuff, you can go ahead and do it. I would never wish to incorporate this technology into my work at all.” He goes on to say, “I strongly feel that this [CGI] is an insult to life itself…” [2] Miyazak speaking about an AI generated CGI creature several years ago, not this generation of AI.

OpenA’s new GPT-4o model has really leaped ahead in image generation capabilities, and is producing astounding, though as usual imperfect, images. A current meme is to use it to generate Ghiblified version of almost any image.

The point here isn’t that everything could be transformed into Ghibli-style images (which would be not only gross, but kind of useless), rather that the transformation is possible at all.

The latest version of OpenAI‘s image generation technology has resulted in a flood of users sharing images on social media that have been transformed in the style of Studio Ghibli, the legendary Japanese animation studio. [1]

Variety on Ghibli style:

On the one hand, the ability of OpenAI’s latest model to ‘Ghiblify’ images is a sad state of affairs. The reason people love his films is that they are an incredible labour of love and craftsmanship.

Below, just how much work goes into a Ghibli film.

Below, Miyaki talking about AI in 2016:

Since this utter garbage is trending, we should take a look at what Hayao Miyazaki, the founder of Studio Ghibli, said about machine created art. https://t.co/1TMPcFGIJE pic.twitter.com/IvaM9WZL3T
— Nuberodesign (@nuberodesign) March 26, 2025

Conclusion

Like many, if not most people, I love Hayao Miyazaki’s films, they are truly amazing, undeniable works of art and include incredible amounts of human labour and ingenuity. It is not great that we can just recreate, at least the imagery, an abstracted version of the style, and apply it to everything.

Ultimately, many of the images I show above, are really just simplistic themes overlaid on top of existing user interfaces, so not much new there, other than color choices and “hand drawn” lines, perhaps some nice blue sky and some flowers, but it also shows, I think, a desire for something new, a desire for something more human-centric in terms of interfaces, as opposed to the hard blocks of the spreadsheet influenced GUIs of today.

AS WELL… if we can put that aside the appropriation for a second and look at how people are using technology to redraw, to recreate, not copy, but in some ways re-imagine, there is at least something interesting in that we might be able to escape the boring computer interfaces of today. Now, we have to do more than just apply a sort of anime style to them, much more, but I think we can extrapolate what is happening today into something more interesting in the future.