It’s clearly not on the level of never having heard of Disney, since I doubt you’re gonna find anyone on this forum who hasn’t heard of Disney while plenty of people clearly haven’t heard of this Anime studio.
Just because I was curious:
https://i.ibb.co/q3P5yjWz/Screenshot-20250327-064214-Samsung-Internet.jpg
You can see the bump on Studio Ghibli searches over the last day since this trend started.
You can also see that it’s completely and utterly dwarfed by the daily deluge of children waking up and searching “Disney” and then coming home from school and searching “Disney” even more.
Studio Ghibli might be a major anime studio, but it is many orders of magnitude off from being Disney.
OpenAI is definitely pooping the party:
I suppose. After all, they have won two of the six Best Animated Feature Oscars they have been nominated for, and been the only group ever awarded a Cannes Palme d’Or, but it is still possible for people (especially American ones) who pay extremely little attention to the film industry to have never heard of it. To their extreme loss.
This guy converts the LOTR trailer into Studio Ghibli style using AI. Astonishing stuff. We are pretty close to the time when an individual can create a feature-length animated film as a hobby project in a few months in their spare time.
I can’t read the story. Says I need a subscription. Can you summarize? It has been working fine for me since learning of it last night. Just tried it again, and it still seems to work for me, or at least it is producing some kind of cartoony picture that looks like a nice stylized version of the photo I fed. I can’t tell you if it’s actually Studio Ghibil or not, as I’m not a connosieur of that genre, but looks pretty good to me. I can tell you the Sanrio and Simpsons characters are spot-on.
By Wednesday night, the system started to block some users’ requests for images in the style of Ghibli and some — though not all — other artists.
OpenAI said in an explanation about GPT-4o’s latest image generation technology that it was taking a “conservative approach” to images using artists’ work in this model.
“We added a refusal which triggers when a user attempts to generate an image in the style of a living artist,” the company said.
An OpenAI spokesperson told Business Insider on Wednesday night that the company continued to prevent “generations in the style of individual living artists” but did permit “broader studio styles.”
“We’re always learning from real-world use and feedback, and we’ll keep refining our policies as we go,” the spokesperson added.
NVM not for this thread
Looks like I’m luckily not one of the users blocked.
Seeing lots of ChatGPT generated images on Facebook. For some reason they often have a distinctive yellowish tint to them that identifies their origin, even from prompts that obviously had nothing to do with Ghibli.
It’s approaching New Yorker cartoon levels of humor.
I started seeing impressively coherent action figures on Facebook that were obviously from the new ChatGPT (obvious for their next-level quality and for that same strange yellow tint).
One example:
I googled for mentions of ChatGPT action figures and got this page as an example:
I tried creating a couple of figures, but free users are still on the old image generator (and limited to three images per day).
FWIW, here’s what Bing comes up with with the same original prompt. They aren’t that different from what the old ChatGPT generator can do, the main difference being that ChatGPT can dream up novel descriptive text for the cards while Bing/DE3 mostly just reuses text from the prompt (badly). But neither, obviously, approaches the level of sophistication on the new engine.
ChatGPT is now allowing 4o image creation for free users. And it allows Ghibli. Very slow, though. It looks like the limit is five images per day.
You’ve seen many of the successes from ChatGTP’s new image generator, how about a failure?
To get a good image I had to “cheat” and flip the screenshot over.
And the finished project, GhibLeon:
Veo 3 is a big leap forward for AI video especially with the integration of sound, music and dialog.
Here’s an absolutely stunning set of Veo clips on Facebook
It did not surprise me those were Facebook ads for products you could order, but what did surprise me was how bad AI was at making your ‘action figure’ based on your Facebook profile. But then, I expect there to be finite personalization for this item.
I don’t know what you are referencing, but my posted image you linked to has absolutely nothing whatsoever to do with a “Facebook ad for items you can order”.
“The Ultimate Rock Collector” set you posted above. If you spend any time of Facebook you’d recognize it and know what I was talking about.
It’s all good.