ChatGPT-4o Images are UNREAL...

Join My Newsletter for Regular AI Updates 👇🏼

My Links 🔗

Media/Sponsorship Inquiries ✅

Links:

Chapters:
00:00 - Studio Ghibli style Lord of the Rings trailer sequence begins.
01:51 - How LoTR Ghibli was created.
03:56 - Studio Ghibli style Dune trailer sequence begins.
04:59 - Hubspot Sponsorship
06:01 - Realistic photo of Albert Einstein lifting weights.
06:17 - "How to Live in New York" infographic.
06:45 - Realistic photo of a man holding the "How to Live in New York" infographic.
06:56 - "How to Take Your Cat on a Leash Walk" infographic.
07:08 - Realistic photo of a cat looking at the "Cat Leash Walk" infographic.
07:20 - Vintage-style world map.
07:42 - Realistic photo of penguins looking at the vintage world map.
07:54 - Realistic photo of a sloth working at the DMV.
08:02 - Mike Tyson with a pigeon (Studio Ghibli / Anime style).
08:13 - Mike Tyson nude weigh-in (Studio Ghibli / Anime style).
08:44 - Sam Altman's Studio Ghibli style Twitter profile picture.
08:51 - Speaker's Studio Ghibli style Twitter profile picture (Rick Rubin).
09:37 - Generated Wikipedia page for "Cat".
10:08 - All-In Podcast hosts in Studio Ghibli style.
10:18 - Sloth anatomy diagrams (Tobi Lutke example).
10:47 - Ridge Wallet cartoon advertisement.
11:09 - Mr. Beast thumbnail with shark asset extraction.
11:26 - Sam Altman face merged into a meme painting.
11:46 - Mad Men style print advertisement for Obvi products.
12:15 - Skeuomorphic iOS UI examples (To-Do list first).
13:53 - Illustrated travel map (Philippines first).
14:59 - Comic book panels explaining T-cells (Guardians Within Birth).
15:52 - Realistic photograph generated from Rick and Morty cartoon image.

[VidSkipperAI]

Summary (with chapter timestamps): ChatGPT-4o's new native image functionality is enabling users to generate insane visuals, such as Studio Ghibli-style movie trailers and realistic photos, empowering a new generation of artists and transforming various media types.
0:00 🎬 Studio Ghibli Style Trailer
• 🤯 ChatGPT-4o's new native image functionality enables creation of insane visuals, like Studio Ghibli versions of movie trailers.
• 🎨 Visual artists can leverage this technology to empower a new generation of creators, making previously impossible things possible in a short time.
• ⚙️ The process involves screenshotting shots, converting them using OpenAI's native image generation (Sora or ChatGPT), animating in Cling AI and Luma Labs AI, and re-editing in sync with the trailer.

2:17 🎨 Creating with AI Image Generation
• ✍️ Prompts used in Sora include detailed descriptions to recreate scenes in the style of Studio Ghibli, ensuring similar composition, colors, and vibe.
• 🔄 The creator used half Clling and half Luma, switching when one didn't work, and previously went viral for converting a Studio Ghibli movie into live-action.
• 🖼️ Examples include a Dune trailer made in 28 minutes from idea to export using Sora and Cling AI, showcasing the potential of these tools.

6:01 🖼️ Versatility of ChatGPT's Image Model
• 🖼️ ChatGPT's new image model can create realistic photos and infographics with few mistakes, such as a lifelike photo of Albert Einstein in the gym.
• 🗺️ Examples include a detailed map of the world and a 'how-to' infographic for taking a cat on a leash walk, demonstrating the model's versatility.
• 👨‍🎨 Celebrities like Mike Tyson are using the model to create Studio Ghibli-style versions of their photos, highlighting the trend's popularity.

9:25 ✨ The Rise of AI-Generated Art
• ✨ The positive reception of Studio Ghibli-style images has driven the viral use of OpenAI's model, avoiding potential deep fake controversies.
• 🖼️ The model allows users to create images by describing what they want in natural language, lowering the barrier to entry for non-experts.
• 🔑 Data and taste are the most important factors, enabling users to generate images such as Wikipedia pages for cats and plausible anatomy diagrams.

11:58 🚀 Potential and Applications
• 🚀 The potential to create any type of media or asset for business, personal use, or memes is vast, with examples including extracting assets from thumbnails.
• ✨ The model can combine people into memes, create Madmen-style print ads, and generate images in the style of the original iOS.
• 🧪 The model understands text, image, and voice, allowing it to express itself in image, such as writing the snake game code in an image.

14:13 🧪 Biomedical and Cartoon Transformations
• 🧬 Biomedical scientists are using the model to create comic series for children, explaining the life of immune cells with beautiful imagery.
• 📸 The model can transform Rick and Morty cartoons into realistic photographs, though with some uncanny valley effect.
• 📏 Limitations exist in terms of dimensions, as the model struggles to create wider images for Twitter headers.


** Generated using ✨ VidSkipper AI Chrome Plugin

[Justin_Arut]

Ghost towns won't just be sets in Hollywood. Hollywood is about to BE a ghost town.

[jackied]

As someone that's done free-lance graphic design work, I think it's probably over. Clients always have something in their head they want and ask for a bunch of reworks. Now they can just ask for those reworks, and not have to email and wait for it to come back. To then not be happy and want something else. This just cuts all of that out and is cheaper.

[Guustaaf]

Appreciate the subtle, understated facial expressions in the thumbnail. A nice contrast with all these over the top thumbnails you see on other channels.

[BigK-]

The amount of toxicity I’ve seen thrown at people having fun is disgusting, I’m glad you mentioned it

[howephillip]

Animation aside, those of us using tools like Stable Diffusion locally have been doing this for quite some time now through image-to-image and inpainting. The only reason it’s blowing up now is that it’s become accessible to the masses and incredibly easy to use.

I work in 2D animation, and I also want to point out that this definitely doesn’t replace graphic designers or illustrators — instead, it empowers workload scaling. Personally, I may not be the most creative from scratch, but I can easily recreate or refine something in the Adobe suite. For freelancers, this means taking output to the next level.

For illustrators who specialize in creating from-scratch content, there may be slightly fewer opportunities in that area, but they can compensate with expanded productivity. They’ll likely face stronger competition against top-tier talent, but competition drives better results and ultimately makes for a stronger product.

[flow.philosophy]

You made a good point here. A lot of artists get mad at this kind of stuff. To me, those artists are lacking in creativity. As a coder, I'm not worried that I won't be able to code any more. I'm excited about being able to code in one day what used to take 1 month

[jason_v]

7:25 Glad he caught that. When he was correcting just the spelling, I was like, "Uhh...." Lol

[saisrikaranpulluri]

Before it use hallucinate a lot whenever I asked to replace the face or person in this image from other image, but now it is making very few mistakes. That was a lot of thought that was put into. Expecting it to be much better in the future even without any fewer mistakes. Cool video, watching all these creations is fun and entertaining.

[jeffcardA]

i've been taking a shot whenever i see or hear the word ghibli over the past couple of days. i'm about to die from alcohol poisoning.

[kylekoterba]

Matt, I've been saying this for a while but the age of experts is about to end. When there isn't a reason to "learn" the basics the context of expert knowledge is lost forever. This will happen for every discipline that AI perfects above novice professional's ability.

[echofloripa]

To get it wider, ask it to add a thick border (or margin) at the top and at bottom.

[vectoralphaSec]

Jobs/ Careers are going to drastically change by the end of this year.

[wrOngplant]

15:53 An awesome Rick and Morty! I'd watch it :)

[jaredgreen]

Anyone else notice how it seems to go out of its way to show an entire face of every character? Even when the person is facing away

[caine]

your editor did u dirty with that thumbnail 😂

[josephcurrington]

Dude sounded exactly like donut when he said gorgeous!!

[SahilP]

13:25 I hate how Apple transitioned away from skeuomorphic. I would throw insane amount of money if they bring it back, and probably in my opinion they have lost countless customers because a lot of people preferred this design. I know it's hard to code for and maintain but man is it pleasing and intuitive!

[alexpetrov]

The Matrix is cooking. I can already smell it.

[BlissfulForever]

💫💕It makes me SO HAPPY that the world loves Studio Ghibli so much 💫💕