How long will it be until...

I would not doubt that it is already possible... There is already a couple of AIs to generate images, generate video even in 4k, create music, write stories, write software, translate from one computer language to another computer language... I have even used one AI to write prompts for another AI... We live in interesting times...
 
Last edited:
We live in interesting times...
The beginning of.

I was on Youtube and notice a lot of AI generated stuff. I am wondering how long it will be before you can give AI a online book and have it make a full 4k or even 8k movie.
Maybe that’ll be ok for young people. I want real people in my movies, music and life!
 
The beginning of.


Maybe that’ll be ok for young people. I want real people in my movies, music and life!
I said take a book and have AI make it a movie...which means everything is AI generated...no actors involved. Instead of reading the book or listening to a audiobook...next level is a movie.

Now I am not young either. ;)

I remember seeing something awhile back that Actors were basically signing away their rights for the movie companies to be able to use their likeness after they die. When I saw this I figured it would be CGI, but now with AI...I think that is what will happen. Think of how much money they will save after a actor is dead, and they continue his likeness...it is all about the Benjamins.
 
I don't see much difference between CGI/SFX (Computer Generated Imagery/Special Effects) and Artificial Intelligence generated imagery... They are both at the manipulation/direction hand of a human... Just another level of abstraction... Turtles all the way down... :nerd:

BTY, I would not mind seeing the Jimi Hendrix Experience or his Band of Gypsys release a few new songs... :bigsmile:

A fun interview with Scott Ross... https://kevingoetz360.com/scott-ross-vfx-ai-hollywood-future/
 
Last edited:
I think once the next level of AI gets released, which I've read could happen in the next 5 years, the entire landscape is going to change. @NBPK402, someday we'll get to the point of what you're talking... but I'm trying to wrap my head around how much energy that will take to execute. We're already hitting the outer limits of what's possible with the energy we have.

Hopefully, future processing chips will be infinitely more efficient.

I think I'm less concerned with AI art and more concerned with AI-generated fakes of real people being proliferated on social media. That's scary.
 
I think once the next level of AI gets released, which I've read could happen in the next 5 years, the entire landscape is going to change. @NBPK402, someday we'll get to the point of what you're talking... but I'm trying to wrap my head around how much energy that will take to execute. We're already hitting the outer limits of what's possible with the energy we have.

Hopefully, future processing chips will be infinitely more efficient.

I think I'm less concerned with AI art and more concerned with AI-generated fakes of real people being proliferated on social media. That's scary.
With Super computers.. I do not know how much energy they will, use, but I am already scared of deep fake videos.

I watched a TV series or a movie awhile back that was very good, but also very scary as you could be framed for murder without ever being there.
 
I think once the next level of AI gets released, which I've read could happen in the next 5 years, the entire landscape is going to change. @NBPK402, someday we'll get to the point of what you're talking... but I'm trying to wrap my head around how much energy that will take to execute. We're already hitting the outer limits of what's possible with the energy we have.
I don't think anything new will happen... AI has been the Siren's call for the past 50 or so years... Large Language Models do take a lot of computing power to create... Places like Google, Meta, and X compute the entire internet into such a LLM... However to actually query a LLM does not take much computing power... At scale, with thousands to millions of folks making queries of these LLMs does again take a lot of computing power... However, one can create a personal LLM on a desk side PC with just a few Nvidia Graphic Cards with as little energy as running your Video Projector and a nice Stereo or AVR system...

Hopefully, future processing chips will be infinitely more efficient.
Moor's Law still pretty much applies...

I think I'm less concerned with AI art and more concerned with AI-generated fakes of real people being proliferated on social media.
Like any technology, there will be those that will use the tech for good or evil...

And... Modular reactor technology will become available soon to deal with all the power requirements... Back in the day Supercomputers were powered by Kato generators using diesel or natural gas...
 
Last edited:
I was on Youtube and notice a lot of AI generated stuff. I am wondering how long it will be before you can give AI a online book and have it make a full 4k or even 8k movie.
said take a book and have AI make it a movie...which means everything is AI generated...no actors involved. Instead of reading the book or listening to a audiobook...next level is a movie.
Impressive but wrong direction. AI can already give me the text summary of a video.
 
AI can already give me the text summary of a video.
Actually AI can transcribe everything said in a video...

And... Today I asked an AI to read The Cat In A Hat book and create the prompts that would allow another AI to create a video interpretation of the book... And it did... Maybe tomorrow I will feed those prompts into a AI that generates video... :nerd:
 
Some people are in complete denial of AI, downplaying its current capabilities and its infinitely possible future.

I get wanting to ignore it... maybe even being afraid of it. But turning away from it is a mistake. It's here and it's not going anywhere. So, your best bet is to figure out how to get along with it!
 
Some people are in complete denial of AI, downplaying its current capabilities and its infinitely possible future.

I get wanting to ignore it... maybe even being afraid of it. But turning away from it is a mistake. It's here and it's not going anywhere. So, your best bet is to figure out how to get along with it!
So far I’ve found LLMs to be a super useful tool and I will continue to exploit their capabilities… until they turn on us all. 🙂

IMG_1607.jpeg


PS: I still want real humans for music and movies. Creating, playing and acting.

PPS: I asked one of the artificial brains for its opinion on the original premise of the thread. It responded with:

🎯 My Take:
We’re heading into an era where:

  • Books could become movies automatically, customized to the viewer’s preferences (voice, style, even ending).
  • Classic actors could “act” posthumously, but only if rights allow.
  • Smaller creators can compete, bypassing massive studio budgets.
But…
The magic of cinema often lies in human performance. AI might handle 90% of technical tasks, but capturing the raw, lived-in emotional texture of human experience is still a challenge. And even if AI can do it someday — we may still prefer real people.

Much like vinyl vs digital or hand-drawn animation vs CGI, AI films may become a distinct category — convenient, personalized, and cheap — but not necessarily better.

🔚 Bottom Line:

  • Technologically, we’re very close to book-to-movie AI conversion for amateur use.
  • Commercial-quality productions with AI? Probably within 5–10 years, depending on model scaling, coherence, and regulation.
  • Emotionally and culturally? That’s a bigger question. But resistance to fully AI-generated content will persist — especially among those who value authenticity and human artistry.
———————————-

we may still prefer real people.”
Wait.. What? Did you notice that? Its started already. :oops:
JStewart
 
Have you heard the new AI band on Spotify? Velvet something?

It's crazy... it actually sounds good!
 
Yes Velvet Sundown... I have been associated with the film entertainment and VFX/SFX industry since the mid 1980s... This trend of using computers to help produce video and filmed entertainment will not stop... And the music industry has picked up on using AI in the production, mixing and mastering as well as promotional content... This train had left the station a long time ago... Now picking up speed and wondering about it running away... Also thinking about duel use for America's Security plays out when you talk about deep fakes...
 
Curious, what do you think of the Velvet Sundown music?
 
Not exactly what I really like... And its not bad for what it is... I have not heard all their stuff... Haven't yet found a HiRez album yet, just a few U-Tube videos...

As I have said before, I don't have any basic issues with AI generated stuff as there is always a human guiding hand involved... As Scott Ross seems to put it... Humans with talent will make good entertainment... Its not like one morning the AI wakes up and says to itself "Hey I think I will make a movie or write and record a song"... A human prompts it... And that human may or may not have the talent to pull it off...
 
I guess that was my reaction. Not bad for what it is. I'm actually more amazed at what's achievable with technology. It's crazy!
 
I was on Youtube and notice a lot of AI generated stuff. I am wondering how long it will be before you can give AI a online book and have it make a full 4k or even 8k movie.

Youtube AI generated
I must say, it does look impressive though :)
 
A perfect time to come out with the 3rd Tron movie too. I wonder if they will actually use AI in the movie too. I can't believe it took over 40 years to get to the 3rd movie in the series. Def gonna get it when it comes out.
 
@NBPK402 Had said "I don't know where they get the images for the actors and actresses, but they sure look good."

So lets ask an AI about that... :nerd:

AI image generators produce highly realistic human images by using powerful machine learning techniques trained on large datasets of real images and their descriptions. There are two main modern approaches: Generative Adversarial Networks (GANs) and diffusion models.
How it works:
The AI is exposed to millions of photos of real people during training, learning detailed patterns and features that make human faces look authentic.
When you provide a text prompt (like "a portrait of a young woman"), the AI analyzes the prompt, recognizes key visual attributes, and then generates an image using the patterns it learned rather than copying any individual photo.
GANs: In this method, one network (the generator) tries to create a realistic human image, while another (the discriminator) judges how real it looks compared to genuine photos. This back-and-forth process pushes the generator to produce images that are increasingly hard to distinguish from actual photos.
Diffusion models: These start with random noise and gradually refine it, using what they've learned about human faces to "denoise" the image into a crisp, photorealistic portrait. This technique has become especially effective in modern AI image generators due to its ability to produce detailed, nuanced results.
High-quality, nuanced prompts specifying features, lighting, and context (like "soft shadows," "natural sunlight," or "35mm lens") can further enhance how natural the output appears.
Key reasons for realism:
Pattern learning, not copying: The AI builds a statistical understanding of how faces are structured and rendered, allowing it to synthesize entirely new, unique human images that "look real" but do not match any specific person from the dataset.
Advanced refinement: Models use extra steps to sharpen details, improve textures, simulate realistic lighting and shadows, and faithfully represent skin, hair, and other human features.
In summary, the realism of AI-generated human images comes from the models' ability to deeply learn and replicate the complex visual patterns of human faces using massive data and sophisticated neural network architectures... By Perplexity AI...
 
Didigal Productions did a lot of VFX/SFX work for Tron... If I remember correctly they had a Cray 1 computer and then got the first Cray XMP system... They also had IBM and DEC "mini" computers as well as a handful of early graphic workstations... :cool:
 
Last edited:
Back
Top