Just had another chat session with OpenAI's ChatGPT system; that stuff is incredibly fascinating and thought-provoking. Literally feels like talking to alien intelligence. Go on an adventure with us about a human astronaut, Captain Nova. Check it out 🧵 #storytelling#chatgpt
Here, ChatGPT just started the story, but I actually liked its first suggestion so let's just go with it!🧵
Getting a bit too generic? Just ask the system to give it a more original spin.. I'm blown away by how coherent and well-formulated this is. Game-changer..🧵
I didn't like Captain Sam, so here's ChatGPT's second suggestion of making it a Captain Nova instead. 🧵
The story should be stranger, so I suggest that Nova's reality should start to become unhinged from ours. I usually find that ChatGPT just comes up with an answer, no matter how strange the request is.. very impressive.🧵
I wanted things to get weirder, so I asked ChatGPT to have Captain Nova try to increase the distortion.🧵
Here, I ask the bot to take Nova into the other dimension. And I asked it to describe in detail what she saw there. 🧵
Land on the planet and make contact with the aliens ofc: 🧵
Whenever it feels like ChatGPT is trying to wrap up the story, just ask it to continue in a new direction or to go into more detail:🧵
It's insane how good this system is.. damn. I have no doubt that our world will change radically because of tech like this.. let's hope we can use it for good. I personally think that this use right here is pretty good! :D The alternative is Skynet..🧵
I got worried here when ChatGPT named the first scientist, but it got better with the next ones:🧵
Let's end on a cliffhanger. That ends this story.. let me know what you think of this. Are you as impressed/surprised/scared/fascinated by this as I am?
It's one of my absolute favorite aspects of Veo 3 i2v. It's ability to transport elements through the latent space, with this simple but very powerful prompt structure:
“Instantly jump/cut on frame 1. [Describe the new context]"
With this structure, I was able to use my quick drawing as the main subject in all the scenes you saw here. Seeing how the model is able to retain my personal sketch in all of these different contexts.. Feels magical honestly 😍
My personal opinion about the question asked here, should be clear if you've followed along with my posts 🔥🎨
Take a look at how consistent my sketch is, despite the change in lighting and the change in the underlying material.. The way the water distort the image.
Creating something as realistic as this in traditional 3d would be VERY difficult.
Notice how the model stays on the original frame for a short while and then jump/cuts.
You can also use these techniques to transform your input.
Here, the drawing is transformed into an oil panting. You could use a prompt like “Instantly jump/cut on frame 1. The oil painting of the sketch from before, is seen on the wall of the modern museum. A single light shines on it, illuminating the 3 characters in the painting. A group of people is watching the painting."
As you know, my explorations of the Gen AI space is ultimately all about creative control. You should be able to shape the generative matter using all your artistic sensibilities and your aesthetic sense.
OpenAI's Sora is a huge technological leap, but what excites me the most about it is the modalities where it depends on input other than text alone. Such as video to video. Here's an example of how Sora can change an input video.
Base video🧵
Prompt: Change the setting to be in a lush jungle.🧵
Prompt: change the setting to the 1920s with an old school car. make sure to keep the red color. 🧵
So you might have thought that the sketch to video was impressive, and some of you even asked: "But what about 3d?"..
Here's my answer to that.
My sketch transformed to an illustration with Vizcom ai and then turned into a 3d model with Tripo ai. Expect more tests of that workflow in the near future.
#3d #ai
Curiously the system didn't manage to make an actual hole through the donut.. but the worst mistake we can make is to think that that's reason enough to dismiss this tech as useless. Remember how we all used to laugh at the wierd ai generated hands? 😅