daily.xyz

0xf…dd8

ERC721

Early explorations on merging generative stories with generative video: what happens when an artist provides only a single sentence as a prompt for the entire video.For StoryGEN1, I'm building an app that takes as its input something as simple as a single sentence, and then breaks it down into a story, and specifies the elements of a short movie (video, audio, dialogue), then automatically generates everything and puts it all together.People talk about "the Netflix of the future" where the audience member describes what kind of movie they will want to watch. This will become reality in a number of years. A movie consists of a huge amount of creative choices. What happens when the AI makes all of these choices?People often dismissingly talk about AI artist "just writing a prompt". I've spent weeks creating a single AI video, so of course currently there's a lot more involved in creating one. But those comments and my experiences lead me down this path of thinking "what would it mean if it is true" - if one could just write a single prompt, and out comes a short movie?I firmly believe AI acts as a mirror - it has learned a model of us, and then when it reflects it back to us, the image it paints can sometimes be very revealing. Now that I apply the same not only to the visual generative side but the story side, more aspects start to emerge: how does a LLM tell the story of us? How do we behave? How do we feel when see this portrayal?

Name

Roope Rainisto - Woman In Blue Top Fails To Pick Up Man At A Bar

Owner

0xca177D4b3E6d4C2778912B47e6aAe49b7c429819

ID#310007

Rarity

X-HI:15%