"Automating Video Creation with LLMs: My Journey & Future Plans"

Prompt

Script

I started dabbling with LLM models last year to understand their capabilities first-hand. I understood they were good at generating text outputs very well, and few models were becoming good at generating quality images. Although the modality of LLMs (largely) is text, it's clear that the modality of the internet is largely video. While other modes of content can be successful (like long-form blogs, Twitter, newsletters, etc.), they ultimately have a far smaller audience (due to the shorter time spans and just people's laziness to look at small squiggles on their screens). So if you want to win big on the internet and reach the masses, you need to bring video into the mix. But among the different modes of content on the internet, video creation (and good quality videos) is the most time and effort-intensive process. And now that I could see a path to create videos with LLM models, I wanted to explore the possibility of improving the video content creation process. So I started working on ShortsKing, beginning with automating short-form content creation. Although I like what I have built in my spare time, there is so much more that can be done. So I like what Invideo is doing, and directionally that is where I want to go as well. I would love to talk to the team, discuss ideas and possibly get to work on solving quality content creation at scale with Invideo.