Hacker News Reader

Follow up question: How much effort, money and skill would the project of creating a video like this take if you allow yourself the use of AI*, vs if you forbid it.

It's hard to define precisely what "use of AI" is. Generating images with midjourney/stable diffusion would count (even if you edit them), having AI-generated code that's executed by the server of a SaaS you use wouldn't.

All of it, I have tested it and want to automate it currently.

Edit: the biggest hurdle previously was no midjourney API. Now that DALLE-3 is released it is good enough and has an API

I could be wrong but I think all of this is AI generated. The images, the translation, and I think even the voice/audio

The audio is painfully AI right from the start, it lacks meaningful pauses.

We are going to get flooded with this stuff. Most of it will be mediocre and the majority won’t care.