A funny blog post So a while back I read a blog post Izzy Miller where he describes how he used a fine-tuned LamaLLM (Facebooks “leaked” LLM model) to imitate his friends in their years-long ongoing groupchat. Basically he took the group chat in iMessage, dumped it all out into a database and built training sets for each of his friends to emulate their style of speaking, types of jokes, etc.

Or how to wrap something up and gain existential dread It’s a two-for-one. I know that I’m not the only one to get this. And I’m sure there’s cool, silicon-valley, pop-psychology lingo to describe it like, “post completion affectation” or something. Fuck all that. The fact of the matter is there is something scary about putting something out in the world. Let me back up. In a previous post I mentioned that I was working on something, and now it’s getting to the place where I can put it into the world.

So I’m working on something new (no announcement just yet, but I will once it’s time). It’s along the lines of some of the other development posts I’ve had here, and like all good ideas came from a problem I wanted a quick hack to solve and it worked a bit better than I expected. So I decided to turn it into something “real”. You know, that kind of thing that you hack together over a weekend, show to a friend, and they go “let me get that!

So in playing around with ChatGPT some more, I wanted to see how well it did at writing code. I had a thought that I spend a lot of time on Hackernews, and because of it’s simple format, you don’t always know if you’re going to get something interesting or not when you click a link. So I thought to myself, “Self; why don’t you make something that summarizes hackernews for you?

This is just a copy-paste of the conversations that I had with ChatGPT to write parts 1 and 2 so that people who are curious can see what the process looked like and the type of edits that I made to clean things up a bit. It’s very long and I’m not sure how much value it will give folks, but I thought it might be interesting to some. Part 1 Model: Default (GPT-3.

Westworld part 2 Introduction In the first part of this blog series, I shared my video game project that uses ChatGPT for character dialog generation, Unity game engine for building the game, and ElevelLabs for voice generation. I talked about the technologies I used, how I designed and developed the game, the challenges I faced, and some fun findings I discovered during gameplay. The game was inspired by the TV show Westworld, and my goal was to explore generative story telling and emmergent narrative gameplay experience by using AI-generated dialog and interactive decision-making in a Wild West-themed setting.

Introduction Hi there! I’m excited to share with you my latest project - a video game that uses ChatGPT for character dialog generation, Unity game engine for building the game, and ElevelLabs for voice generation. The purpose of this blog post is to give you an insight into the development process, challenges faced, and learnings gained throughout the project. I believe that the use of AI in video game development has the potential to create more immersive and dynamic gameplay experiences.

This blog post will make you mad So fair warning here; This blog post is going to make you angry. Regardless of your political beliefs, there’s going to be something in here that’s going to piss you off. And that’s okay! Most of the time, angry is an easy emotion to get people to feel so that they keep engaging in content and clicking on ads. Outrage-based content is all the “rage” (get it?