I’m enjoying my exploration into AI image-generation quest. It’s been a rewarding experience, but I can’t shake the feeling I need to start each post like this with an apology because, you know, AI image generation.

Unlike other OpenAI’s DALLE or Midjourney, Stable Diffusion is open source and does not require using the cloud. You can run it at home, on your own hardware, provided you have enough juice for it (an average gaming rig should be good enough to start).

Stable Diffusion gets a bad wrap because… well, Let me ask you this: What happens when you give a boy a tool that can create any picture he wants? What do you think is the first thing that boy will create the second the door closes and the lights are off?

But that doesn’t mean you can’t use it for other things. Having AI image generation as open source makes a big difference. It’s up to you to use it as you see fit. It’s up to you to train it and provide it with images (you can create a model based completely on images of yourself or your artwork; stealing copyrighted work is not necessary). It’s up to you to learn how to use it, and man, there’s a lot to learn.

If you’re a fan of FOSS and Linux, you’d probably like to know that Stable Diffusion can run on Linux. In fact, that’s its native environment. It’s not as nearly as simple as typing a prompt into Imagen, Google’s AI image creation tool, but if you’re a Linux veteran, I doubt that’s a problem for you.

And I hope you’ll try it. We need more people who can develop and understand AI technology that is independent of big tech AI poison that is everywhere. We actually have a chance to develop a moral, responsible, and private AI tool - all the good stuff we FOSS techies appreciate. And now is the time. Before the entire field is swallowed completely by capitalistic greed.