I've been meticulously rebuilding my laptop over the last few days, installing only a handful of pieces of software. I was really heartbroken to find that one of my favorite pieces of open-source software, one that I depend on almost daily, has become infected with the AI virus.
It's become clear to me that at this point, any corporate or otherwise closed piece of software must be assumed to contain AI slop but the sadder and more surprising realization is that the same is true for open source software as well. Any open source project that doesn't boldly claim to be AI-free in the project description is likely to contain AI-written code as well as AI features.
This left me distraught for a few days, but this morning I'm starting to sort things out.
First, I'm not going to escape using and depending on things that use AI today, tomorrow and probably not for awhile. If keeping it out of my tiny list of software necessities isn't possible, what hope is there? Even Linux itself is falling into this trap, so there's no obvious, immediate way out through the existing open-source ecosystem.
Until I can resolve my dependency on these things, I'm going to have to find a way to accept them. That doesn't mean giving-up or becoming complacent, it's just means that this is the way it is until I can do something about it.
I've thought of two things I can do about it.
In the short-term, it's possible to utilize the openness of open-source to go back in time to a point before this nightmare began. It is possible to fork the open-source projects I depend on at a point in time before they became tainted. It's not easy to tell exactly when this happened so I'm going to do a little homework on when LLM-generated code became a thing and then fork all the projects I need on the same date. This means I'll be running "out of date" code and that's fine, right now my favorite computer is from 1983 and runs "software" (firmware?) from 1985.
The long-term solution is to depend on things that are made by people I trust. That trust includes a commitment to making the things themselves which by nature excludes the use of AI. Surprisingly I only know a couple of people like this, at least in the software development space. Effectively this means I'm doing it alone for now, but I'm hopeful that there will be others who want to join me soon.
It's very hard to let-go of the feeling that all of this is unnecessary, that it could easily be avoided. It's really a blow to my respect for open-source software that so many projects have taken this path. Anyone who bothers to look can see that AI (and especially LLM's) is a fundamentally and irredeemably fascist technology and once you have this knowledge continuing to use it can be interpreted as nothing other than being complicit with fascism. I can't believe that some of the developers I've grown to love and respect are fascists, so I can only hope that they have been keeping themselves in the dark.
The good news is that there are at least two paths forward, and I think both are complimentary. The result will be a big step forward, one that we've needed to take for awhile.
Jason J. Gullickson, 2026