By Rama, CC BY-SA 3.0 fr, Link
Kismet, a robot head made in the 1990s, is a machine that can recognize and simulate emotions.
I used Plex for years. Plex is a self-hosted media server. In short, you can install Plex on your own computer and use it to stream music, media, and images to your different devices. Trick is, Plex is not free and their 'lifetime pass', a way to skip subscriptions, has increased in value by hundreds of dollars in less than a year. Ples is very easy to set up so that you can stream your media from the Internet to your different devices.
Then, there's Jellyfin, which is open-source software that's built by, well, anyone who's willing to join. It's like Plex, but harder to set up, but it's entirely free of charge to use. The problem here, is that a few of the top code maintainers have dropped out of the project. One of the problematic reasons, stated by one of the maintainers, seems to be attitudes and AI use in Jellyfin.
Then, there's Codeberg, which is a platform that offers Git hosting. In other word, it's like GitHub but without being owned by Microsoft for nefarious purposes, like training AI from anybody's code.
Now, Codeberg have decided to not allow vibe code on their platform. There are several decisions behind this:
- AI-company web crawlers constantly scour Codeberg for code to steal and use to train their LLMs (large-language models that are used by AI)
- the cost of AI on our climate, meaning that mostly horridly useless AI are drawing us closer to the precipice, more than nearly any other technology
- vibe coders are cut off from their community
- hardware—RAM and hard drives, also GPUs—cost a lot more today because of how hardware is being en-masse hoarded by AI companies
I mostly agree with these points, but I'm not sure about the communal spirit. Sure, trusting a sycophantic stochastic parrot feels like a great idea, and while communities differ in strength, capacity, and experience, I think using humans as your community for knowledge instead of AI is like learning to have fun outside of taking heroin; while shooting heroin offers quick release and payback, the long-term results are, for most people, fucking disastrous, perhaps especially to those around the addicted person.
Jellyfin and Codeberg people have naturally got the stick for their stands against AI and the effects of AI, mostly from people who say 'nuh, the new way of coding is here to stay'. Some people in Jellyfin have even graciously pointed out that a lot of AI has been used to write the software, which seems to be a true statement.
As I'm writing this, I listen to the new album by The Strokes. The album is Reality Awaits, and a little more than two minutes into the song 'Liar's Remorse' I hear this:
Listen to the AI tell you what to do today
A year back, the author Dave Eggers visited OpenAI and discussed AI with them:
The effect of ChatGPT on educators’ lives is catastrophic. Whether you intended to do it or not, you’ve made every teacher’s life infinitely more difficult than it was two years ago. So, just let that settle in… If students are using it to compose, which is the biggest tragedy of all, they’ll never learn to write. And their voice is stolen from them. They’ll never have the ability to say their truth and tell their own story. And that’s silencing an entire generation or two.
When I speak with humans, I get a kick. I just listened to a conversation between Nick Cave and Louis Theroux and thought of Cave's Q&A website named The Red Hand Files. This post is one example of how Cave selects questions and thoughtfully answers them. It's a lovely thing to see a person sit down, think, and then tell us of their conclusions.
It's a lot more worth than simply parroting what ChatGPT or any other AI orifice gives you. Naturally, people judge you off the value of your thoughts; I'd rather you said 'I have no idea' rather than just pass me something that AI gave you. At least I'd get a human with real emotions, not autocorrect on bad steroids.
I work as a writer. That's my job: I write. Part of that job is to review what people write, to follow their line of thought, mostly to learn what they've told me. If they use AI, their non-thinking is most often evident: even if the text is factually correct, the human who's given me the answer doesn't truly understand what they've given me as an answer, which makes me feel that they believe AI is an oracle that solves their problems:
- I get a question
- I don't ponder the contents of the question
- I feed the question into an AI machine
- I get an answer
- I send the answer to the person who's given me the question
Imagine being that person and by the coffee machine bumping into the person to whom you've sent the question. You get follow-up questions. You don't know how to answer them.
I don't want to end up in that state. I much rather prefer to say 'I have no idea what you just said, can you please explain it to me as though I'm a five-year old?' than 'Oh here's the answer' and then spout utter trash.
I'm not a fan of AI unless it's used very sparingly and doesn't sponsor capitalist companies that don't care much for humanity. I recommend to see this 1.5-minute-long video to see what I mean: