Self-hosting

I'm also quitting or rather retiring this year. a couple years ago I pivoted from being a contract/consultant dev to doing consultation work to clean up AI slop that various companies produced.

I had hoped that eventually my clients would get it, would realize that utilizing LLMs for end to end builds and vibe coding your way to a final product didn't work but...no...they aren't getting it. They refuse to acknowledge it.

I'm tired. I'm tired of the same song and dance with every client "hey this app we made isn't scaling and there's a lot of issues with it, can you run a code review and tell is where we went wrong?" "sure you used AI to build this from end to end via a vibe coder who doesn't know any of the languages the AI used. you need to start over with actual real devs" and nope. they just find another vibe coder that they feel can write better prompts.

So I'm quitting. I can't do it anymore. I was hoping this bubble would have burst by now but the way, especially recently, companies are fighting tooth and nail to keep these shit producing agents going because they're way too invested now is worrying. IF this bubble bursts it's going to take EVERYONE with it. Everyone. so I'm getting out now and hope my money is still worth something when it does burst.

Oh I've been on a similar journey, after a year long break I'm back in the education space and things have only gotten worse.

I voiced concern about professors using LLMs to create presentations. The sloppy images with nonsense text are one thing, but it's hallucinating on actual content. Since I'm learning, I can't be certain if something on this topic is true or an hallucination. My criticism was understood as me just not being tech savvy enough for AI.

I'll probably drop the course and get my money back, I could get a chatgpt subscription for a lot less..

I've beefed up my legal insurance because I believe the only way to hammer some sense into people and businesses is to sue when their agentic ai makes mistakes. Currently nobody seems accountable for the output and that has to change.

You join a meeting with a coworker. Your coworker has enabled an AI tool to automatically take notes and summarize the meeting. They do not ask for consent to turn it on. The tool mischaracterizes what you discuss.

I just had a Teams meeting where a business partner added their AI assistant to the call and said that it would record audio/video for training purposes.

They didn't ask for permission or anything, just joined and turned it on. Like wtf? Where does it store the data? What is it used for?

If I wasn't forced to attend I would have left then and there.

I'm convinced that the way we use AI and the way it is marketed by corporations will force us to recognize the fundamental divide between people who genuinely care for human progress and building a better tomorrow and those who either don't care at all or have a very warped and egocentric view of progress. I'm glad that what is described in this blog post hasn't reached my day job yet, but I'd be lying if I said I'm not worried that it's only a matter of time. It's difficult to not feel extremely disillusioned by the current state of the world.

That's a good article, thanks. The core beliefs they articulate at the end really resonate for me, but it was getting hard to convince people they were worthwhile even ten years ago, long before LLMs were infecting people's minds.

• Things that are worth doing are worth doing well.
• Things that are done well require time and effort.
• You make meaning through the doing.
• Ideas are common; effort is not.
• There are no shortcuts.

just watching humanity burn hard-earned resources on something arguably worse than arms race and dotcom/blockchain craze combined. Got to outlast it, in spite.

I feil this one. I had co-workers that used the office picture of another one to give it to chatgpt to change it to a specific style. I was feeling so weird about this... it's hard to explain, like I was revulsed by this

Ouff, this article hits hard… and makes me rather glad I'm trying to do my own thing.

This one hit me hard. It described a lot of what I've been feeling myself lately.

I know I'm in the minority here, but I don't blame AI for the conditions he's describing at all. I'm a little jealous that as an older millennial, he got to experience the golden years of tech work where everyone was getting rich off work marketed as meaningful and socially progressive. Us younger folks that got into tech because of that era are kicking ourselves for not being born a decade or two earlier.

As a gen z-er, I've only experienced exploitation. Skeleton crews where you are saddled with way too much work at all times, and your seniors have no time to train you to do it properly, so you bury yourself in a cycle of burn out and tech debt. Oh, and our starting wages have likely not increased since OP graduated college. So my perspective is that work for large corporations is a joke, and no one actually cares about the output beyond how much money they can extract out of shittier (i.e., cheaper) work. This enshittification of the workplace is why people are using AI first and asking questions never. I don't blame them. I'm using copilot for side projects and it's 10x faster at coding than I am, although I agree with OP, the code can be sloppy and should absolutely require human supervision.

I think what he hasn't quite arrived at as the logical conclusion of his laments is that tech workers need to unionize. It sucks because I do think people of his generation who benefited most from the tech boom would never consider that they would benefit from class consciousness (a lot of them aren't just temporarily embarrassed millionaires, they are actually ashamed millionaires). But yeah, if he wants privacy protections in the workplace to be taken seriously, if he wants assurance that AI will not literally take his job because it was trained to do just that by his company, if he wants to find meaning in human connection, he's looking for a union.