Fuck AI

I'm a line worker in a factory, and I recently managed to give a presentation on "AI" to a group of office workers (it went well!). One of the people there is in regular contact with the C_Os but fortunately is pretty reasonable. His attitude is "We have this problem; what tools do we have to fix it", and so isn't impressed by " AI" yet. The C_Os, alas, insist it's the future. They keep hammering on at him to get everybody to integrate "AI" in their workflows, but they have no idea how to actually do that (let alone what the factory actually does), they just say "We have this tool, use it somehow".

The reasonable manager asked me how I would respond if a C_O said we would get left behind if we don't embrace " AI". I quipped that it's fine to be left behind when everybody else is running towards a cliff. I was pretty proud of that one.

I'm so sick of fixing AI slop code, especially because there's no love for people who fix the slop, only for the people who shipped the slop.

They have to justify the cost of the consultants they paid to tell them to spend money on it.

Any boss ramming a tool down their workers throats without understanding it or validating it's usefulness is not a particularly good boss.

It's a productivity miracle for manager because they can bullshit their job faster and easier.

We just had an all hands where they were circlejerking about how incredible “AI” is. Then they started talking about OKRs around using that shit on a regular basis.

On the one hand, I’m more than a little peeved that none of the pointed and cogent concerns that I have raised on personal, professional, hobbyist, sustainability, environmental, public infrastructure, psychological, social, or cultural grounds - backed up with multiple articles and scientific studies that I have provided links to in previous all-hands meetings - have been met with anything more than hand-waving before being simply ignored outright.

On the other hand, I’m just going to make a fucking cron job pointed at a script that hits the LLM API they’re logging usage on, asking it to summarize the contents, intent, capabilities, advantages, and drawbacks of random GitHub repos over a certain SLOC count. There’s a part of me that feels bad for using such a wasteful service like in such a wasteful fashion. But there’s another part of me that is more than happy to waste their fucking money on LLM tokens if they’re gonna try to make me waste my time like that.

I don’t work with computers or coding, yet even in early childhood education/therapy some people are pushing for AI. Someone used it to make “busy scene” pictures for students to find specific things in. I hate using them. Prior to this, we used “busy scene” images that are easy to find online, full of quirky, funny details that the kids enjoy spotting.

But I can barely look at the slop images that were generated. So many of the characters have faces that look like wax figures left in the hot summer sun. The “toys” in the scene are nonsensical shapes somewhere between unusable building blocks and poorly-formed puzzle pieces. Looking at the previous, human-made pictures brought me joy, but this AI garbage is a mess that makes me sad. There’s no direction, no fun details to find, just a chaotic, repetitive scene. I bet the kids I work with could draw something more interesting than this.

Bosses aren't oblivious, AI isn't for the workers benefit. They need the workers to use the AI, so it can improve and begin to replace them.

Our new tech lead loves fucking AI, which let's him refactor our terraform (I was already doing that), write pipelines in gitlab, and lots of other shiny cool things (after many many many attempts, if his commit history is any indication).

Funnily, he won't touch our legacy code. Like, he just answers "that's outside my perimeter" when he's clearly the one who should be helping us handle that shit. Also it's for a mission critical part of our company. But no, outside his perimeter. Gee I wonder why.

I was asked to create a simple script... Great, I could have knocked that out in maybe 3 or 4 hours.

Boss insisted I use A.I. ... Fine whatever.

The code it spit out was OK, but didn't work... So I took it and started re coding and fixing the bugs.

It took over 3 hours to get that sloppy code to a working state.

Boss asked why it took so long, ai works in seconds. He didn't understand that I had to fix that crap code he forced me to use

Look, ai does pattern matching like a champ. But it can not create... It doesn't imagine...

Management never has a clue what their employees actually do day-to-day. We're just another black box to them, tracked on a spreadsheet by accounting. Stuff goes in, stuff comes out, you can't explain that.

Honestly at this point AI is bad and human critical thinking is the worst I've ever seen in my life.

I know people that I expect would collapse inward without AI holding their hands, and here's the surprise of this statement. Can't wait to see it happen. I'm really holding on for the implosions and REALLY hope they happen when I'm nearby.

I have a simple anwer why managers think its smart and workers things its dumb. The managers see all kinds of documentaion from workers and to them the AI slop look the same. It looks the same due to the fact that the managers never take the time to comprehend what they are reading.

I used some AI at work to do some stuff in polars, because I don't really know that library very well.

As a result I have a function that does what I asked for (I wrote tests), but I don't understand it and didn't really learn anything. Not a great trade.

Most of my conversations with my management is forced to be talking them out of the heinous baloney they're convinced of because "Gemini says.." No boss, Gemini made some shit up. Scroll past it or stop wasting my time.

Dilbert manager energy

And the only reason they can get away with not charging the training and computation costs is bunch of rich people essentially gambling a small portion of their generational wealth.

it's just great at pretending to do something, good enough to trick stupid execs

One of my college professors is mandating chat gpt