update:
March 24, 2026 - As if to show me up, Disney cancelled their AI deal the day after I published this. So. Um. Yay humans!
A few years ago, a friend and I worked on a weekly publication just for fun. Every Sunday we would regret not doing anything for it all week and crunch for pretty much the whole day to make sure it was out by Monday. There was no reason for this, other than that we just really wanted to keep to a schedule.
One of the sections that we included was providing fortune telling. We did a somewhat traditional format using zodiacs, but in addition to that, we added in a bonus tarot reading. Neither of us are professional seers and so we had to learn as we went. And while it was fun to divinate, it was also a bit of a challenge.
Tarot decks have a LOT of cards to work with, each with their own meanings (both regularly and inverted!) so it follows that there is also a lot to keep track of when doing a reading. And to do twelve? Well, it was a pretty massive undertaking considering it was such a small part of our magazine. We needed a solution and so we got to work making one.
Everyone who knows me knows that when it comes to data, I’m a freak in the sheets… the spreadsheets, that is! [hold for applause]
(It makes sense if you see all the other tabs and also can understand what I was doing, I can’t understand that though so good luck)
So anyway, because I love working in cell coding so much, I figured we could cut down on our manual labor by developing a system in which we have all the different cards and their meanings in a database and then call from that database a set of cards to then speed up the process of getting fortunes ready to go. There was still a bit of effort to do, interpretations, writing prose, that sort of thing, but the technical bulk was much less and it allowed for us to get more done efficiently.
I am firmly of the belief that human nature is to be as lazy as possible. This is, of course, in conflict with nature itself in that we must work to survive, but despite this, the effort folds into a desire to do nothing. We enjoy doing things, sure, but was also enjoy not doing things and when we look at human technology, more often than not the creations we use to propel society tend to be ones that reduce things we have to do.
Typically, I am all for this sort of path of least resistance leading innovation. It’s why I like to make calculators in Excel and why I try and get out of doing things (if you know me irl no you don’t, mind your business). But I also just think that calculators in Excel are really neat. You can put in a code that will, with the right conditions, populate a field based on whatever is in a dataset. With enough effort, it would be possible to build a pretty sophisticated chat bot in Excel.
From what I understand of LLMs, they very basically work similarly to that type of Excel lookup, using keywords to pull in data from a set. In theory, and with all I’ve said before this backing it up, LLM “AI” is really very cool to me! Unfortunately, theories don’t always pan out as one might expect.
If you’ve been living under a rock the past few years, how’d you stumble on my neocities site of all places, what??? We all know about AI, I don’t need to give it the modern Surf Dracula treatment. Tons of environmental damage, financial ruin, job market devastation, all for really bad images and videos that no one cares about except for the person who summoned them. Or like, the future, I guess, if you think AI is good for some reason (You’re allowed to! And I am going to judge you for it, sorry not sorry). Why is it, though, that an advanced calculator can be so cool to me while a far more advanced calculator is so repulsive? It has to do with the soul of it all, or rather the lack thereof.
First, chatbots are neat, I don’t care. We can all pretend like those early chatbots on the old web weren’t majorly fun but then we’d all be liars. But why are those fun where ChatGPT is oppressive? Novel chat bots like Cleverbot were fun and interesting. Importantly, though, they were upfront. When users went to Cleverbot, under the text input was a message: “This bot learns to be clever from real people, and its AI can 'say' things you may think inappropriate.” While not particularly detailed, the implication is that what goes into the machine can eventually come out of it. By engaging with it, users provided their consent, something that those running the modern iterations of this entertaining toy either do not understand or do not care for.
(Don’t act like you didn’t try to trick it)
When an AI company is building a dataset, it’s generally not forthcoming about it. For example, a book piracy ring was taken advantage of by AI companies to load up on texts. Reddit, Youtube, all these social media sites that users have been contributing to for decades are all feeding grounds for the machine under the guise of “well, it’s public, therefore it’s for me.” It’s the equivalent of saying “obviously she was asking for it, look where she was posting.” But the thing is, something being visible does not mean that anyone can walk up and use it as they wish. Don’t believe me? Go to your nearest fast food restaurant and try to paint the lobby without asking.
Of course, what we put online, we understand it may be seen. We WANT it to be seen. If this was a collection of entirely private thoughts I’d write it in a private diary, not publish it on the web. But these are still MY words and my work and just because I consent to people reading it doesn’t mean I agree that anyone can take them from the page and plaster them how they want. AI companies tend to disagree, which is interesting because for some reason they operate on a different ruleset than the rest of us.
In 2000, Napster was sued because it allowed users to share music files over the internet, bypassing traditional purchases and leading to large fines and the collapse of Napster. In 2024, Suno AI was sued because it, probably, built its entire training set on copyrighted works. Although they claimed that nothing it output was like any copyrighted music, they settled in 2025 and plan to have a new mysterious dataset to generate sound with in 2026. What’s the difference between these two? Both stole music, but is it fair to say that the intent is important in determining why one is less wrong? Sharing a song with someone is a celebration of art. Having a machine dissect and regurgitate a song is not.
(Fun fact, one of the Napster founders went on to be a big part of Facebook, wow!)
Importantly, though, Napster is dead and Suno is still going strong, somehow. And even worse, it seems the music industry was the most bullish against these sorts of practices. Shortly after Disney announced an anti-AI lawsuit, they announced that they were dropping it and instead investing billions in AI. Most companies don’t seem to want to touch the subject, yet they all silently wait to take advantage of it, testing the waters here and there and waiting for the collective intolerance to drop enough so that they can flood us with endless cheap garbage.
But that’s the thing that makes it all so silly. They know it has a bad reputation, generally speaking, so they try to be subversive with it. Oops, we forgot to take out AI used as a placeholder. Oh gee well actually we just use AI as a tool to help out a little and that’s why long-standing systems suddenly crashed and cause major outages. It’s not OUR fault please understand!
I can’t understand. I can’t get behind any of this. It isn’t a construct designed with purpose, it is a pool from which the eviscerated thoughts and passions of human beings are dredged to form a semblance of life. But it’s not life, it has no intention behind it.
My tarot generator might seem a bit soulless, especially to a real fortune teller, and I will grant that compared to physically shuffling a deck and drawing cards, there is less personal care behind it. However, it is also something that was created with purpose and meaning. Each description was written out, not stolen. Every interpretation was decided by a person looking at a draw as a whole. What makes this so different from simply asking Grok to give a fortune is that the tarot calculator is a tool to replace a physical deck and whatever answer Grok provides is not a tool. It is homogenous generic nothing based on the mashed ghosts of writings past.
Maybe I’m being too harsh. After all, I did say that searching for convenience is human nature, right? As much as I firmly believe that, I also believe we vitally need adversity. Fulfillment doesn’t come from being lazy. If that were the case, people who do nothing would be the mentally and emotionally fittest and I think we all know that’s not typically the case. Laziness only feels good when it’s achieved and having an LLM vomit up a hodgepodge of past achievements does not provide that same sense of appreciation.
I’m not into Dark Souls. I have tried it but it’s very oppressive atmospherically in a way that just makes it difficult for me to enjoy because I’m a soft thing who likes joy too much and also zombies freak me out and I can watch movies about them but in a game? I don’t think so, get away ahhhhhhhhhhhhhhhh. Um. Anyway. I like them as games in theory, though, and I play similar titles like Nioh 2 or, less similarly maybe, Monster Hunter. A common issue is the difficulty which acts as a barrier to entry for some players. These games don’t hold your hand and guide you to success (except recently, MONSTER HUNTER WHAT ARE YOU DOING!!! okay, focus, Lopp, we’re doing a thing here), instead trusting the player to figure out how to play them and progress. Sometimes it takes a lot of failed attempts before it finally clicks and when it does there is a feeling of elation because oh my GOSH I DID IT YES I RULE!!!
(Image unrelated)
Imagine, though, instead if at the first sign of difficulty, the game yanked the controls away from you and performed on its own, deftly felling whatever boss that was in the way. Does that euphoria still exist? Surely, some doofus is out there thinking “of course, it feels good to beat the game, pushing buttons is the same thing, you’re not really doing backflips, duhhhh” but this is just some strawman I made up to be just as disingenuous as its response. There is no satisfaction because there is no overcoming. Just like we can feel fear or happiness when watching a character in a movie but never feel like WE are IN the movie.
When an LLM does all the “work” (in quotations because also they hallucinate so is that even work? If you tell your mom you took out the trash but actually didn’t, then that doesn’t make the trash out it just makes you a liar who didn’t do the work), it takes away any chance of feeling anything. The process is trivialized and so are the results. People wonder why no one wants to watch their AI movies or listen to their AI music or read their AI books, but they never take into consideration that nothing went into these things, no work, no passion, no care. The reason humans appreciate art is because, even if it’s subconsciously, we recognize the enormity of the effort and talent that went into these pieces. And despite LLMs being a pool of stolen human works, nothing actually went into the output except a random word generator.
To me, that random word generator has the potential to be really interesting and I like the concept of it a lot. Unfortunately, it has a lot of inherent problems. I feel at this point I’ve expressed enough disdain for the substance that makes up their datasets. However, that specifically extends to the other horrible side of this monstrosity which is the people who are running these things are disgusting, immoral, and unethical.
Disgust isn’t that easy to quantify, being a gut feeling (haha get it?) and all, but by explaining the other two, maybe it will make more sense. People running LLMs are immoral and unethical because of a simple reason. They are operating with billions of dollars, have insanely vast pools of wealth to call upon, have exceptional connections in a variety of industries, the reach to make pretty much anything happen, and even with all that they still turn to digital piracy, something for which regular people are STILL facing punishment. I’m not exactly anti-digital piracy, but I am very much opposed to unfair treatment. And when there are legal cases, as mentioned they are brushed away or strings are pulled to have the plaintiffs change their minds entirely. All of these artists who are involuntarily feeding the machine COULD be offered compensation, the money is their, but it’s apparently more expedient and cost-conscious to just bypass all ownership rights. Which, that seems to be a running theme with these people.
(Hey look at that, another Image unrelated)
“We want to make the world a better place,” the corners of their lips creak upwards as there eyes remain frozen. It is a simple lie they peddle to justify all they break. They are making the world a better place! Except for artists and analysts forced out of their careers as CEOs chase new trends. They are making the world a better place! As they buy sprawling tracts of land that could be farms or houses or just nature, even. Land that will one day hold massive buildings brimming with energy, sucking up every drop of water nearby, raising cost of living prices to unprecedented levels. These sites don’t even create jobs, though. A controversial data center in Ohio is expected to create 10 jobs at the cost of millions to everyone in the surrounding area. So I guess I have to ask, what are the benefits of these LLMs?
So far, the only “good” (quotations because this is being really really generous) is that companies can save money. By eliminating humans. And how’s that working out for everyone? Well, Windows 11 seems to be. Um. Yikes. Well, okay, how’s the Cloudflare code? Oof… Uhhhhhhhhh okay but surely the art is good? How many Oscars went to AI this year? Hm, that actually did okay because apparently Netflix’s Frankenstein won some production and effects awards. They had AI, so I guess that settles it, the billions of investment was worth it. Everyone go home, stop caring about life and living, plug into Grok.
I don’t think this has to be a situation where there is a total victory for either side in the debate on modern LLMs. As tragic as they are, there is some potential for these programs to be useful. However, in its current state with how its being run, there’s no other reasonable response than to despise this anti-life, anti-thought, gross stupid bad calculator. Resistance is easy, we aren’t bound to some dystopian future. Make your own fate (and calculators).