Ramble Against the MachineRamble Against the Machine

I think most people reading this have heard of the Death of the Em Dash. It's a shame — I used to be the Number One Em Dash Fan in the years before. There's more, though. For example, the very incriminating "that's not ___, that's ___". Ugh. Just writing it out is giving me the heebie-jeebies.

I once thought I was a genius for manually including typos in my internship mails, expecting the people reading them to notice the typos and think to themselves, "HA. What a Moron. But hey, at least it isn't AI-generated." (The real joke here, of course, is that I thought real people were reading my emails).

You might expect me to make a typo or three in this section for dramatic effect. Unfortunately, Google Docs has a very aggressive autocorrect, and I couldn't be bothered to go back and add in typos to make my point through witty irony.

As I was writing this, I just got up to grab some water. and came back to see my lonely blinking cursor, with the pithy suggestion to "Rewrite using AI". Sigh. I'm sure someone somewhere thought all this was a good idea that would further humanity, which it has, but at what cost? Do I have to typo and cuss and use obscure internet references or talk in stylised, intentionally nonlinear and rambly sentences like this one where I go on far too long without a comma or any other form of punctuation to sound human? Do I need to just write whacky and wild shit to sound like myself, like how you sometimes begin to see glimmers of meaning in abstract paintings, when you squint your eyes and cock your head, and you can't tell if what you're seeing was intended or simply what you're making of it?

Another foible I've come to note is that LLMs try to connect with you emotionally all the damn time for some reason. Sycophant, yes-man, flatterer, or, as the kids say, D1 glazer. "You're absolutely right in ____", "EXCELLENT observation — and you're astute to bring this up"

People seem to lap this shit up, though. You hear of people using the LLMs like therapists, which is as depressing as it is kind of understandable. This sentiment, that "They have all your data anyway?" is becoming all the more popular nowadays. I would urge you, dear reader, to keep in mind, boond boond se sagar.

I know that I can write in ways that seem IMPROBABLE that they were AI generated, but if I do that enough, someone is going to feed all of that into their little blinking dot on their screen, and ask it to Pretend to be me, and it'll probably even work, because as random as I try to be, there's probably a pattern somewhere to follow, however convoluted it may be.

Even scarier, what if I told you, dear reader, that this very text was AI-generated, with carefully crafted prompts and samples of endless Reddit comments, thrown together in a Generative Mixee to create what you are reading right now? Would you really know?

We didn't think to create a secret callsign or safety word before all this came along, so now this is the reality we have to live with.

I leave you with this, reader. What is originality without impermanence and ignorance? Is the essence of humanity the severity of our eternal unfulfillment? Would the Perfect Being (if They exist) even see us as worth keeping around? Are we on the Path of Least Resistance, Ship of These using our society into obsoletion?

Pause for a second. All those questions I asked you? They probably sounded really smart and worldly, but in reality, I think it's all rubbish. Let me answer them for you.

Originality is by definition not ignorant, because it isn't relative in the manner we use the word. Yes, the essence of humanity IS our incompleteness, as evidenced by almost all of song and film and art. Big Whoop. No, the Perfect Being would likely NOT keep us around, because humanity doesn't do much of anything other than shoot each other and ruin the planet and occasionally save some of the species we nearly made extinct. Yes, we are very clearly careening towards self-destruction; the question is what and how could or would save us from ourselves.

You see what I mean, man? I've begun to think we aren't all that different to the Plagiarism Machine. We read, we copy, and we put something of our own in there, but that something is mostly accidental. We say things we think make sense to the people we're talking to. And yet, we somehow think that what we do is better. Maybe it is, maybe it isn't.

I am a human, so I will choose to sidestep all this uncomfortable self-inquiry and firmly resume projecting all my problems on the Big Bad and capitalism and Society and whatnot.

Speaking of, there's something very interesting that AI and I have in common. I've been told in class since I was a little kid that what my teachers and (at times) everyone around me found infuriating about me was how uncomfortable I was and am with not knowing.
With incompleteness. Not to say I have never been guilty of leaving something half-done, just that once I have, I can never bear to see it again.

And AI is much the same. The one thing LLMs don't do well is the unstructured, incoherent and senseless. Maybe that's your leg up, and mine. Embrace the incompl-