I have seen LLMs used to karma farm in order to create sellable accounts, but this account is 12yo which I find more disturbing because it means it used to be a real person’s and is now being used for something else.
All Reddit is, is bots upvoting bots. But the shareholders are fucking morons so it doesn’t really matter. Reddit is dead.
How do you know it’s AI? If we’re all being exposed to its quirks, there will be subconscious feedback on our writing styles.
It’s the classic AI writing style — restating the original point in unnecessarily elaborate terms, explaining the joke as if the reader needs a literature review, setting up a false dichotomy, and ending with a neat little generalized takeaway. Add some em dashes and words like “rather than” and “as though,” and you’ve got the whole package. It’s not any one quirk; it’s the cumulative effect of sounding polished, verbose, and weirdly over-explanatory without actually adding much.
Posting highly formatted but substanceless empty fluff is not a “writing style”, it’s spam. A human is certainly capable of doing that if they set their mind to it, but occams razor strongly points to automation.
i guess the em dash is a red flag too
its likely AI bots, upvoting, commenting on AI bots.
Lmfao, umbrage.
Umbrage is our word we need to take it back
People should award arguments on the merits. Why should where the argument came from influence evaluation of its present merit?
At many times I’ve been accused of
- spamming
- being AI
- being not real
- being too weird to be real
Etc. gonna need more than that. Nouveau McCarthyism needs to justify its continued existsnce otherwise GUH-BYE!
Why should the argument source matter? Because the whole point of social media is for people to communicate with other people. We want to share ideas with each other.
If all you can do is copy/paste, you’re a failure. If you were in school, you’d fail for plagiarism. You’re wasting electricity and wasting our time. Sorry.
Or, if all you want to do is share someone else’s idea, just put a URL to the place where they said it. Don’t pretend it’s your contribution when it’s not.
deleted by creator
The problem with automating social media posts is that they definitionally cannot have “merit”. The primary function of slop is not to “argue” but to “flood the zone with shit” and create a plausible consensus for whatever opinion the person running the LLMs wants. We saw this with paid humans before LLMs. It’s not new.
A human will never be able to win a debate against an LLM bot “on the merits” because their function is to waste your time. An LLM cannot stand on merit, cannot hold a position, and cannot construct an argument. The words that are output by an LLM have no connection to any inside mind or model of the world. The only way to win a debate with an LLM is the same way you win a debate with a troll: identify the hallmarks and refuse to engage.
Im just saying, the McCarthyist argument of “its AI/LLM” has to be dealt with at some point. If an LLM can argue equally or better than ahuman, focus should probably shift to the axtual argumeent as the nexus of discrediting its alleged AI source, rather than shrieking “ITS AI/LLM!” being sufficeint to shut down discussion
An LLM cannot “argue” full stop. If you find yourself in a debate with the text outputted by an LLM bot you have accomplished nothing besides wasting your own time (which is the entire point).
deleted by creator
One cannot win an argument that does not exist
deleted by creator
I’m starting to understand why you often find yourself accused of being a bot… 🙃
How many people use social media to talk to LLMs instead of people? I don’t think most people use social media under the impression comments they respond to and read are AI, but actual people.
deleted by creator
You may enjoy talking to chat bots, but I don’t have interest in that.
