Quick Note: Grok the Fashbot
[the title of this post is meant to be read to the tune of "Rock the Casbah"]
If you're reading this post, you probably know the following accursed series of clauses: Elon Musk's personal Large Language Model-built chatbot, Grok, which operates as a user on X, formerly Twitter, has once again been tweaked. This time, the micro-adjustment to its underlying algorithm is intended to produce genuine straight-up Nazi propaganda. Talking about advocating for a second Holocaust, "noticing" common last names (a Nazi meme about "noticing" who "controls the media"), etc.
As should be incredibly obvious by now: Elon Musk is a Nazi. His companies exist to further his eugenicist goals. X, formerly Twitter, and by extension the official X chatbot Grok, are no exception. He and anyone who works for him are effectively Nazis in any meaningful sense of the word.
But I want to emphasize another aspect of this whole evil, stupid, evil-stupid debacle.
Elon Musk is not an original guy. He doesn't have a lot of ideas. Instead, he sees something, it rattles around in his empty skull for a couple months-to-years, and he then regurgitates it. There was a time - eons ago - when one of his more infamous online posts was him just reposting other peoples' memes with the watermarks filed off. When his companies are successful (really just Tesla, AFAIK - and PayPal back in the day, if you want to give him credit for that), it's not because of original ideas from Elon, but rather because of a combination of his own vast private fortune, massive government grants, and a knack for riding existing waves.
He's also not a subtle guy. His humor is Cringe. His gestures are broad. He couldn't be more on-the-nose if he were a pair of glasses.
So here's Elon Musk with his dumbass chatbot, named after a genuinely-useful term from Heinlein's Stranger in a Strange Land (I used to use "grok" in conversation. Less so these days). He dickers sloppily with the code in an attempt to troll Gavin Newsom, or to advocate for genocide, so that when users ask Grok what's going, they get served a warped, Muskified version of reality.
What's to stop other LLM-based "AI" companies from doing the same thing, in a more subtle, less Musky way? What's to stop ChatGPT from, say, deprioritizing leftist thought? What's to stop Microsoft Copilot from making sure to never mention any ties between Bill Gates and Jeffrey Epstein? What's to stop any of these companies from making subtle adjustments to the little magical man who lives inside the chatbots?
Never trust a source you can't trace. Never trust "AI." Even if you insist on using it - and please, don't, but I understand I can't stop everyone just by begging - double- and triple-check the facts it gives you. It is very, very dangerous to turn your reality over to a black box, especially when that black box is controlled by some of the richest, least-morally-sound people on the planet. Not all of them are as unsubtle as Elon Musk. Although to be fair, they're all pretty unsubtle. Having billions of dollars has a stultifying effect.
This post brought to you by the Butlerian Jihad: "Ridding the world of Thinking Machines since ten thousand years from now."
