I’ve been largely skeptical since I finished the AI concentration in CompSci for my undergrad. I’m gonna prob draw flak from this audience for saying this… bu

I’ve been largely skeptical since I finished the AI concentration in CompSci for my undergrad. I’m gonna prob draw flak from this audience for saying this… but the con is right there in the name. Artificial Intelligence. We don’t know what intelligence truly is. And we damn sure don’t know what an “artificial” one is. But there’s a spark that it strikes in us. An inner curiosity. A fear. Who are we? Will we always be the smartest thing on the planet? The last one seems to particularly strike a fear in those with an outsized reliance on a perceived intelligence for their self worth. Elon?… does he read HN? I’m not just being cutesy with language here. It’s a field that natural draws in those technologies that appear to be human, but may or may not be useful. That’s been happening since Eliza. Marvin Minsky even admitted to playing this card for grant money. Notice it’s not a field named after a particular technology. And it’s certainly not aimed after solving a specific problem. Anybody on here ever mention that what they wanted billions for was to create shakespearean plays in the style Tupac or what not? I’m guessing not. And I’m not saying that I’m immune to the fascination. That’s why I studied it in undergrad. It’s not that these technologies are worthless. ChatGPT is very impressive in what it does and if you’d asked me 5 years ago if we would have had this by now, I wouldn’t ever had thought so. And it’s not that the field doesn’t concern me. But my major concern isn’t AI super intelligence. It’s AI super stupidity. Like when some glamor chasing politician gets sold a crappy product and it shuts down a power grid or goes awry on a major weapon system. I’m embarassed to admit that the Skynet scenario is actually potentially forseeable in our future. And that Rocko’s Basilisk nonsense? Come on, guys. It’s like we’re not even trying anymore… And I’m gonna call my shot here. You know what’s far and above the most useful technique in my experience that could at all be called AI? Bayes Law. From the 1700’s or whatever. And Bayes didn’t even publish his own work. Doesn’t take a billion dollars of capital or a supply of GPU’s for that one. Coincidence? Prob not. edit: Marvin needed his last name.