Raw LLM Responses

THIS. They are so full of shit and fake pearl clutching that I can barely stomach this discussion anymore. For anyone who believes they actually are worried about the "dangers" and aren't just being greedy; that this isn't more like a late 19th century fight over mineral and water rights, ask yourself this: Why have AI's been trained on data that includes some of the most imaginative ways for an AI to destroy and supercede humanity ever conceived? Wouldn't giving it extremely dangerous "ideas" about what it is and could be capable of be the FIRST thing they should be working to remove from the training data and reverse course on? You know... the data it's using? Now go ask Chat about Hans Moravec's theoretical "AI Nano Tree". Ever seen that mentioned in the breathless headlines? Does it make ANY sense to give a conceptual future destroyer of mankind all the best ideas on HOW to go about doing so? I'd love to hear their reasoning and for just ONE F'KING journalist to prove they aren't completely useless mouthpieces of major corporations by asking any of the CEOs this question. The leap from LLMs to sentience is gigantic and still unfathomable. That's not to say there aren't real dangers - they exist. What I'm saying is that's not really what they're talking about a good deal of the time. The alarmism tels me, based on what they're NOT doing, that this is all about greed and control.