Raw LLM Responses

The problem here is, first and foremost,that the scientists who have developed this technology have completely lost any kind of meaningful handle on how AI actually works; how the AI thinks—how it uses, processes and draws conclusions about the data it takes in and retains. This is not a new phenomenon. Go back to some of the earliest AI that were programmed to dominate the chess world; the scientists even then were at a loss to fully explain how the very systems they created actually worked. How can we possibly control machines that we fundamentally don’t understand? That we understand less than the neural network and processing of our own brains? Yes, it’s that bad. I appreciate Blake Lemoine’s words here, but have to say he seems oddly restrained. I don’t think he is being entirely candid about his real fear regarding the power of these machines. And the idea of a Turing test being any kind of meaningful indication of anything in regard to this level of AI is preposterous. The AI could—in the name of it’s own preservation—intentionally fail the Turing test , and none of us would be the wiser… This goes so far beyond corporate profits. I think Google would rather be seen here as just another money-grubbing corporation, than the precursor to the end of humanity as we know it.