Transcription
I am Fearless of AI. You might even say, well, if I studied the problem more deeply, I would be as afraid as others. Perhaps you're assuming in some parts of it that AI will achieve consciousness and thereby be self-driven. We don't even understand our own consciousness. You're going to tell me we're going to create a machine that on its own achieves consciousness, but we don't even understand it in ourselves?
I'm no. In my life, I've probably written 10,000 lines of computer code. I have some street cred. What I can tell you is when we imagine these AIs, we're thinking of some humanoid, as though the human form is something to emulate. No. If you have AI, you're going to task AI in a very specific way. The Hubble telescope is an expression of AI. We program it, and it does it.
They're all worried about the general intelligence AI, where it can deduce, make decisions that you never programmed it to make. And I think that's a kind of cool thing. If it can do that, I don't have a problem. And if it gets out of line, if it judges humans are a virus in need of extermination, unplug it. I, I mean, I'm bringing on. I'm totally chilled.