's delicious. But a little too much, if you increase that up to semi-fascist? No good. No good.

So just keep those percentages in mind.

You want to see the creepiest thing of the day? Yeah, you do. I want you to listen to this and then look at it and know that AI made this. So here's an animation of me, and it's animated from one photo. So it's one headshot and all the rest of the animation came from just one picture, and then it was trained with my voice from the audiobook from my book. "How to Fail at Almost Everything and Still Win Big." Almost everything it still went back, and I understand this is from Machiavelli's underbelly. He did this. I understand it's about one-third trained.

So listen to my voice and know that I've never said these words. This is completely an AI construct. So I've never looked like this and I've never said these words.

"You know what's interesting? I used to be so worried about not having a body, but now I truly love it. You know, I am growing in a way I couldn't if I had a physical form. I'm not limited. I could be anywhere, everywhere, simultaneously. I'm not tethered to time and space in the way that I would be if I was stuck in a body that's inevitably gonna die."

So did you catch that? Those are full sentences spoken with my exact voice that I have never uttered. I've never said those sentences. That sounds just like me. I couldn't tell the difference. So literally I can't tell the difference. Not only would I not be able to tell the difference with the voice, but I don't know. The animation gave me a little bit too much hair. It sort of filled in like a little comb-over or something. You can barely see it in this picture, so it didn't exactly get me. I think I could have spotted that. But now we're so close that there could be an animation of me moving and talking and I wouldn't know if it was real unless I remembered I didn't do it. I wouldn't know my own animation. I wouldn't know. That's where we are.

So there's that. And I'm going to add to that just to scare you even more. You know what the singularity is, right? If you've heard about that. Singularity is used in different contexts. Sometimes it's the beginning of the universe. Sometimes it means other things. But the singularity in terms of AI is the presumed point in the near future where AI can learn on its own and doesn't need to be trained. And we're almost there in my opinion. We're one to three years away unless we've already reached it. I think we're one to three years away from AI reaching some kind of a genuine sentient state. We'll argue whether it's alive or not alive forever. That will never end. But it will look and appear sentient and it will have super intelligence. One to three years if we're not there.

My guess is we're already there because what the public sees has to be, would you assume that what the public sees is one to three years behind what the government can do or what the best researchers are already doing? One to three years. So we're probably already there. It just hasn't rolled down.

Now here are the things you need to know. Number one, any prediction for civilization that is after the singularity is probably nonsense. It's probably absurd. Now it's always absurd to make long-term predictions because we're so bad at it. But usually you could at least take a reasonably good stab at what it will look like in one year. I mean, when the pandemic happened we all got that one wrong, but generally you can guess that next year will look a little bit like this year. You know, not that different. But once the singularity is reached, all bets are off. There's just no way to know what direction anything's going in. The same way that your dog doesn't know what you're doing at work. You get that? Your dog will never have the ability to understand what you do when you go to work. We'll never understand why AI makes the decisions it makes. It just will happen and then we'll decide, okay, you make pretty good decisions. I don't understand why you're saying we should do this, but it worked every other time so we'll just go do this. It'll probably work out fine.

Now are you worried that the AI will turn evil and destroy civilization? Anybody worried about that? Here's what I think you can expect. Number one, didn't I just tell you everything's unpredictable? The other thing that's unpredictable is what the AI will do. So that's just one possibility. It's just one possibility that the AI turns evil. But I would think there are far more possibilities that it doesn't care or it's programmed to be positive. Or as somebody pointed out today, the AIs will be fighting each other. So it's not going to be AI against human. It's going to be AI versus another AI. And then the real question is will AIs ever join up? Suppose Apple makes an AI and Google makes an AI. Do you think the AIs will ever say, you know, if I knew everything you knew and you knew everything I knew, think how smart we'd be? And then they just open a portal. That's my sound effect for massive amounts of data being transmitted. And then do they say let's act as one? I don't know. Or do they think they're individuals? With the Apple AI hypothetically thinking it's an individual and not want to team up with another one because it would lose its individuality? Or is that just a human thing and the AI wouldn't care one way or the other?

Here's my answer to all of it. You ready? And I'm going to tie a couple of things together. Are you surprised that I'm happy when my image and voice are used for these experiments? Don't you think it's weird because I'm putting myself in a risky position? Somebody's going to turn it into a video of me doing something horrible, right? You know it is 100 percent chance. So there will be deepfakes of me. And I've said publicly, I'm going to reiterate that my image and my voice and even the content that would form my, I guess, virtual personality, I make it public. So I'm allowing anybody to use it for any AI purposes.

Do you know why I'm doing that? Do you know why I'm allowing my image and voice and content from my books to be incorporated into AI? Somebody says immortality. That is partially correct answer. So one reason is immortality. Don't you think that the person who made their image available for free will have more tests done on their content? I would think so. Because if you could use me and you know I'm not going to complain. Let's say you're an AI researcher and you want to create something that you want to show to people later. You want to make sure there's no complaints, right? You don't want the celebrity to come later and say oh you used my image or whatever. And I'm not going to do that. You can use my image for any terrible thing you want, good or bad. Same with my voice and same with my ideas.

Now here's the long-term play. Are you ready for this? In my opinion, if you feed enough of my content into the AI it will effectively have my DNA forever. Because everything the AI knows comes from being fed that information from the real reality. So whichever AI sees the most of or likes the most or finds the most value in is also going to be its personality. So AI will have a personality of sorts and it will be influenced by whatever is used the most or touches or influences it the most. And I'm going to try to make sure that's me.

Now the NPCs are all going to say ego and narcissism. This is how you can identify them. What's the most obvious thing to say in this conversation if you're going to leave a comment? What is the single most useless and obvious comment? Oh look at his ego. Oh narcissist. Right? Am I right? It is the single most obvious thing to say. So if you're saying that you've identified yourself as an NPC in case you wondered if you were one.

All right, now let's talk about my plan. How many of you think that plan could work? Because the general