rain just went from oh you took care of yourself and now you can take care of your family easily. How about the tribe? How about other people?

So you can see the arc of my career followed that pattern. And I'm way less rich than those other guys. Like way less. Not even close. So I think it's instinct. You say it's ego. I don't think so. I think it's instinct. So that's my take.

So I'm gonna, so here's the full model and you compare it to yours. There's either these people who got super rich and still want power even though they have practically unlimited power for anything they want, right? Bill Gates can make pretty much anything happen right? Does he need more power? I don't know. I think it's people working on instinct. And in the Soros case I think he's poorly maybe poorly served by somebody down the line.

Now let me test you. Let me test you. Does my assumption, and this is just an assumption it's not an allegation just an assumption. It was speculation. I'll weaken it from assumption. Assumption is even higher than speculation. I'm going to take it all the way down to speculation. So all of our speculations are equal because we can't read his mind. We can only see what he does. And I would put my fortunes, I would bet them on there because there's corruption below him and that explains everything.

What do you think? Did I sell it? Disagree wholeheartedly. I see people say I see no way. I saw some people say yes. Now I would argue that those of you who have been so deeply in the Soros is evil camp are unable to hear what I'm saying. Those of you who didn't really have a strong opinion one way or the other probably heard my explanation and some of them said oh that looks better than the other explanation.

I think if you can't make the leap on this one you have to check your thinking. Just step back. I'm not even going to say you're wrong because that part we don't know right. So hear me carefully. I'm not saying you're wrong. Do you get that? Do you get really clearly that I'm not saying you're wrong? Whatever your interpretation is you could be totally right. Totally right. It just doesn't fit the facts as well as mine. That's all.

All right here's an interesting thing. You know I told you that AI especially with the help of Machiavelli's Underbelly on Twitter I've been testing whether AI can write a Dilbert comic strip. And so far it's shockingly close but not there.

I made the mistake, somebody's telling me that I have a strong opinion and that therefore I am biased. That is correct. That is correct. That's why I tell you to check my work the same as you should check your own. But here's what you should look for. What would be the source of my bias? If I had had a strong opinion one way or the other before this started then that would be a good speculation. But I think you can say that my bias from the start is I just didn't understand. Would you agree? Would you agree with my characterization of my own public statements of Soros that from the beginning I've said I'm just puzzled. I just don't understand.

If you approach a topic from I'm confused and then you form a point of view it's less likely to be confirmation bias. Do you get that? Confirmation bias is when you're just agreeing with what you already said. Oh all the evidence supports what I already said. That's not what I did. It was literally yesterday when I realized that corruption could explain all the elements. So I didn't have anything like a firm opinion and still don't. That's why I'm so adamant about saying that this is speculation. But so is yours. We're all on the same level of credibility which is none basically.

So back to humor. So AI did a good job of coming up with things that sounded funny and I cleverly but incorrectly suggested that if AI were fed the six dimensions of humor, a formula that I came up with years ago for what makes something funny. The formula is that if any two of six dimensions are used in a joke it's, it could be funny but not necessarily, but it's a minimum requirement. The minimum requirement is two of six dimensions.

If you haven't heard it the six dimensions are is it recognizable, you know something that happens in your life, is it naughty, cute, clever, bizarre, and there's another one I forget. But if you use two of the six you get it. So I said hey if you were to feed this formula into AI could then it beat me? Because I believed it had not been done. But it turns out that Machiavelli's Underbelly had done exactly that and fed my entire explanation of the six dimensions of humor into it. It still did not create a joke that I think was great. It was missing one thing.

But this is going to really blow your mind. There's only one thing that it couldn't do. It knew how to write. It knew what a joke was. It knew the formula for writing a joke. It had the right topic but it didn't quite hit. There's one thing I say all the time that it needs to have to write jokes that's missing. A/B testing.

Do you know how I know if something is funny? I'm A/B testing it in my own brain. Because there's still one thing I can do that AI can't do. Well that's a lie. It's one thing that AI hasn't done yet but could really easily do. That's the problem. The problem is it can't A/B test on its own because it doesn't have a sense of humor. But a human has a sense of humor. This developed in whatever weird way we develop.

So when I'm going through all the options I'm saying if he says this is funny, if he says this is a bit funny, if he says this I actually laugh. I A/B test instantly. If it makes me laugh that's the requirement. It has to make me laugh. The computer doesn't have that. It'll put on its best shot. It'll try one thing its best shot and then it's done. If I did that I couldn't write humor either. If all I did is put down my first draft you would never read my comic. It doesn't do the second drafts. It doesn't edit and it doesn't test it with the audience.

Now here's what's going to put me out of business. All you need is a Twitter account that the AI has access to that it can tweet. And then you say AI write me a joke about George Soros or critical race theory. And then the AI gives a first draft. Here's my first draft. And then it tweets it and says what do you think about this? And then humans some way instantly because jokes are easy to read might be three sentences. Within five minutes the AI has hundreds of responses. It only needs ten. Ten responses would be perfectly adequate to say if something is funny or not. Because people are similar enough that if three out of ten thought it was hilarious that would be a hit right? You know you can't get ten out of ten.

But if you got three out of ten people say that is just really funny you would have a best-selling book, number one comic. Three out of ten is huge for humor. So the only thing AI doesn't do to write jokes it could do tomorrow. It just had to know what I just told you. All it needs to do is add the one thing that I have that it doesn't have and it could do it better than I could do it. Because if I could test all of my jokes instantly on Twitter with other people I wouldn't use my own brain to do it. Because the limitation of my own brain is that I'm only getting the things I think are funny. And cartoonists become a little weird over time. We need edgier stuff to laugh because we deal with humor all the time. So you just need more.

And so sometimes there are things I don't think are funny that the audience would. And some of the actually some of the most popular Dilbert comics of all time I thought were some of my weaker efforts but the public thought it was my best work. So I can't tell but hey AI would know what the best one is because it would have data. It tested it.

So you add the first draft, A/B testing, second draft, A/B testing. You add that to the humor formula and AI already can do my job.

Did you see the pictures the AI drew of my comic strip? And you noticed that the heads of the characters were like misshapen. How hard would it be to teach the AI hey we see your first draft of these comics but we should remind you that on humans as well as comics the ears on both sides are symmetrical. The eyes are symmetrical. So you want symmetry. Whatever's on the left should look like the right. Believe it or not AI doesn't seem to know that yet in the context of creating you know Dilbert comics. But how hard would it be for AI to learn that the ear on the left needs to look like the ear on the right or it's not right? Pretty easy. Pretty easy.

All right that should scare you. You know a lot of us thought that art would be last. The art would be the last thing that the AI could do. And for probably 20 years I've been saying you are so wrong. Art is going to be the first thing. Art isn't going to be the last thing. It's going to be the first thing.

And the reason that I can say that is because of my talent stack. I do creative stuff for a living in a lot of different realms you know visual art and writing etc. Tweeting, speaking. So I have the creative experience down. But my background is more economics and rational business and you know even engineering type of thinking. So I bring sort of an engineering approach into the creative world.

And what I can see, what I can see is my own method. I can see my own system. I'm not sure other artists see it as clearly if they don't have sort of an engineering mindset going into it. And so what I can see is that all art is just a formula. Art is just a formula. And then you add the first draft and the A/B testing part of it and you're done.

I knew that from the jump. I knew that art would be first. And in my opinion the art that the AI is doing right now is better than human art. So let me call it. And by the way it doesn't matter what you say. It matters what artists say. This is one of those situations where it takes an artist to tell you when the AI is better. And I'm telling you that right now. I've seen a lot of AI art in the last month. It's better.

If you told me what do you want to put on your wall I would put the AI on my wall right now. It's not even close in my opinion. The AI art has already lapped human art and it will never return. It's over. You know people might buy human art because it came from humans and you know they're always going to see things they like but the AI art is just better. It's just better.

Calling it better is silly. No it's not. No it's not better. Better in this context means that it lights up the parts of the brain the art is supposed to light up and it does it more thoroughly than other art. That's pretty objective. You know I know art is you know it when you see it blah blah but that's just not true. It's just not true. What's true is good art is done well is very objectively done well and you could find the formula to it.

You know how many of you know that if you were to do a portrait there's a let's say a design element or several they pretty much have to use. If you go outside those design expectations you don't have art. So art has always been formula. It's just that the artist didn't necessarily know they were following one. And you thought it was maybe magic and art because you didn't know the formula either.

But if you do create art and you have some background and you know rational thinking you can see instantly that it's just craft. Art is mostly craft plus luck. There's some people who make things and it's just magic on the first draft and that's just luck. You know if you tried to make artistic things all day long and you had some skill at it you're going to nail it once in a while. It's just luck. You're not talking about the same art that you're talking about. I think I am but I don't know. There's a priceless plexiglas box called art. Well I think that's more about marketing than anything else.

Is luck really genius? Well let me give you one of the reframes that's in my *How to Fail* book. Well that's weird about luck. The old frame is that you can't control luck. The reframe, the one that I offer, is that you can totally control luck. You just have to go where luck exists. So go where there's more energy, more things happening, more connections, more networking, more jobs, faster growing industry. A growing industry not a shrinking industry. You just go where the action is and then you know put yourself into it and you're going to have all kinds of opportunities. But if you sit on your couch at home luck isn't going to find you at home.

So you would say that's not luck. No it is luck because you still need the accidental opportunity. You still need to notice when somebody said something. Somebody named their couch lucky. Yeah okay that's one way to play it.

Jeff says you do realize the engineer artist thing works both ways. Excellent point. The engineer science thing works both ways too. Yeah there is a hypothesis that I found stronger than I thought it would be that artists anticipate science and therefore engineering. Right? Do yo