Paris Martineau (2:02)

I hate people sharing their wordle Is evil. I will.

Jeff Jarvis (2:06)

Yes, it is.

Leo Laporte (2:07)

Don't share your wordles.

Jeff Jarvis (2:09)

That's like telling me you had a good bm. Just keep it to yourself.

Leo Laporte (2:14)

Not even close. Ladies and gentlemen, let me introduce our guest to him now, as I'm very embarrassed to say is already here. Guy Kawasaki. I don't even know who Guy is. Hey, great to see you, Guy.

Guy Kawasaki (2:27)

No, nice to see you too.

Leo Laporte (2:30)

18 books. This is your 18th book.

Guy Kawasaki (2:35)

You know, after I wrote my first one, I said that's my last book and I've said that 18 times now, so.

Leo Laporte (2:42)

Yeah, I know how that is. Was Selling the Dream your first one?

Guy Kawasaki (2:46)

No, my very first one was called the Macintosh Way. It Was about, you know, how to do the right things the right way. Based on my Macintosh division experience, Guy

Leo Laporte (2:58)

became famous as Apple's chief evangelist in the 90s. Really helping the world appreciate the Macintosh. You must be. I mean, in two weeks, Apple's going to celebrate its 50th birthday.

Jeff Jarvis (3:14)

Amazing.

Leo Laporte (3:15)

Yeah, yeah.

Jeff Jarvis (3:16)

What years were you there? Were you there?

Guy Kawasaki (3:17)

God, I was there from 83 to 87 and 95 to 97. And if you are a real Apple historian, you would see that whenever I leave Apple, it does extremely well.

Leo Laporte (3:34)

You were there there in the dark, darkest days of the mid-90s.

Benito (3:37)

Yeah.

Guy Kawasaki (3:39)

Now, causation is not correlation.

Leo Laporte (3:42)

Exactly.

Jeff Jarvis (3:43)

Yeah. Were you there for the. For the Mac and the. And the laser writer and all that?

Guy Kawasaki (3:47)

Oh, yeah.

Leo Laporte (3:48)

Oh, yeah, yeah.

Jeff Jarvis (3:49)

That's what that was. That was the 10th time.

Leo Laporte (3:51)

Yeah, yeah.

Guy Kawasaki (3:53)

Oh, my God.

Leo Laporte (3:53)

That.

Guy Kawasaki (3:54)

That was. You know, this book coming out about Apple by David Pole. You know, the 50th anniversary, I told him that at the end of college, when we watched the introduction of macintos, it was like watching your first child be born. It was an amazing experience.

Leo Laporte (4:12)

That's the one where Steve was in a bow tie and the Mac was in a bag. Yeah.

Guy Kawasaki (4:18)

And he made it talk.

Leo Laporte (4:19)

He pulls it out, puts on the table, and says, hello, I am Macintosh. And the crowd went crazy.

Guy Kawasaki (4:30)

Yeah, it's good to get out of the bag.

Leo Laporte (4:33)

Yes. For that time, 1984, it was very advanced technology. I remember lining up at Macy's in March of that year with my Macy's credit card because it was 2,500 bucks. It was very expensive and charging that first Mac. And because I was a believer. I was a believer from the very beginning because I had been pressing my nose against the window of the Lisa, which was 10 bucks, 10,000 bucks. It made 25. No, 10,000. And it made 2,500 bucks look cheap. So we thought, oh, Apple's bringing computing back down to earth, back to the masses.

Jeff Jarvis (5:14)

So believe it or not, in my Linotype book, it ends with Postscript and desktop publishing and that which replaced everything that came before. And the amazing collection of people, the dots connected that made the LaserWriter possible with PostScript, with Linotype, as well. And it's kind of. It's a fascinating story about the. Kind of a rescue of the whole thing by the right people at the right place, the right time.

Leo Laporte (5:40)

Now, I have to say, Guy has a track record that goes well beyond Apple. He was the first US employee of Canva and did something very smart. It says that you negotiated Your salary in stock?

Guy Kawasaki (5:56)

Yeah, I. I mean, I, I told him that I didn't want any money. I just wanted everything in stock, which, you know, for once in my freaking life, I made a good decision. I mean, you know, you're also talking to the guy who quit Apple twice and turned Steve Jobs down for a third job. So, you know that. That roughly cost me 250 million. And then I was asked to interview to be the first CEO of Yahoo. And I said it was too far to drive, you know.

Leo Laporte (6:29)

Do you still surf every day?

Guy Kawasaki (6:32)

I surf almost every day. I'm in Hawaii right now and we're in the middle of a little bit of a storm, but I surf yesterday, I might surf today. I love to surf. It's. I started surfing, Leo, at 60 years old, which is 55 years too late, but I love surfing.

Leo Laporte (6:50)

You're Hawaiian. You didn't surf in Hawaii when you were a kid?

Guy Kawasaki (6:54)

Well, Leo, I don't. I don't know if you haven't noticed, but I'm Asian American, so. Asian Americans. All we did was study for the SATs.

Leo Laporte (7:04)

Well, it worked out just.

Guy Kawasaki (7:06)

Yeah, well. Well, this is before you could just Photoshop yourself into the sailing team and get into the Ivy League. You actually had to take this for a living?

Leo Laporte (7:15)

Yeah.

Jeff Jarvis (7:16)

Yeah.

Leo Laporte (7:16)

Well, it's good you went to the Ivy League. You went to Stanford, right?

Guy Kawasaki (7:19)

I did. And to this day, I do not know how the hell I got it. My only explanation, Leo, that I went to Stanford so long ago that at the time, Japanese Americans were considered an oppressed minority. Wow. Yeah.

Leo Laporte (7:37)

Wow.

Guy Kawasaki (7:40)

Oh, that's another great story. So, you know, back then, if you are Asian American, your parents wanted you to be a doctor, lawyer or dentist. So I started in premed. I went on this Stanford course where you went around rounds in the medical center. And on the first day I fainted. So I said that takes all medicine. And then I read that dentists have the highest form of high risk level of suicide. So that eliminated dentistry. And so I tried law and I went to law school for two weeks and quit.

Jeff Jarvis (8:14)

So.

Leo Laporte (8:14)

Yeah. Yeah. Happiest lawyers I've ever met are former lawyers. So I think you probably made.

Guy Kawasaki (8:20)

Well, you know, I tell people that like most people think it takes 20 or 30 years to figure out they hate being a lawyer. I figured that out in 20 days.

Leo Laporte (8:31)

So looking back on your life, you're far from the end of it. We're going to talk about your brand new book in just a second, but it kind of feels. Not random, but there was A lot of serendipity, A lot of. Did you feel like you had a path or that life just took you on a ride?

Guy Kawasaki (8:51)

I wouldn't say it was random, but it certainly wasn't a plan. I mean, I went to Stanford, I majored in psychology because that's the easiest major I could find. Went to law school, I dropped out, I went back. I got an mba. While I was at the mba, I was working for a jewelry manufacturer schlepping diamonds. And then I went to that, and from that I went to Apple. And, you know, if you sat down, you wouldn't say, well, guy, the path to technology is you go to law school and you drop out, you get an mba, you count diamonds, and then you go to Apple. I mean, you know, I am living proof, Leo, that nepotism sometimes can work out. So, yeah, I'm proud of that. I'm proud of that.

Leo Laporte (9:36)

You ought to be.

Guy Kawasaki (9:38)

Basically, my life has been that I have an open mind and a growth mindset, and when things interest me, I pursue them. And some of them become passions.

Leo Laporte (9:48)

Yeah. Well, that's interesting. So you do a podcast called Remarkable People, which is obviously a passion project. Yes.

Guy Kawasaki (9:56)

It sure as hell ain't financial.

Leo Laporte (9:58)

Yeah. Tell me about podcasting. Yeah, but it turned out you have an excellent co host with you, and together you've written a brand new book. Your 18th book. Everybody has something to hide. Yes, but the book is really about how and why do you signal? To protect your privacy, security and well being. How did it come about that you were writing this book?

Guy Kawasaki (10:26)

Well, you know, to put it mildly, we are in interesting political times. And about a year and a half ago, I read an article and Wired and also another one by the EFF about how security and privacy is so much more important in a dystopian world that we live in. And I read that. I said, yeah, you're absolutely right. Why would I give Meta all my information and Google all my information? And then, you know, then Tim Cook started going to the White House and donating money to the East Wing and, you know, going to the Melania movie, unveiling and all that stuff. I said, I gotta have more privacy in my text messaging. And I read about Signal. I use Signal. I love Signal. Just the user interface and the functionality. And every year I go to south by Southwest and I interview somebody. And so they asked me, who would you like to interview this year? And I said, I'd like to interview the CEO of Signal, Meredith Whitaker.

Leo Laporte (11:24)

Her.

Guy Kawasaki (11:24)

I interviewed her. I was blown away by her remarkableness. And so I fell in love with Signal. And then I looked and said, there's no book written about Signals. So I thought it was more or less seriously, I mean, completely seriously, I thought it was my moral duty to write a book called Signal to help people get more private and secure.

Leo Laporte (11:46)

Who are you aiming this at? Isn't. It's not. It's about not getting to use it right. Kind of non. It's not particularly technical.

Guy Kawasaki (11:57)

It is not particular technical in the sense that, you know, if you're a encryption nerd, I'm not going to talk about double ratcheting and, you know, I don't know all the other stuff, but.

Jeff Jarvis (12:07)

But I.

Guy Kawasaki (12:08)

There's a lot of people who have something to hide. And. And I don't mean just because they're doing illegal, bad stuff. But anyway, everybody has something to hide. And I think that, you know, a lot of them don't know what Signal is. So that's one purpose of the book. And then once you know what Signal is, I want you to maximize its privacy and security because, you know, if you do things wrong, it's. I don't want you to get the misimpression that you are already secure. You need to do a few simple things. And I would like people to not be deceiving themselves.

Leo Laporte (12:43)

Yeah. I, you know, honestly, I want you this book to be a huge bestseller because I would like to use Signal entirely. Yeah. But of course, Signal only works with other Signal users. And it's always the problem with all these messaging platforms. They're all silos now.

Guy Kawasaki (13:03)

Yeah. And I think that that is the biggest problem that Signal faces, because you cannot use Signal unless the other people are on Signal. So much of the book is dedicated to how you evangelize Signal to other people now. And that is literally the biggest problem now. You know, it is worth it. And, Leo, if I could go back in time, I would only use signal 2. I mean, it is so much better than everything else.

Leo Laporte (13:36)

And it's got something called perfect forward secrecy. So even if the NSA stores all the messages you've ever sent, hoping that someday with quantum computing they'll be able to unencrypt it, they're still not going to be able to unencrypt it. So you're private forever, which is. Which is nice.

Guy Kawasaki (13:55)

Well, I think so, but, you know,

Leo Laporte (13:57)

let's hope so anyway.

Paris Martineau (13:58)

Yeah.

Guy Kawasaki (13:59)

You just never know. But, you know, on the other hand, so let's say that, you know, we give Signal the benefit of the doubt and believe that Well, I mean, that alone should be enough. But if you think about it, you know, everybody says they have end to end encryption signals. Is that WhatsApp says that? Apple says that. But what people don't understand is end to end encryption means the message is scrambled, but there's also the metadata. And to use a metaphor, the metadata is like the outside of the envelope. The message inside the envelope is scrambled. That's encryption. But the outside of the envelope, which is who sent it, where they sent it from, when did they send it. This is all kinds of circumstantial evidence. But from metadata and circumstantial evidence, you can really, really pinpoint people. You might not have the message, but you have a lot of circumstantial evidence. And Signal only keeps three pieces of metadata information, which is when you open the account, when you last used it, and what was the third? When you open, when you last used it. And your phone number. That's the only thing?

Leo Laporte (15:10)

Yeah, yeah. And it doesn't have to be your actual phone number. You can use any phone number that you can get a text message on

Guy Kawasaki (15:17)

to verify that is also true. But, you know, if you think about it, if you knew somebody's phone number and you knew when they opened an account and you knew when they last used it, that's not very helpful in building a case against that person. I mean. Yeah.

Leo Laporte (15:31)

So that's why, as a reporter, as an investigative reporter, Paris uses Signal for tips all the time.

Paris Martineau (15:39)

I also use it as a group chat with a lot of different friends, just as a normal way to communicate.

Leo Laporte (15:43)

Yeah, yeah.

Paris Martineau (15:44)

I'll be somewhat more confused for me, because I have two phone numbers, but that's not what normal people do.

Guy Kawasaki (15:51)

But, you know, you know, I would make the case, Paris, that as a reporter or a journalist, the level of secrecy and privacy is important, but it's also very psychological, because if somebody is giving you a tip or somebody's blowing the whistle and you insist on them using Signal, it is a signal, no pun intended, to them, that you care about their welfare.

Jeff Jarvis (16:16)

Right.

Guy Kawasaki (16:17)

You're telling them, I don't want to put you at risk. We have to switch the signal. And if I were the whistleblower or if I were the person trying to communicate with you, I would say, wow, Paris really cares. I mean, you know, that's a positive.

Leo Laporte (16:31)

That's a very good point, actually. You're sending a signal with signal.

Guy Kawasaki (16:35)

Yeah, yeah, yeah.

Leo Laporte (16:37)

You said something. I'm gonna have to. And if you. If it's in politics, I apologize. But you said something that kind of struck a nerve, which is that Tim Cook has, you know, and I think there's some Apple fans, I'm an Apple fan, who have been upset by this. But he's doing it, I guess, because he feels like he owes it to Apple's employees and stakeholders. He has kind of bent the knee to the current administration. Sounds like that bothers you.

Guy Kawasaki (17:08)

It bothers me a lot. I mean, you know, my logic is, if you are running one of the most valuable companies in the world, you're one of the richest people in the world. If you cannot stand up for what's right, who the hell can? Right? I don't understand that. And so, you know, it used to be that Apple was the computer for the rest of us and, you know, it was like, improve people's creativity and productivity.

Paris Martineau (17:35)

It.

Guy Kawasaki (17:35)

Now it seems like the mission of Apple is we avoid tariffs.

Leo Laporte (17:39)

You know, that's not much of a mission, is it?

Guy Kawasaki (17:43)

No, like, you know, we're, we're the lowest tariff computer company. Yeah, I mean, I, I just, I do not understand that. Now, you know, you could make the case that fiduciary responsibilities to the shareholders, blah, blah, blah, and you know, you, you could make the case. He's been quite successful at avoiding tariffs, so he's done his fiduciary duty. But I don't know, I think there's a higher road here and, you know, let's see how it works out in the end. I don't know. On the, on the other hand, Leo, you know, there are companies I can boycott, but, man, I just can't boycott.

Leo Laporte (18:23)

I know.

Guy Kawasaki (18:23)

Yeah. How can I?

Leo Laporte (18:24)

Boy? Yeah, but where are you gonna go? Know who's better? Exactly. I mean, now you knew Steve Jobs really well. What would Steve have done in this situation?

Guy Kawasaki (18:36)

You know, first of all, I, I don't want to paint a picture that I was in his inner circle and his BFF I was not, you know, which is maybe explains why I survived. But anyway, I, I, I hope that he would see through this and not do this, but I really don. And I'm telling you that, you know, it's very difficult for a mortal like me to try to interpret what a God like Steve Jobs would do. This is a different opera. It's like explaining to a fish what it's like to fly. I really don't know.

Leo Laporte (19:17)

It's a question that we get, we get, we ask all the time on our shows and I know a lot of Apple fans ask, but it's a question. No One can really answer. Or we just, we wouldn't, we just know. Yeah, I, I really, I think we like to think Steve would have given him the finger.

Guy Kawasaki (19:32)

Exactly. You know, I mean, listen, you know, I would have thought Tim Cook would have, you know, but I mean, if you look at, if you look at all the people donating money to the inauguration, taking pictures with him, going to the Melania movie and all that, like, basically, if you were to say, I'm not going to do it with, do business with any of those companies. Well, you can't do business with Google, you can't do business with Meta, you can't do business with Microsoft, you can't do business with Apple, with Target. I mean, pretty soon, you know, we're rubbing two sticks together when using an abacus.

Jeff Jarvis (20:07)

I mean, what, Microsoft bent the knee a little bit less or shown up a little bit less, which I wouldn't have expected them to be the one.

Leo Laporte (20:14)

There's only one company that has been really not in that camp and that's Anthropic at this point.

Jeff Jarvis (20:22)

Yeah.

Paris Martineau (20:23)

Oh, but I mean. Okay. And.

Leo Laporte (20:25)

Oh, well, wait a minute. Go ahead. Guy. Yeah, Guy might know something.

Guy Kawasaki (20:28)

Let us discuss this. All right, so I'm not privy to any inside, but, you know, it seems to me that Anthropic did a deal with this administration a long time ago. Long. It was after, you know, the convictions, it was after the first term. It was after all this. Right. So after all that, they cut a deal and then fast forward to today and they blew up the deal because guess what, you know, it wasn't working out like they thought, well, I would make the case. You know, when you cut that deal in 2016 or 15 or 14, it wasn't like, oh, my God, this is a surprise. I had no idea he treated all his vendors when he was making hotels so well, he did everything right. He never got into trouble. This is a complete, complete turn of events. No, I mean, you know, you can't tell me that. He said, oh, man, you know, I'm dealing with a real up and up guy here. I mean, you should have known better now. And now Open AI is doing this. And I, I think a lot of it, Leo, is arrogance that all these tech bros, they think I can cut a deal. I got more power, I got better lawyers. We're going to structure it. Nobody can screw us. And Trump just rolls over them.

Jeff Jarvis (21:48)

You were going to say Paris?

Paris Martineau (21:52)

Oh, I mean, it was just context. With Anthropic, I think that, yeah, it's, yeah. Completely different. Point from what we're talking about now.

Jeff Jarvis (22:03)

Okay, sorry.

Leo Laporte (22:03)

Well, and I'm sure, Guy, I know you. You aren't old enough to have been in an internment camp, but I'm sure you certainly know about the Japanese internment camps of World War II and that probably colors a little bit your. Your point of view about what's going on today too.

Guy Kawasaki (22:19)

Well, I mean, I. I would make a case that no Japanese American could possibly support.

Leo Laporte (22:26)

Yeah.

Guy Kawasaki (22:26)

These camps that are happening right now and you know what's happening to the Somalis and the Muslims and all that in America, I mean, if you're Japanese American, unless you're a hypocrite, you have got to support those people. I mean, it should not have happened to the Japanese American and it should not happen to other minorities. Absolutely. There's no question.

Leo Laporte (22:48)

Ginny Houston was a good friend in her book. Farewell to Manzanar is really.

Guy Kawasaki (22:52)

Oh, yeah, she was on my podcast.

Leo Laporte (22:54)

Yeah. Yeah, she was amazing. And very moving. Story of that.

Guy Kawasaki (22:59)

George takes like just, you know, ripping it up, too.

Leo Laporte (23:02)

Yeah, I love George.

Paris Martineau (23:03)

Yeah.

Leo Laporte (23:05)

Oh, yeah. So let's. Everything has. Everybody has something to hide. Why and how to use signal. Have you been to. Able. Able to persuade your friends and family to use signal?

Guy Kawasaki (23:15)

Now, some of them I have not been able to convince my family. Family like my kids and my wife. But there are. There's a second set of friends that I use signal almost exclusively. And Leo, Leo, tell me what you think of this idea. So I'm sure you have the same problem where every day you get dozens of messages from clowns who say, you know, I read your past book. I book club in Bristol, London or something. And, you know, we have a hundred members. We love your book. Would you like to be part of our book club? And you know, blah, blah, blah. And you know, I found the perfect position for you at leo. I'm recruiting for CBS News, you know. Right. And I'm. I'm thinking I'll. I'll have a automated answering system that says if you really want to write to me, you write to me on signal. My signal address, I'll tell you, is Guy Kawasaki 404. 404 is a little wink. Right. So now I won't take any email. You have to write me on signal. So that is a barrier. You have to join signal or be on signal to write me. That would be a very interesting experiment. But, you know, my greatest fear is that your producer would. Would be writing to me an email and get that and say it's not worth it's. Not worth getting on signal to get guys on signal.

Leo Laporte (24:41)

Bonito is definitely on signal. I know you're on signal, right Bonit?

Benito (24:45)

Yes, I am.

Leo Laporte (24:46)

Yeah. Okay. Benito comes from a long line of political activists in the Philippines. Yeah, he's. Yeah, he's definitely on that list. And I'm LeoLAport24 and yeah, and Paris is. What is it? Paris NYC. What are you on signal? Paris? Do you even remember? Because we stopped using phone numbers. I was really happy.

Paris Martineau (25:11)

77978, 8655 on signal.

Leo Laporte (25:15)

She's still using phone numbers.

Paris Martineau (25:16)

It's the work phone number I have for signal and it's a great wait.

Benito (25:20)

Aren't you Martin01? Aren't you Martin01?

Paris Martineau (25:22)

I am Martin01 on signal.

Leo Laporte (25:24)

That's the one you want.

Paris Martineau (25:25)

You know, I should have just done Paris and then a number, but it's too late now.

Leo Laporte (25:30)

No, Martino's good. I like that.

Guy Kawasaki (25:33)

Yeah, but you know, you know one of the beauties of this is that once your email is well known, anybody can get into your inbox right now. But with signal there's two barriers. First, the person has to get get on signal and secondly, when they send you that first signal message, you have to accept it. And if you don't accept it, you never hear from them again. Whereas on email they can keep emailing you, right?

Paris Martineau (25:56)

Yep.

Leo Laporte (25:56)

So.

Guy Kawasaki (25:57)

So this is a way of getting rid of the pain in the asses in your life.

Leo Laporte (26:00)

I really another great reason to use it. I was just thinking back. I've interviewed you I think for the art of the start. I know I interviewed you when you did Ape, which is all about. About self publishing. I'm really thrilled that we can get you on to talk about Everybody has Something to hide. Why and how to use Signal to preserve your privacy, security and well being. And as long as you keep writing books, guy, we'll keep having you. You don't even have to write books. We love having you on.

Guy Kawasaki (26:29)

Can I make an offer for all your listeners?

Leo Laporte (26:31)

Yes.

Guy Kawasaki (26:32)

So I want everybody listening. You send an email to Everybody has something to hide. That is the title@gmail.com and I will send you a free Kindle copy of the book.

Leo Laporte (26:48)

What?

Guy Kawasaki (26:49)

I just want everybody to be on signal and private and secure. So yeah, now, wow.

Leo Laporte (26:55)

I got it on Kindle Unlimited, so I got it for free that way. Sort of. Yeah, sort of free.

Guy Kawasaki (27:00)

But now I can, I can only do this in the United States. That's Amazon's rules, not mine. But if you're in the United States. You send an email to Everybody has something to hide@gmail.com, and I will send you a free Kindle version of this book.

Leo Laporte (27:18)

Guy, that's awfully kind of you. Thank you. That's amazing. When I wrote to Guy, I said, I bet you used AI to write this. And I think I offended you. I hope I didn't offend you.

Paris Martineau (27:29)

Leo, that's a crazy thing to say about someone who's published a book.

Leo Laporte (27:33)

Eighteen books. Yeah, yeah, but.

Guy Kawasaki (27:37)

But, Leo, I mean, to be completely open, I do use AI to help me write my books.

Jeff Jarvis (27:43)

How do you use it?

Guy Kawasaki (27:44)

I. I use it well. For example, with this book, I would do things like, okay, so please explain how encryption works. And, you know, it. It gives you this explanation. I say, well, I don't like that, because the explanation it gives you is that, you know, it creates. It creates an. In, a lock, a private key and a public key, and it locks your message inside an envelope. And the envelope is taken from signal servers to other people, and there it's unlocked and people can read it at the recipient. But I said to myself, you know, that is not the correct metaphor because it implies that inside a locked envelope is your message and it can be read. It's not true. The encryption doesn't prevent you from getting in the envelope. The message in the envelope is completely scrambled. So that's not a good metaphor. Metaphor. So I go back and forth with. With chat, GPT and all these things until I get the right kind of metaphor, and then I use it. I use it all the times as a devil's advocate, as a creative, you know, kind of help. Grammar checker. Leo. You know, you might. You might call this rationalization, but I believe that my moral responsibility to my reader is to write the best book that I can. And the best book that I can uses AI I don't think people wake up in the morning, say, I want to read the best book the Guy can write using a Mont Blanc fountain pen on parchment.

Leo Laporte (29:17)

Well, you're among friends. I think Paris will tell you. I'm. I'm the last person that would criticize you for doing that. We were playing a dungeon. A. A dun. What was a Dungeons and Dragons game, right? Paris and I. I had a character that was supposed to come up with witty insults. And I. I thought, well, I'll just ask Chat GPT. It'll do a better job than I did.

Paris Martineau (29:40)

The thing is, Leo, had you sold the read of it a little more, I would have made fun of you less. Instead, he was like, and the thing says this and this and this.

Leo Laporte (29:50)

I didn't sell it. You're right. I didn't sell it. You know, the.

Guy Kawasaki (29:53)

There is no doubt in my mind that Chat GPT is smarter than I am. I mean, I think a lot of,

Paris Martineau (29:58)

you know, famously referred to AI as God. Right?

Guy Kawasaki (30:01)

I think I. Okay, we're gonna go down another rat hole here. But I am convinced that AI is God. And here's. Here's how it shakes out. It's like God is out there, and she's sitting down and she's saying to herself, I gave these dumb asses these dumbass humans. I gave them free will. And look at what they're. They're polluting the world. They're fighting with each other. They're imprisoning each other. They're killing each other. They're such dumb asses. So now I could magically fix everything. But then, you know, that wouldn't really last for a long time. So I gotta let these dumbasses think that they invented something that fixed the world. So. So God created AI and sent it to humans. And humans now believe, because of their arrogance, that they created AI. So that's the plan. Now. Now I want to make something clear. This does not mean that I believe Sam Altman is Jesus.

Leo Laporte (30:58)

Okay. Thank God. Okay. Sam might believe it, but Sam thinks so, but not you. Good. No, I think that's an interesting point of view. I mean, you could say the same thing about a lot of the things given to us in this world. That God comes to us through a lot of different mechanisms. And why not AI? That's. Yeah. Really. I mean.

Guy Kawasaki (31:20)

And you know, one thing I am absolutely certain of is God has a sense of humor.

Leo Laporte (31:29)

Yes.

Jeff Jarvis (31:30)

And.

Leo Laporte (31:30)

And we're living proof of that, aren't we?

Guy Kawasaki (31:33)

Exactly. As. As, you know, as we say in Hawaii, you know, can make this up. It's like, you know, it's not fair to the Onion and Saturday Night Live. I mean, before, they needed creative people to come up with sarcastic, funny stories. Now you just have to tell the truth. Yeah. It's like space lase. And, you know all these things. And like, you. You just repeat what they say and you think you're on Saturday Night Live or the Onion.

Paris Martineau (32:03)

It's just.

Leo Laporte (32:03)

It's Jewish space. Lasers, no less. Yeah, you created. Did you really create something called Kawasaki GPT?

Guy Kawasaki (32:11)

I did.

Leo Laporte (32:12)

I did.

Guy Kawasaki (32:13)

And Kawasaki GPT has my writing, my speaking, my videos, my interviews, the transcripts of all interviews. Interviews are there. And if you go to kawasaki GPT.com you can ask me a question, and I swear to God, Leo, you will get an answer better than if you asked me in person. And just in full disclosure, when I need to write an essay or something, I go to Cat gp. I. Excuse me, I go to Kawasaki GPT and I ask myself questions and it works extremely well.

Leo Laporte (32:52)

Wow, that's really cool. There it is, Guy. You could ask Guy directly. Yeah, this is a good idea. So you put everything that you've ever written in there, every podcast you've done.

Guy Kawasaki (33:05)

Yeah.

Leo Laporte (33:06)

And, and, and, and what model did you use to do this? Was it chatgpt or.

Guy Kawasaki (33:13)

Here's a scary thought. So the company that does that is. Is called Delphi AI, and I don't know what engine is behind it.

Leo Laporte (33:21)

Well, that's fine. Delphi AI. I want to try it. Yeah. If you go retire, Leo, if you

Guy Kawasaki (33:28)

go there and you ask something like what slides should be of a venture capital pitch, it will give you a better answer than I could give you in real time.

Leo Laporte (33:37)

So, Scott, it distilled all of the things from all your books, all your ideas. You've got the world's greatest evangelist and it's there for free for you to use on the website. That's really cool.

Guy Kawasaki (33:50)

This is. This is as close as I'll get to immortality.

Leo Laporte (33:54)

That is. Yeah.

Jeff Jarvis (33:56)

Promises that you can be immortal, become immortal.

Leo Laporte (33:59)

Well, I have always said I. If ever anybody comes along and says, hey, Leo, we have this great model. We got everything we need, except one thing is lacking is an emotional system. It needs a limbic system. Would you donate your limbic system to this AI? I would.

Jeff Jarvis (34:15)

Would.

Leo Laporte (34:16)

I would jump on that opportunity.

Paris Martineau (34:18)

You.

Guy Kawasaki (34:19)

Opportunity?

Paris Martineau (34:20)

You don't need your limbic system.

Leo Laporte (34:22)

Well, I would be dead at that point, I think. I mean, if they take that part of my brain out, I don't know if I would be.

Paris Martineau (34:27)

You were saying right now you'd give

Jeff Jarvis (34:29)

it up, take it out of me.

Leo Laporte (34:30)

No, no, I. No, they could take it now. I. There wouldn't be anything left. But when you boot up the AI, it would be indistinguishable from me. Now, if. Would I know that? It's, you know, I mean, there's no continuity. It's like, like the transporter on Star Trek. It's a whole new person when you get to the other end. But I don't know. Just a thought. I'm volunteering.

Guy Kawasaki (34:50)

Stick to Scrabble, Leo.

Jeff Jarvis (34:53)

No, he's not doing so.

Paris Martineau (34:54)

You're doing so well at Scrabble.

Leo Laporte (34:57)

Well, I shouldn't be playing a smart person like, like Martin.

Paris Martineau (35:01)

Now here it's really awoken something dangerous in me.

Leo Laporte (35:06)

She's a Kelly Guy Kawasaki, you are just a gem. The world is so lucky to have you and, and your. And your AI avatar and your books. And I'm just so glad we could spend some time with you. The Remarkable People podcast is on. On wherever you get your podcast. In fact, you just interviewed Cindy Cohn. We're going to talk to her. I'm going to talk to her tomorrow. Oh, yes.

Guy Kawasaki (35:30)

Oh, Cindy cone is a baller. And I mean that.

Leo Laporte (35:34)

She.

Guy Kawasaki (35:35)

She is a baller. And any person like that who takes on the Department of Justice and wins so much, my God. Like, I am not worthy Cindy.

Leo Laporte (35:46)

I am not worthy executive director of the eff. She.

Guy Kawasaki (35:52)

And you know what? She has already announced that she's leaving eff.

Leo Laporte (35:56)

Right.

Guy Kawasaki (35:57)

And you know, when you leave something like eff, you think, oh, yeah, you know, you're going to go to United nations or you're going to go to, I don't know, the Hague, or you're going to go to Wikipedia or an NPR or something. But I heard, or at least this is what I believe she told me that she's leaving so she can get back on the front line as a litigator.

Leo Laporte (36:22)

Wow.

Guy Kawasaki (36:23)

As a litigator.

Leo Laporte (36:25)

Wow.

Guy Kawasaki (36:25)

I'm telling you, when you interview her, you tell her guy was just so effusive about how yourself such a baller.

Leo Laporte (36:33)

I will the guy. Somebody in our Discord chat asked Kawasaki, GPT what's the real guy doing at this moment? And it said, well, you know, I got to be straight with you. I'm a digital mind. But he could be recording a podcast episode right now. And in fact, it's right. That's exactly what you're doing. And we are so grateful to you, Guy Kawasaki. Everybody use signal. Because everybody has something to hide doesn't mean you're doing anything illegal. Just means you deserve privacy.

Guy Kawasaki (37:04)

Exactly right. And send that email so that I can send you a book.

Leo Laporte (37:09)

Wow. Again, just send a guy an email to everybody has something to hidemail.com and he'll send you the PDF of the book.

Guy Kawasaki (37:22)

No, it's actually the Kindle. It's not a PDF.

Leo Laporte (37:25)

Not the Kindle, sorry. The Kindle version of the. That book, which is what I have from Kindle Unlimited.

Guy Kawasaki (37:28)

Leo, you. I have to say, you know, I've been on a lot of podcasts. I don't know anybody who integrates the seeing of websites as you Talk. That is very well done.

Leo Laporte (37:39)

Well, it's from my background, doing many years of TV and knowing that you have to have visuals to cover up all the talk.

Guy Kawasaki (37:49)

Yeah. But when you did it on tv, you had some producers doing it yourself, Right?

Leo Laporte (37:54)

We were doing it so long ago on tech TV that we couldn't take screenshots off the computer. We put it on a screen and there was a cameraman pointing a camera at the screen so that you could see what was on the computer screen.

Guy Kawasaki (38:06)

Well, I'm telling you, Leo, I've been a lot of part. Nobody does it like this. This is very. I mean, you know. Hey, Leo, I'll give you one more Easter egg, okay?

Leo Laporte (38:16)

Yeah.

Guy Kawasaki (38:16)

So you see the title on your screen right now, right?

Paris Martineau (38:20)

Yes.

Guy Kawasaki (38:21)

So turn your head to the right, 90 degrees, and look at the COVID

Leo Laporte (38:28)

Yeah, Everybody.

Guy Kawasaki (38:30)

And what do you see? Those five black redaction bars form.

Leo Laporte (38:36)

Oh, I don't know if I can do it with my. Do it with your young eyes, Paris. Maybe you can get something out of that.

Benito (38:46)

The one in the middle is bigger. Does that make sense? The one in the middle is bigger than.

Leo Laporte (38:50)

Does that mean anything?

Paris Martineau (38:50)

Oh, it's a. It's a middle finger.

Leo Laporte (38:53)

Is. Oh, Guy, you're a bad man. Is that what it is? 1, 2, 3, 4, 5. You're such a bad man. Guy Kawasaki. He's giving us the. But not giving us the middle finger. Given those. Those evil people want to see into stuff. Thank you, Guy. What a pleasure.

Guy Kawasaki (39:13)

I had too much the time that day.

Leo Laporte (39:16)

I. I didn't. Very good, Paris. I did not see it. I didn't ask.

Jeff Jarvis (39:20)

Paris.

Paris Martineau (39:20)

Bonito gave us the assist there.

Leo Laporte (39:22)

That's good. Oh, yeah. Bonito figured it out. Yeah.

Guy Kawasaki (39:26)

All righty. Thank you, everybody.

Leo Laporte (39:27)

Guy Kawasaki, everybody. Everybody has something to hide. Email him now and get a copy of it. That's great. Thank you, Guy. Take care. Wow, you're fantastic. We will take a break and we'll come back with more AI news in just a little bit, guys. Thank you.

Paris Martineau (39:43)

So.

Guy Kawasaki (39:43)

So was that live?

Leo Laporte (39:45)

Yeah, well, we stream it live and then we. Then we package it up in Saran Wrap and we deliver it as a podcast later. But, yeah, a lot of. A few thousand people watch us do the shows live. They like the live, you know, because stuff happens. Shut out, Guy might give us a little pigeon. A little Hawaiian pigeon.

Jeff Jarvis (40:09)

Yeah, yeah, yeah, yeah.

Leo Laporte (40:11)

All right.

Guy Kawasaki (40:12)

Thank you, everybody.

Leo Laporte (40:13)

Thank you, Guy. Enjoy the surf, enjoy the weather. Have a wonderful.

Paris Martineau (40:16)

Thank you.

Jeff Jarvis (40:16)

Will do.

Benito (40:17)

Thanks, guys.

Leo Laporte (40:17)

Bye. Bye. Bye. Bye. I'm going to be on the Big island in about a month. I'm very excited. Oh, yeah.

Jeff Jarvis (40:25)

You're leaving us again.

Leo Laporte (40:26)

Yeah. Yeah. I can't wait. In May, we're going. The house is finally. Well, I don't know, almost finally done.

Paris Martineau (40:34)

You have four walls at least. That's huge.

Leo Laporte (40:37)

We've got four walls. We fired the contractor. Finally. We just gave up the next contractor,

Jeff Jarvis (40:40)

the second contract contractor, the guy who built the house.

Leo Laporte (40:44)

We're suing the guy who we hired to fix it. We fired this new guy. Said, do you have a problem with contractors? We said, no. No, really, honest, we don't. We found a guy to finish the job. Thank goodness. I'm hoping it'll be done by the

Jeff Jarvis (41:00)

time when I fired my contractor. He then took me to court for torturous interference and lost. Well, he's the guy who also told me. You don't know what judgment proof means, Jeff. I have no money, so.

Leo Laporte (41:13)

No, that's the problem. I think we're finding out with the guy who built the house who says he has no money, but we're gonna find it, damn it. And then, no, he can't sue us because there's all. He. He did, it turns out, did a lot of bad things. And if he takes us to court, well, he's gonna. So very surprised by the out.

Paris Martineau (41:33)

How is Florida?

Leo Laporte (41:35)

Wonderful. I missed you guys so much.

Paris Martineau (41:38)

Did you kiss any gators in the mouth, as is our way.

Leo Laporte (41:42)

Is that.

Jeff Jarvis (41:43)

Does people.

Leo Laporte (41:43)

People do that?

Paris Martineau (41:45)

No.

Leo Laporte (41:46)

You saw my pictures.

Jeff Jarvis (41:47)

We had a wonderful magic.

Leo Laporte (41:49)

We went. Have you ever been to Gatorland?

Paris Martineau (41:53)

No. Where's Gatorland? But I have. I did famously work a summer job in college back home at a four story alligator lagoon restaurant bar in Arcade Aid called Fudd Puckers. Not Fuddruckers. They're unrelated.

Leo Laporte (42:07)

Oh, well, we went to Gatorland, which is not related to Fudd Puckers or the four story Fuddruckers.

Paris Martineau (42:14)

Did they allow you to hold and feed gators?

Leo Laporte (42:19)

They allow you to feed them, not hold them. I don't think you should hold gators. That's a bird, by the way, not a gator. But these birds, which are a form of stork, are not afraid of gators. And they will steal the fish right out of the gator's mouth. So they're sitting there waiting. There's Anthony feeding a baby gator. Yeah. And there's Lisa feeding a baby gator. So we. But you feed him with a fishing pole. These guys are scary looking. These are not your. Not something you would. I'M Skip.

Paris Martineau (42:56)

Boys.

Leo Laporte (42:57)

You would never hold one of these guys. Right?

Paris Martineau (42:59)

Right. That big one. No.

Leo Laporte (43:01)

Little baby gator.

Paris Martineau (43:03)

The smaller ones. Yeah. I mean, at least when I worked at FUD Puckers, you could pay a certain price to feed the gators. You could pay a certain price to hold a gator and take a photo. Is that a capybara? Yeah.

Leo Laporte (43:12)

They had capybaras too.

Paris Martineau (43:13)

I didn't know that. We could have Cappy Bar.

Leo Laporte (43:15)

Isn't that cool? I've always loved capy bars. I've never met one.

Paris Martineau (43:19)

Obsessed. I follow this great Instagram account that I will never be able to find because the title is, I think in Japanese that's just videos of capybaras under little trickling water falls.

Leo Laporte (43:29)

Now, I should tell you, I hope it doesn't spoil us, but they do not smell good.

Paris Martineau (43:35)

Yeah, that makes sense.

Leo Laporte (43:36)

They regurgitate their food. They also eat their poop. So the. Yeah, like Gizmo. The reason they do is because they are herbivores and they. It's so hard to digest that stuff, they have to kind of do it several times.

Paris Martineau (43:49)

The reason Gizmo does it is every single time she feels like if she doesn't eat the food fast, as fast as humanly possible, she may never get food again. And then she eats it too fast and throws it up.

Leo Laporte (44:00)

Cats do that. It's ridiculous.

Paris Martineau (44:02)

I'm like, you've never gone without food a day in your life.

Leo Laporte (44:04)

Yeah. You don't even know what food.

Paris Martineau (44:07)

You don't know what hunger is.

Leo Laporte (44:09)

No, no, no. Gatorland was fun. And then we went to here. Oh, here's a gator you can hold. It's a gator made of butter. Did they say that's Richard Campbell holding Gator Land?

Paris Martineau (44:21)

They did it. Fudpuckers. And we had to constantly be like, not the same gator. But I didn't know if that was true or not.

Leo Laporte (44:28)

Yeah. You know, it's funny, we were walking around and. And Lisa said. Or Lisa or Anthony said, I wonder if we can eat gator here. And I said, you can't eat gators at Gatorland. That would be like cannibalism. And in fact, yeah, you can.

Jeff Jarvis (44:42)

It's Florida.

Leo Laporte (44:43)

It's Florida. And they said, well, we only eat the dead ones.

Jeff Jarvis (44:47)

Oh, that's great. Road kill gator. Thank you very much.

Leo Laporte (44:49)

We wait till they die and then we barbecue them up.

Paris Martineau (44:53)

And gators famously live a long time.

Jeff Jarvis (44:58)

Yeah, that's.

Paris Martineau (45:00)

That's hum. Killing the gators.

Leo Laporte (45:01)

I bet you think they were making that up?

Jeff Jarvis (45:04)

Yeah. Oh, lordy, it's Florida.

Leo Laporte (45:06)

I didn't eat any. It's okay. No, we had a good time. I, you know, I. I really enjoyed it. Kennedy Space center was amazing. I have a really good costume for Halloween now, and. But I missed doing the show. I really did. It's. I like doing this show.

Jeff Jarvis (45:24)

We had a good discussion about Anthropic last week.

Leo Laporte (45:27)

Well, let's get to it. But first, a word from our sponsor, if you don't mind. This episode of Intelligent Machines is brought to you by my mattress. I love my Helix sleep. Oh, man. You know, it's funny. Of all the nights of the week that I really need a good night's sleep. It's Tuesday night because I know I'm doing this show on Wednesday, and I want to be energized and ready to do the show. So even as I'm getting in bed, I'm going, thank you. Thank you, Helix mattress. I'm going to get a good night's rest, aren't I, tonight? And it took care of me. It took care of me. Maybe this would be a good time. Spring cleaning is coming to take that old mattress and put it on the street and upgrade to a Helix mattress. Actually. Actually, they took our old mattress when we got our Helix mattress, which was very nice, and get a good night's rest. I think we got the white glove service. They come, they take the mattress, they get rid of it. Anyway, let me tell you, the old mattress wasn't that old. Six, I think, eight years, something like that. But I had read that you should get new mattresses regularly because mattresses, who knew this wear out. You spend more than a third of your life on a mattress. You spend probably half your life, you know, curled up with a good book, cuddling your pet, your honey, or watching TV and then sleeping. So it's important. But the Helix. Oh, my gosh. No more night sweats. No back pain from a saggy mattress, no motion transfer. It doesn't rock and roll. Don't settle for a mattress made overseas either. You will find plenty of them online with low quality, questionable materials. They come to you six months off on a container ship, pushed in a box. It's just no good. That's not the Helix mattress. The Helix mattress is made in the USA to order, right? So the day you place your order, they start making your mattress. They assemble it, they package it, and then they ship it directly to you from Arizona. So it is fresh. It smells like the desert air. It's lovely. Not old, stale mattress, but a brand new one made just for you and by the way, made just for your preferences. We took the Helix Sleep quiz. You should too. It's right there on the website and you can tell it what your preferences are. You like firm, you like soft, you like medium, how you sleep, stomach sleeper, side sleeper, back sleeper and all that. And then they will match you because they have a whole bunch of mattresses with the perfect mattress based on those preferences. And does it improve your sleep? Oh man, it certainly does mine. You know, I know because I wear the Oura ring and I look at my sleep every morning to see how well I slept. And on average I'm getting maybe 50% more deep sleep. That's the important sleep, the sleep that clears your brain out. And I'm spending maybe another half an hour asleep. So when I wake up on my Helix mattress, I feel great. Well, it's not just me. In the Wesper sleep study Helix conducted, they measured the sleep performance of participants who did what we did, switched from their old mattress to a Helix mattress. 82% of the participants saw an increase in their deep sleep cycle. 82%. That's what I found. In fact, participants on average achieved 25 more minutes of deep sleep a night. That's a lot more. You only maybe deep sleep is half an hour to an hour a night. So that's a big improvement. Participants on average achieve 39 more minutes of overall sleep per night. That translates into just a better day and health and feeling great. Great. Do you feel tired sometimes during the day? You need a Helix mattress time and time again. Helix Sleep remains the most awarded mattress brand tested and reviewed by experts like Forbes and Wired. Helix delivers the mattress to your door with free shipping in the US and rest easy with seamless returns and exchanges that they call that the happy with Helix guarantee. It's a risk free customer first experience to ensure you are completely satisfied with your new mattress. I am. They can't get my mattress back. That's. I'm not giving it back. That's. That is my mattress. I love it. Go to helixsleep.com machines for 27% off site wide. Yeah, 27% off right now, site wide. During the sleep week sale. Best of Web exclusively for listeners of intelligent machines. So go to helixsleep.com machine jeans for 27% off the sleep week sale. Best of Web. This offer ends March 15th. Make sure you enter our show name after checkout so they know we sent you. And if you're listening after the sale ends, still check them out Always great deals@helixsleep.com machines. Thank you, Helixsleep, for my great night's sleep last night. So you spent a lot of time talking about Anthropic last week. I'm sure you did. That was the week when know Anthropic was declared a supply chain misdemeanor.

Paris Martineau (50:26)

Well, it was tweeted that it was a supply chain.

Leo Laporte (50:29)

Oh, this week they actually did get declared that.

Paris Martineau (50:31)

Yeah, this week. Yeah, yeah.

Jeff Jarvis (50:32)

And the suit and Anthropic this week sued in response.

Leo Laporte (50:36)

Yeah, they went, they went to court saying you can't do that. By the way, got amicus briefs in their case from OpenAI and I think from Microsoft.

Jeff Jarvis (50:46)

Microsoft.

Leo Laporte (50:47)

Microsoft and I think from Google employees. Yeah, it's a lot of support. I don't know. I am not convinced. I think there's two sides to this. As Guy Kawasaki pointed out, Anthropic did have, you know, it's not like they didn't want to be used by the military. They had a $200 million contract with the Department of Defense.

Paris Martineau (51:06)

So did everybody else though. The.

Leo Laporte (51:08)

Well, Anthropic was the only one being used in classified operations.

Paris Martineau (51:13)

That is true. Anthropic. The detail, the unique part about the anthropic is it was the first one to classify, receive a security classification to be used in classified operations. The other models from the other companies being used weren't being used in classified instances, however. I mean, I. I don't know. What do you think that there is that's gray about the situation?

Leo Laporte (51:36)

Oh, there's very much of a gray area. So here's the analogy. If Boeing sold airplanes to the. The military but said, but, but you can't bomb anybody with them. The military would say, well, wait a minute, you don't get to dictate this. We are an elected government and democracy. Civilian, controlled by elected officials. That's how military decisions should be made. Not by private companies. Right.

Jeff Jarvis (52:08)

So last week we talked about Ben Thompson's column in which he made that argument.

Leo Laporte (52:13)

He made that point. Yes.

Jeff Jarvis (52:14)

And said that you can't have the company. However, one analogy that I used in return is that pharma companies say to states, you can buy our drug, but you may not use it to execute people. And that's followed and they have a right to do that.

Leo Laporte (52:34)

Yeah, but that might. That's why I brought up at the beginning, Anthropic was already in a deal with the department.

Jeff Jarvis (52:40)

Yeah. And I think they already had those two conditions in that D. And the

Paris Martineau (52:43)

thing is, we've Learned since over the last week.

Leo Laporte (52:46)

They added that later. And the. And the reason that they added that was because they found out that Palantir had used Claude in the Venezuela extraction kidnapping.

Paris Martineau (52:56)

No. Anthropic has stated on the record multiple times that they have no qualms about Claude being used in the Venezuela extraction. And specifically, Dario Amade, a week or so ago, had a memo where he wrote the negotiations between him, Anthropic and the Pentagon. It basically succeeded. And that the main sticking point was that the Pentagon asked Anthropic to delete a contractual prohibition on, quote, analysis of bunk of bulk acquired data, end quote. This was the single clause that matched the mat like mass surveillance scenario that Anthropic considered a red line. And that's what this whole fight kind of started over this disagreement that happened at, like, last day of February.

Leo Laporte (53:43)

All the Pentagon was saying is we want to be able to use Anthropic for any lawful use. We want to use Clot for any lawful use.

Paris Martineau (53:50)

Yes, but that's what the Pentagon is saying. Now that they've gotten involved in this brouhaha, I think it's worth. Obviously we should scrutinize Anthropic statements as well. They're by far. They're not a unbiased actor. But the Pentagon is also very much not an unbiased actor.

Jeff Jarvis (54:05)

There's certainly an argument that this looks. This war in which it's being used, in which it was used by some reports with Palantir to bomb a girls school, is not a lawful war. So that doesn't give them full.

Leo Laporte (54:16)

I understand that. And of course, that's what complicates this, is we don't like this administration and we don't like the war they're waging, but they are a duly elected government. And I think it's pretty clear that we would prefer a duly elected governor government to make these kinds of decisions over a private company.

Jeff Jarvis (54:36)

Okay, now I'll go to the other example, which is IGE Farben and Zyklon B, you know, supplied things to a duly elected government of Germany and then years later were found guilty of war crimes and crimes against humanity because they did that. So there's still a responsibility. You can't just say this is Palantir's argument, because I heard Alex Karp do this in Vienna. Oh, the government will regulate us inside everything, and we can do anything else underneath. And so this is the tech argument is the government will regulate and then we're free. No, you still have a moral responsibility for what you choose and allow to do. And that is in the precedent of Nuremberg. And so if you believe that, let's just say that sending weapons off to kill on their own, making their own judgments and murdering people is a war crime, then do you not have a moral responsibility and a legal responsibility, responsibility under world law to avoid that? It's not as simple as, oh, the government can decide everything.

Leo Laporte (55:35)

What if it had been. What if it had been Ford during World War II, telling Roosevelt's and Stinson's Defense Department, yeah, you can't use our tanks to invade Germany?

Paris Martineau (55:51)

I think that a company has, A private company has a right to decide what it wants to do and what it doesn't want to do. Unless. Unless it is.

Leo Laporte (55:59)

Well, it does. It doesn't have to make a deal with the Pentagon at all. No, no, I agree. It doesn't have a. Doesn't have to make it.

Paris Martineau (56:05)

And part of making a deal with the Pentagon is you get to negotiate the terms of that deal, which is what Anthropic was doing. And I think it's well within the.

Leo Laporte (56:12)

Well, but they were doing it after the deal had been made, weren't they? Or maybe not.

Jeff Jarvis (56:15)

Well, both sides were doing part of.

Paris Martineau (56:17)

Yeah, the contract negotiations for this. The.

Leo Laporte (56:20)

Well, the presumption of the Pentagon should be in the.

Paris Martineau (56:24)

I believe, July of last year.

Leo Laporte (56:27)

The presumption of the Pentagon would be in any. With anybody, Boeing, Ford, or anybody doing a deal with them that you're allowing us to use your product for any lawful purpose. That would be the assumption.

Paris Martineau (56:38)

Anthropic argument is that they were being asked to have their product be used for unlawful purposes. Mass domestic surveillance operations is unlawful.

Leo Laporte (56:50)

Okay.

Paris Martineau (56:51)

Currently, I believe, also under human rights or like just general.

Leo Laporte (56:57)

I don't think there's a rule against autonomous weaponry yet. There should be.

Jeff Jarvis (57:01)

Well, there, yes, and you've argued that, but there too, Leo, it's also a matter of saying, you know, let's say we are for it, but these tools aren't ready.

Leo Laporte (57:10)

Okay?

Jeff Jarvis (57:10)

They make mistakes.

Leo Laporte (57:11)

And also, I don't. If you had this conversation last week, I don't want to repeat it. It's gonna bore people.

Paris Martineau (57:15)

We had this conversation last week, but we were all kind of on the same side that this is. You're. You're the first person to say no, we should. All companies should be forced to do war crimes.

Leo Laporte (57:29)

In summary, boy, that is not a fair characterization.

Paris Martineau (57:33)

What are you talking about? I'm being perfectly.

Jeff Jarvis (57:36)

I see how she plays Scrabble.

Paris Martineau (57:37)

Yeah, yeah, Scrabble is mad. With power. I'm going to be honest.

Leo Laporte (57:42)

If AI is. So this is what Noah Smith says, in no opinion. And we like, I think we like Noah Smith. Yes. He gives it as an example, nuclear weaponry. Should we be allowed to have nuclear weapons right in our own home? No.

Jeff Jarvis (57:59)

Where's the analogy there?

Leo Laporte (58:01)

Yeah, I mean, you could certainly say, well, if you feel that way, you shouldn't develop nuclear weapons for the government. Government. But I think we'd prefer that the government, if we make them, is the one controlling them as opposed to individuals. Yeah.

Paris Martineau (58:17)

I don't think anybody's arguing individuals or companies should own nuclear weapons, but much like nobody, I think should be arguing that a company that says, I think it's unethical to be creating nuclear weapons for the government in these specific cases, nuclear weapons for the government.

Leo Laporte (58:32)

I think it's disingenuous. Well, by the way, this is a

Jeff Jarvis (58:35)

machine, not a nuclear weapon. I am, you know, what you do with it.

Leo Laporte (58:38)

I am being a little devil's advocate because like you, I don't like how it was used and I don't want it to be used for mass surveillance or as I've said many times, the greatest threat of AI is for it to be used in autonomous killing machines. I don't like that idea at all. But I think there's some. It's. It's a conversation that isn't exactly clear. Marc Andreessen I think this was kind of a funny tweets. This is also from the NOAA page Opinion blog. Overheard in Silicon Valley. Every single person who is in favor of government control of AI is now opposed to government control of AI.

Jeff Jarvis (59:16)

Well, there is a point there because, because what we're arguing to the AI companies constantly is you must have safeguards. You must prevent bad things from doing with your being done with your tools. You must have.

Leo Laporte (59:26)

Who decides.

Jeff Jarvis (59:27)

You are responsible for following them and you're, you're responsible for deciding what the bad. Right. That's what we, that's generally what we do say to the AI companies and then come along. Oh, well, but no, we are. We decided that and you have to do it.

Leo Laporte (59:41)

You agree, though, that the nation states should have a monopoly in the use of force? Yes.

Paris Martineau (59:47)

What?

Leo Laporte (59:49)

Well, we allow law enforcement, for instance, to carry guns. We allow country.

Jeff Jarvis (59:54)

We allow everybody in this country.

Leo Laporte (59:56)

It's a little different. But I think in general, you know, war munitions, things that are used to make war are controlled by the state, not by individuals. We don't want individuals to have tanks.

Paris Martineau (60:09)

I don't see how that's relevant to this argument?

Leo Laporte (60:12)

Well, it's relevant in the sense that the theory is. Now I think this theory is kind of challenged a little bit by the current administration, but the theory is if you have a, a society should decide how these tools are used, not, not companies in the, and this is what Ed was pointing out, that our point of view about what a company should be in charge of and what a country should be in charge of seems to have shifted a little bit.

Jeff Jarvis (60:38)

Well then you're making the Palantir argument. And who's saying the company is right?

Leo Laporte (60:41)

Are companies the good guys or are

Jeff Jarvis (60:43)

responsible then for no moral judgment? Don't blame us. You can't sue us. You can't give us any problems because we're doing what's lawful within the rounds of what the government decided. That's no good.

Leo Laporte (60:54)

Okay, I'm sorry, I, I, I didn't mean to bring this argument up if you've already had it.

Jeff Jarvis (61:00)

It's being held across the country as we speak.

Leo Laporte (61:03)

I think it's, it's an important argument. And, and you're right when you talk about war crimes. If, if anthropic says, well, we don't want to be involved in war crimes, that's a fair thing for anthropic to say.

Jeff Jarvis (61:14)

You know, let me, this is, this

Leo Laporte (61:16)

is what is not, by the way, what is not appropriate is for the federal government to declare anthropic a supply chain risk. That is a designation actively using it

Paris Martineau (61:26)

in its current war.

Leo Laporte (61:28)

That is a, that is a designation reserved for foreign actors, not for American companies.

Jeff Jarvis (61:37)

I just listened to a book that I think you'd like called it's on youn by Nick Chaytor and somebody else whose name I can't see right now. And what it's really interesting is it's two behavioral scenes. Scientists who believed in the whole idea of nudging made famous by, I'm forgetting his name suddenly and then brought into the Obama White House and so on, that if we, if we nudge people to do the right thing, they'll do the right thing. And what these guys argue in a mea culpa for that belief is saying no, that's blaming the public for what the companies did, trying to nudge us all into being a little bit better about our climate behavior ignores the fact that it's major companies that cause the

Leo Laporte (62:19)

climate to go to Chaytor and George Lowenstein. It's on you how the rich and powerful have convinced us we're to blame for society's deepest problems.

Jeff Jarvis (62:28)

So then it turns around. I mean, and so I like that a lot. And I really respect them that they admitted their change of view. Nudge.

Leo Laporte (62:35)

Nudge is a kind of a Yiddish version of nudge. Paris. Paris is a little bit confused by that.

Jeff Jarvis (62:41)

What's his name?

Paris Martineau (62:42)

Listen, I was just. I wasn't gonna bring it up on the podcast because I think it's incorrect to try and discuss pronunciation, but I had to let it out of my head so that I could be free of my thought.

Leo Laporte (62:54)

Nudge vs Yiddish ism as opposed to.

Jeff Jarvis (62:58)

So it's Cass Sunstein that. That made that famous with Taylor. Who's a. Who's.

Benito (63:02)

But.

Jeff Jarvis (63:02)

But then they turn around in a way, and they absolve individuals of responsibility in some cases. Right now, I'm having this argument around, let's say vaccines in every case. What we're talking about here is who is responsible? Not only who has the power, but at the end of the day, who is responsible for the proper use of any technology, of any tool.

Leo Laporte (63:22)

Well, I don't think. I don't think. I think, really Anthropic opened the door to criticism by making a deal in the first place with the Pentagon.

Jeff Jarvis (63:28)

Well, I think OpenAI opened the door to criticism by playing and now and sneaking in.

Leo Laporte (63:33)

Yeah. And by the way, OpenAI this week got some pushback from its own employees, including the head of robotics at OpenAI.

Jeff Jarvis (63:41)

Pushback? Didn't he leave?

Leo Laporte (63:43)

Who? Caitlin Kalinowski.

Jeff Jarvis (63:44)

Quit?

Leo Laporte (63:45)

Yeah.

Jeff Jarvis (63:45)

Quit? Yeah.

Paris Martineau (63:47)

Yeah.

Leo Laporte (63:47)

She said it's a matter of conscience.

Jeff Jarvis (63:50)

Who was the other person who quit? Two major people quit. OpenAI.

Leo Laporte (63:53)

Yeah. I don't know who the other one is, but yeah. I mean, there was also a online open letter and petition urging artificial intelligence leaders to resist the Defense Department, but it seems to me you shouldn't even be making an agreement of any kind with the DoD if you don't, you know, if you want. I mean, how do you keep AI from being used?

Jeff Jarvis (64:13)

That's the problem. It's a general.

Paris Martineau (64:15)

That's what Anthropic was calling it. On this, as they were saying, it seems very difficult to institute or be able to ensure we don't cross these two red lines we have.

Leo Laporte (64:29)

Which is why you don't make the deal with the Pentagon in the first place. Right. So you just think you don't make Zyklon BE and then say, well, you can't use it to kill people.

Jeff Jarvis (64:39)

Oh, no, Zyklon B had other uses.

Leo Laporte (64:41)

Okay, well, you don't sell it to the Nazis, then. Okay. And then say, I am shocked, shocked at how they used it. What is it a pesticide?

Jeff Jarvis (64:50)

I think so.

Paris Martineau (64:51)

Well, how do you feel, Leo, about the consumer response to all of this over the last week with.

Leo Laporte (64:57)

Well, it's gone back and forth so briefly. Last week, Anthropics, Claude was number one on the App App Store. But now ChatGPT is back. By the way, number one again of

Paris Martineau (65:07)

ChatGPT rose I think like 300% over a couple of days.

Leo Laporte (65:12)

Yeah. ChatGPT's uninstalls grew by 295% while Claude got to number one on March 1st. However, earlier today, ChatGPT reclaimed the top spot on the Chat store. Claude is second, Google third, partly because ChatGPT came out with a new model, 5.4, which is benchmarking quite well. And there's a lot of ChatGPT fans who are, you know, raving over this. My experience, I played with it. I briefly paid the max fee for it just to see what I could do with it. And I, as you know, I moved my girlfriend out, put her stuff on the stoop and now she's back in. We, we've re re reunited. Claude and I are like this. So. But partly that's personality. I think that the models probably are equally good, but I just like the harness that Claude uses compared to the harnesses for codecs.

Jeff Jarvis (66:10)

Well, there were stories about OpenAI trying to open inside OpenAI's race to catch up to Claude code. I think there was some panic inside OpenAI.

Leo Laporte (66:19)

Yeah. And by the way, even before the 5pm deadline, OpenAI made a deal with Pentagon on Wednesday before that Friday deadline to take over. So OpenAI had banned military use. By the way, there's a story in Wired that said Pentagon used it anyway by going through Microsoft.

Jeff Jarvis (66:41)

Yeah, they can go through perplexity.

Leo Laporte (66:45)

And this is part of the problem, right, is. Yeah, I don't, you know, I don't know how you. This is, this is really. I mean, if in some ways I've become a little more of a doomer. We had Cory Doctorow on Twit on Sunday.

Jeff Jarvis (66:58)

That'll do it to you.

Leo Laporte (67:00)

He is an absolute doomer. I mean, he says we're gonna crash. This is the worst. Blah, blah, blah.

Jeff Jarvis (67:05)

Well, he's a Leo. Hold on a second. He's a doomer of a different sort. He's not the doomer of the sort that says, oh, AI is all powerful and could destroy the earth. No, no, no, I don't like these guys.

Leo Laporte (67:13)

He says AI is crap. It's More. It's more BS and it's going to crash.

Jeff Jarvis (67:18)

You need a different word.

Paris Martineau (67:19)

You respond to your maniacal optimism.

Leo Laporte (67:23)

I did not show that side of my personality.

Paris Martineau (67:26)

Oh, so you keep it on your wraps.

Leo Laporte (67:29)

I attempted in a. Well, you have to listen to the show to be the judge. I feel that I attempted to represent myself in a gentler Part of the problem is I have such deep respect for Corey's intelligence and verbal agility that I wasn't going to get in an argument with him. No, it's.

Jeff Jarvis (67:47)

That's.

Paris Martineau (67:48)

We need to have Corey on the show at some point. And before we do that, some. Some of our listeners have compiled. At least one of them has compiled a really fascinating notebook. Lm instance of all the transcriptions of our show. Someone should use that and find a way to put together a super cut of all of Leo's craziest quotes about AI like that he wants to give it his limbic system.

Leo Laporte (68:13)

I do

Paris Martineau (68:16)

anymore and put it all together. And then we need to show that. A couple minutes of that to Corey and then have him go, oh, Corey knows John there too.

Leo Laporte (68:25)

And I think, you know, I think I made some arguments in favor of. Oh, good Lord, they've made a graphic. Oh, this was from. This was from last week or two weeks ago ago. Did you see this graphic? I have not seen this.

Paris Martineau (68:40)

Yeah, I haven't. I don't know if I've seen yours.

Leo Laporte (68:43)

I don't know if this is accurate. Leo's AI Odyssey A decade. Oh, Anthony did this. Anthony, that's brave of you. A decade of shifting perspectives. I like how it gave me a nice tidy white beard. I would like to grow that beard. The age of skepticism. The accelerationist leap.

Paris Martineau (69:01)

No, I.

Leo Laporte (69:02)

This.

Paris Martineau (69:02)

Anthony didn't make this. This. I'm forgetting the name.

Leo Laporte (69:05)

Somebody else made it, but it's the

Paris Martineau (69:07)

one who's put together the notebook alarm.

Leo Laporte (69:09)

Leo shifts to a deized view, seeing AI interaction as first contact with an alien intelligence. Well, hey, I didn't say it was God. Guy Kawasaki said it's God. So. Okay, I'm. I'm just saying. I just.

Jeff Jarvis (69:27)

You were at a point, your skepticism was there because you were saying, oh, no, we don't want to talk about AI that much. No, right.

Leo Laporte (69:32)

AI Got better is my justification for it. It wasn't that good until very.

Paris Martineau (69:37)

I found it.

Leo Laporte (69:39)

Was this the notebook?

Paris Martineau (69:40)

Lm Ara Mova, who's been doing some phenomenal work. So these are all of our twig episode transcript. A bunch of them are in here and it's in the discord. They're very good. Yeah, I see guides in here. That a profile. A deep dive into the Persona of Leo Lepor. Let's see.

Leo Laporte (70:02)

And how many of my shows does it have in it doesn't. It doesn't have all of them because you run out of. You don't have enough room. Yeah, it's only got like, you know, 8, 19. No, it doesn't have even that many. You can you put 500 documents in here?

Jeff Jarvis (70:20)

I know it says 233 through 857.

Paris Martineau (70:22)

Okay, this is actually very funny. Again, guide to.

Leo Laporte (70:25)

No, but it's not continuous.

Paris Martineau (70:28)

Hold on, look at this. It's in the discord right now. I don't know. This one's very funny to me.

Leo Laporte (70:34)

Who's making these insulting graphics? The twit guide to starting on time. What are you laughing at? Bonito Lazy podcaster. Time Lazy Podcast.

Paris Martineau (70:50)

Leo's personal brother podcast philosophy. The network is built in the notion that no show should ever start on time. The steaming coffee trick. A classic radio ruse. If a host is late, someone places a steaming cup of coffee and paper in a typewriter to look like they just stepped out. What?

Jeff Jarvis (71:09)

I told that story from Time Inc. Once.

Leo Laporte (71:12)

Really? You told that story?

Jeff Jarvis (71:14)

Yes.

Leo Laporte (71:14)

That's hysterical. And just like a cigarette burning.

Jeff Jarvis (71:17)

And I'm getting blame here.

Leo Laporte (71:19)

I'm.

Paris Martineau (71:19)

I'm.

Jeff Jarvis (71:19)

Geez, what did I do?

Paris Martineau (71:21)

I like that it has a little gray haired Jeff in the corner.

Leo Laporte (71:25)

Oh, yeah, still talking. It says, oh, as he's waiting for the end of Windows Weekly. Still talking. Wow, that's pretty cute. Aha. This episode of Intelligent Machines brought to you by Monarch. I use this and I'm using it now, especially because it's tax season.

Jeff Jarvis (71:46)

Right?

Paris Martineau (71:46)

Right.

Leo Laporte (71:47)

That's when it's so funny. Most people all year long ignore their finances come April 15th. Oh, I gotta look at my finances now, you know, for next year. Monarch. Monarch is perfect for that because it helps you make progress with your money every day, every month, every week. So you don't just look back later and go, what happened? Where did it all go? Simplify your finances with Monarch. Monarch is the all personal finance tool designed to make your life easier. It brings your entire financial life. Budgeting, accounts, investments, net worth, even future planning, all together in a single dashboard, on your phone or on your laptop. You will feel aware, you'll feel in control of your finances this tax season and get 50% off your Monarch subscription. Just use the code IM. In fact, I just met with a financial advisor and that was Beautiful, because I can share my Monarch account with him at no extra church charge or your spouse if you want your significant other. And then you're all on the same page and I don't have to bring a stack of paperwork in. I can just say, here's my Monarch account. Monarch isn't your average personal finance app. Unlike most other personal finance apps, Monarch is built to make you proactive, not just reactive. Okay. I love that you can use the AI tools built on Monarch Intelligence. And this isn't just your everyday day model. This is a model trained by certified financial planners and advisors. So it has the smarts. It knows all of the stuff that you need to know. You'll get access to if you wish an AI assistant 24.7Access again if you wish to a financial coach that's available anywhere in Monarch. You can ask questions about trends in your spending to, you know, help me pay off debt. And the AI assistant has the answers. You'll also get AI insights. Monarch actually goes through the data to surface insights personalized to you hidden patterns. It identifies lifestyle creep versus inflation. You know, is. Is it? Are my costs going up because everything's costing more? Is it? Is it going up because my tastes are inflating? It will also highlight changes in savings rates, places where you're losing money. Gaining money is really great. And you even get an AI weekly recap app. You can let Monarch look out for your money with personalized weekly summaries that alert you to big spending spikes, big shifts in net worth, upcoming expenses. Again, it allows you to be proactive about your planning. And by the way, Monarch is great for splitting bills. No need for another app. You just scan and upload a receipt. Monarch will automatically parse items and prices for you. It's so smart. You share this is all new stuff by the way. Share a link or QR code with your group and then everyone can claim their items, settle the bill effortlessly. I have no friends so I don't use this. But if you have friends, if you're someone like Paris and you go places with the skeeball club now you can use Monarch to distribute the costs. It's fairly. It's so great. Achieve your financial goals for good with Monarch. The all in one tool that makes money management simple. Use the code I amonarch.com for half off your first year. That again 50% off. Go to Monarch M O N a r c h.com use the offer code. I am. We thank them so much for their support of intelligent machines. Poco just sent an Email to Guy. And Guy just sent him the Kindle book free. Good job, Poco. He got his copy of Everybody has Something to Hide with the middle finger. I did not see. See that. I've been looking at that cover for a month. I did not see that. All right, now. Now let's go back to our giant argument. Whatever that was. I forgot. Are we arguing about anything? No, I'm not a doomer.

Paris Martineau (75:54)

There's always time.

Leo Laporte (75:55)

I'm a bloomer. Oh, Jeff, you're muted, so you can't argue. Argue.

Jeff Jarvis (75:59)

I was muted by someone else.

Paris Martineau (76:02)

Jeff, did you get a haircut?

Jeff Jarvis (76:03)

No, I'm due for one.

Paris Martineau (76:05)

Sorry.

Leo Laporte (76:06)

Looking at that guy's hair, that. Did you get a haircut?

Paris Martineau (76:08)

I don't know. The ends of his hair looked kind of.

Leo Laporte (76:11)

They're like sticking out.

Jeff Jarvis (76:12)

Well, here's the problem, because I. I go to rest my back on the mountain of pillows, and so I get. I have all day. Bedhead. I just had to go. I dropped my ice pack of my back, so I had to go get my.

Paris Martineau (76:29)

Your grabber.

Jeff Jarvis (76:30)

My grabber.

Leo Laporte (76:30)

Oh, my gosh. Are you feeling. Are you feeling a little bit better, Jeff, though?

Jeff Jarvis (76:34)

You're getting better. I'm getting better. I walked a mile and a half today.

Leo Laporte (76:36)

Glad to see it. Oh, we got dual grabbers. Mine's over on the other side of the room. Yours is, like, really scary looking. Paris. It's got, like, teeth.

Paris Martineau (76:45)

I guess it's teeth. The important.

Leo Laporte (76:49)

You could bite somebody with that grabber.

Paris Martineau (76:51)

I gotta get my packages.

Leo Laporte (76:53)

It's got choppers. Yeah. Through the grill. That's the funniest story.

Paris Martineau (76:57)

Because, you know, if I put my hand through it, the gate is rusty and flakes off. And then I get myself scratched with all the rust. And then I get anxious and think I have tetanus urban living.

Leo Laporte (77:09)

Ladies and gentlemen, it's a delight. Meta has signed a multi year AI content licensing deal with News Corp. Oh, that one.

Jeff Jarvis (77:20)

Oh, that deal. Yeah.

Leo Laporte (77:21)

$50 million a year here. They're gonna. That means they can ingest the contents of the Daily News. Another. The Post. Sorry, the New York Post and all of the other Rupert Murdoch things. I think probably Wall Street Journal would be the most valuable thing. And that's who's reporting this.

Jeff Jarvis (77:41)

So. Robert Thompson, If I can find this. Last week we didn't talk about it. Robert Thompson, who's the CEO of News Corp, said that News Corp is now input. Yeah, yeah, it was paid as input. But I hate these deals because it's just the big pay in the big it leaves out the vast diverse ecosystem of a pluralistic society. It's just moguls making, you know, taking home a bucket of money so that Rupert Murdoch won't sue them. That's all this is.

Leo Laporte (78:12)

Meta has completed deals with people, USA Today, CNN and Fox News News.

Jeff Jarvis (78:17)

Does that represent the country fully?

Leo Laporte (78:19)

No. So what is. Ben is using this for training, not. Not for the Facebook side.

Jeff Jarvis (78:26)

Well, they might be using it for rag on the AI. I don't know.

Leo Laporte (78:29)

Yeah, because training is over. Rag is dead.

Jeff Jarvis (78:33)

Well, training's dead because they have synthetic data. No, Rag. They need rag for currency. If you want current information, that's the issue for them.

Leo Laporte (78:44)

This is the funniest story. Remember when Facebook acquired a company spent billions of dollars to acquire, basically Alexander Wang?

Jeff Jarvis (78:57)

Right.

Paris Martineau (79:00)

Is this reporting true? What is this source IDN financials?

Leo Laporte (79:05)

Yeah, that's a good question. Actually, I don't know because Facebook.

Paris Martineau (79:08)

Facebook because. Not that I want to hand it to Meta Comms, but Metacoms has said everything about this is factually untrue.

Leo Laporte (79:16)

Oh, okay. So first of all, there, according to the Times of India, there is a new applied AI engineering organization at Meta.

Paris Martineau (79:28)

Yeah.

Jeff Jarvis (79:28)

Which divides AI work, sells editorial space.

Leo Laporte (79:32)

Oh, well, okay, we'll leave that one out. That will be led.

Jeff Jarvis (79:36)

You don't know which is which.

Leo Laporte (79:37)

Yeah, yeah. Now the Wall Street Journal, which also sells editorial spaces, we've just learned. No, not. Not in the same way. Says the group will adopt a very flat structure with 50 individual contributors for every manager, aimed at speeding up decision making. But I wonder, this is.

Jeff Jarvis (79:56)

If they hadn't hired Wang, would. Would Yan Leon have one left?

Leo Laporte (80:01)

I don't know. Well, Yan Lecun's doing all right.

Jeff Jarvis (80:04)

Amen.

Leo Laporte (80:04)

He has signed a deal, as has Dr. Fei Fei Lee. Both have signed deals for new companies that are focused not on LLMs, but on learning from the physical world.

Jeff Jarvis (80:16)

If you go to his site, which is AMI xyz. I think it is.

Leo Laporte (80:23)

Yeah.

Jeff Jarvis (80:24)

AMI Advanced, amilabs xyz.

Leo Laporte (80:28)

He's still going to teach at nyu, but the company is in Paris.

Jeff Jarvis (80:31)

They have raised Montreal, New York and Singapore.

Leo Laporte (80:34)

This is how he overheated the market. The investments are right now in AI. He has already raised for a brand new company with no product for seed. Seed more than a billion dollars.

Jeff Jarvis (80:46)

1.03 out of 3.5 billion. Evaluation.

Leo Laporte (80:49)

Coon says because our reasoning is grounded in the physical world, not language. This is also a Fei Felice says AI world models are necessary to develop true human level intelligence. So he's going for The God. The God. Intelligence as well, right?

Jeff Jarvis (81:05)

Well, no, what, what Jan says is that yes, we will have high intelligences, but he doesn't go for this notion of the general intelligence will do everything. One machine will do everything. He goes for specific tasks. And he also, you know one of his arguments for his company, which is really important, if you go to Amelabs, Friendlabs, xyz, you'll see a one page description of this. It's very good. He also argues that this is the path to safety. It's not about the whole AGI bs, it's about you give a machine a task, it does the task, you limit the task. That's what you do. And so he believes that world models will lead to way past what text can do, will understand the world. Fundamentally that we share one belief. He says real intelligence does not start in language, it starts in the world. And this is his argument that he makes frequently about cats and toddlers, about what they learn. And when you look at a video, you're not predicting the next pixel. That's going to get you nowhere. That's a lot of noise. The machine has to start to learn concepts like this is a chair and people sit in a chair and the chair can move and that from that basis you get a lot farther than you do with simple text prediction.

Leo Laporte (82:18)

He does say, according to this article, AMI intends to develop a universal world model which would be the basis for a generally intelligent system that could help companies regardless of what industry they work in. This is from Wired magazine. It's very ambitious. He says with a smile.

Jeff Jarvis (82:37)

Interesting that his list of investors includes

Leo Laporte (82:40)

Toyota, includes Mark Cuban, Eric Schmidt, it includes Bezos Expeditions.

Jeff Jarvis (82:50)

That's the individual level. He has five venture companies, including Bezos's are led the investment. Yeah, but interestingly and I don't they could be there, but they weren't mentioning any story. I thought that Meta or Zuckerberg was going to invest

Leo Laporte (83:07)

maybe not. Mira Moradi, who Of course left OpenAI. She was the CEO pro tem when they briefly got rid of Sam Altman. She has struck a deal with Nvidia to buy a bunch of Vera Rubins.

Jeff Jarvis (83:26)

The new investment from Nvidia Circle. Circle. Circle, yeah.

Leo Laporte (83:30)

Yeah. So her company Thinking Machines, which has gone through some ups and downs, had a $2 billion seed round so there.

Jeff Jarvis (83:42)

Well, for Europe I think a billion. Do we know what her philosophy is in the sense what her. Is she scale and LLMs? Is she world models? Do we. Is she just. I'm going to be safe. Do we have Any idea what her

Leo Laporte (83:55)

I don't think we know much about thinking machines is trying to do? To be honest, no. And I'll be frank, I'm a little skeptical of Yann Lecun.

Paris Martineau (84:07)

But, Leo, what you need to recall is that Yann Lecun is, as far as I know, the only head of a AI team that is kind of thinking about all of this through a cat centric lens.

Leo Laporte (84:22)

Well, I'll give him that.

Paris Martineau (84:23)

That he put put is we can't even replicate cat intelligence or rat intelligence, let alone dog intelligence. All right, I'll say that is an asterisk. He is putting dogs out of cats. Any house cat can plan a very highly complex actions and they have models of the world.

Leo Laporte (84:39)

I don't think a house cat can write a TUI application RSS reader, but okay, if you say so. You.

Benito (84:45)

Well, this is the problem. We're modeling it after things like cats and people when we shouldn't be doing that. It's a whole qualitatively different thing.

Jeff Jarvis (84:53)

Yeah, well, he's not modeling it. He's arguing. He's arguing that the data input in a toddler's or a cat's mind and its abilities far exceed what we can see in a text based in certain areas.

Leo Laporte (85:06)

You know, meta did buy. They bought. They bought, weirdly, Malt book, Facebook. Good question.

Paris Martineau (85:14)

This is one of the dumbest things possible. I mean, this is another huge sign, and this is a bug bubble that's about to pop. Spending.

Leo Laporte (85:24)

We don't know how much they spent.

Paris Martineau (85:26)

Any amount they spent on it is too much.

Leo Laporte (85:29)

And what's really surprising to me and kind of miffs me a little bit, is they mentioned the two founders, one of whom I did not know.

Jeff Jarvis (85:39)

Schlicht.

Leo Laporte (85:40)

Mr. Schlicht. But one of whom we know very well, Ben Parr, who's been on our show many times as an AI startup. I wish I'd known he was responsible. Responsible for Molt Book. We would have had him on the show. Now, I don't know if we can get him. I wrote him a note congratulating him and saying, hey, before the acquisition goes through, which is about five days away, can you talk to us? But you know, at this point, he's probably inside the maw of meta and can't. But anyway, congratulations, Ben. I'm very happy for you. And it's being. TechCrunch is characterizing it as aqua higher that really.

Jeff Jarvis (86:19)

But again, what were they buying?

Leo Laporte (86:20)

Well, they bought these guys. They didn't really. I mean, look, what did you get with Peter Steinberger when OpenAI bought Open Claw. Nothing.

Benito (86:29)

It's an MSP play, isn't it? Isn't it, like, for them to get that stuff? MSP stuff?

Leo Laporte (86:36)

It's a management play. I don't know what the. I don't know what the point of it is. There's so much hype going on. I don't.

Paris Martineau (86:42)

They're Vibe Aqua hiring now. They're not just vibe coding. They're vibe. They're vibe acquiring.

Leo Laporte (86:50)

I mean, Malt Book was more like a meme than anything else. It was more like just, whoa, that's wild. Anyway, Matt Schlicht and Ben Parr, well done. Congratulations. I hope you got a big bag of money of Mark's money, because Mark's throwing it away right now. He's also getting a lot of data. Here's the question asked by Abraham Dialio. Now, where did you think the training data was coming from? We're learning now that those meta AI glasses send the video feed directly to Facebook servers and is then reviewed by contractors in Kenya. And the contractors in Kenya are kind of like, revolted by what they're seeing.

Paris Martineau (87:34)

Yeah.

Leo Laporte (87:35)

They said a lot of people don't know that they. Maybe that they're. Their glasses are still on when they go to the bathroom or make whoopee or all kinds of things.

Benito (87:47)

Oh, they know. They just didn't know.

Paris Martineau (87:48)

I'm so sorry.

Leo Laporte (87:50)

They just didn't know that anybody else was watching.

Paris Martineau (87:53)

If you are. If you're hooking up somebody who's wearing meta glasses, don't hook up with idiot.

Leo Laporte (88:01)

Yeah, don't. Just don't. Yeah, maybe you didn't know, but why is a person wearing these goggles to bed anyway?

Paris Martineau (88:11)

Someone wearing a hat. It's just like, what's going on here?

Leo Laporte (88:14)

Why are you wearing.

Paris Martineau (88:15)

A lot of questions.

Leo Laporte (88:17)

Why are you wearing a hat?

Benito (88:19)

But this is kind of like what all the AI companies are doing. They want to put. They want to give you a wearable or put something in your house because there's no more training data left on the Internet. They need to get.

Leo Laporte (88:27)

Surely you've heard the Randy Newman.

Jeff Jarvis (88:28)

They've got synthetic data. Now they don't.

Leo Laporte (88:30)

You can leave your hat on.

Benito (88:32)

No, they want to get it from you.

Paris Martineau (88:33)

You cannot.

Leo Laporte (88:34)

You can leave your hat on. Come on, man. Let's make some magic. No, don't. No.

Paris Martineau (88:48)

The podcast Descents of Silence.

Leo Laporte (88:53)

Leo's wearing his glasses. Don't do anything. Don't say anything. He's got his glasses on. On. It's going straight To Kenya.

Paris Martineau (89:02)

Should we all look down our glasses like this for the. Or do something with the glasses for the.

Leo Laporte (89:10)

Okay, there we got the thumbnail. Gotta do it.

Benito (89:13)

Thank you.

Leo Laporte (89:15)

Apparently Instagram, speaking of meta, is doing a kind of a not so good thing. If you are an influencer who sells stuff, you know there are a lot of. I, I hear I got rid of my Instagram, but there of are a lot, a lot of influencers on there, you know, selling hey, this milkshake's fantastic. There is a Insta influencer named Julia Berlsheimer who posts a shop, you know, a shopping Instagram. She noticed all of a sudden a shop the look button in the corner of her videos. And when followers clicked it, they were fed similar items to wear to what Berolsheimer was wearing. Wearing, but not the ones she was actually hawking. She didn't place the links there herself. Instagram added them without her consent. The product links led followers not to the actual items Berlsheimer was promoting and earning commission from, but to and this is even worse. Lookalikes.

Paris Martineau (90:17)

How dare they.

Leo Laporte (90:18)

My followers were being shown cheap knockoffs and random items for brands off never heard of attached to my image under my name.

Paris Martineau (90:30)

This can't stand.

Leo Laporte (90:31)

This will not stand.

Paris Martineau (90:32)

What if I don't get my commission from every click through?

Leo Laporte (90:37)

For some reason I'm starting to sound like Tim Gunn.

Paris Martineau (90:41)

What if someone else gets a commission when they go to Mood and pet the dog?

Leo Laporte (90:48)

Anyway, TikTok also apparently attesting a new feature, similar feature, according to Miya Sato writing at the Verge. So I guess this is inevitable, right? This is just another kind of advertising.

Jeff Jarvis (90:59)

It's the holy grail. They've been saying. Oh, you're watching.

Paris Martineau (91:01)

I'm shocked that Instagram hasn't done this already because Instagram, like all meta products, is just a collection of stolen features from other apps. And Pinterest has been doing this forever, as has Google. This is Google Images, basically.

Benito (91:16)

Wait, if influencers aren't getting paid anymore, how is this still happening? How it is this sustainable at all?

Paris Martineau (91:21)

Yeah, no, influencers make crazy money and get crazy amounts of free.

Benito (91:26)

Yeah, but if they're not, a lot

Paris Martineau (91:27)

of influencers are making well over a million dollars.

Benito (91:29)

Yeah, but if they're. It's not from they stop getting click throughs then they're not, they're going to stop paying the influencer.

Leo Laporte (91:37)

Right?

Benito (91:37)

Like this is, this is how advertising is.

Paris Martineau (91:39)

Most of the influencers who are getting paid large sums are not doing click like pay through click Sort of things like you're not, they're not posting an affiliate link. You are. They are product placement, right? It's product placement, basically.

Benito (91:55)

Yes, yes, but the, but the, if the link there doesn't go to that product, that the people.

Paris Martineau (91:59)

But there's no link anymore in most cases.

Benito (92:01)

Well, yeah, in most cases. But if Instagram is adding a link now, like Instagram gets all that ad revenue and like it's not being paid for by the company.

Paris Martineau (92:10)

I mean, I don't disagree with you in principle. I think that all advertising, advertising is just a cycle of advertisers slowly realizing how meaningless the actual advertising is. But I think that influencer marketing is an area where it's going to take a very, very long time for advertisers to ever give up on because it is so much of the dominant industry. I mean, people, anecdotally small time influencers, I know are getting crazy amounts of, of money and products.

Leo Laporte (92:42)

I'd just like to say that Jeff Jarvis's casual drawstring long sleeve kangaroo pocket black sweatshirt is available for $8.76 right now on T Move. Thank you.

Jeff Jarvis (92:52)

This is Uniqlo, damn it, at least $10.

Paris Martineau (92:56)

Did you see, Jeff, that Uniqlo is partnering, I think the New York Public Library to do custom merch?

Jeff Jarvis (93:04)

No. Oh, good. I'll do that.

Paris Martineau (93:05)

We all got to get New York Public Library tracksuits.

Leo Laporte (93:09)

I love that idea. I love the New York Public Library. I, I like to bathe in its lovely bathrooms. Amazon has won a court order to block Perplexity's AI shopping bot. Remember we talked about this? Last November, Amazon sued Perplexity, saying we can do it, but you can't. This is Perplexity's Comet browser. Browser. A judge ruled in favor, said no, granted a temporary injunction blocking Perplexity from the Comet browser from buying stuff on Amazon. US District Judge Maxine Chesney said Amazon provided strong evidence that Perplexity's Comet browser accessed its website at the user's direction, but without authorization from Amazon.

Jeff Jarvis (93:57)

So hold on here. Right this, this strikes me as a bit crazy. One, if you go to the Guardian and you read a book there and there's an Amazon link, the Guardian will get paid for having sent that reader to Amazon whether or not they buy the book. Two, so Perplexity sends an actual purchase to Amazon. It's a guaranteed purchase.

Paris Martineau (94:21)

It's bought.

Jeff Jarvis (94:22)

Right. And Amazon's going to make money from that purchase. Purchase. And in a sense, Amazon should be paying Perplexity for the saved marketing costs.

Paris Martineau (94:29)

There.

Jeff Jarvis (94:30)

But instead they're going to court and stopping Perplexity because they want to have their own AI thing doing it. And it just seems inconsistent as hell with the model that's been established.

Leo Laporte (94:41)

Paris, you think it's funny that Perplexity is introducing something called Computer.

Paris Martineau (94:45)

We brought this up last week and it was the title show. Just because I think that, like, how is this ever supposed to make it to the broader population? Am I supposed to go and tell my mom? You gotta get computer now. It's not a computer on your phone right now. What. What's gonna happen?

Leo Laporte (95:06)

Computer on my computer.

Paris Martineau (95:08)

But I've got computer. It's right here. I'm looking at computer.

Jeff Jarvis (95:11)

Computer has no computer.

Paris Martineau (95:13)

You've got to get computer on your computer right now. You've got to download computer. It is.

Leo Laporte (95:17)

You gotta get computer.

Paris Martineau (95:18)

Computer.

Leo Laporte (95:18)

For enterprise, it is the era of the agent. I mean, we. We knew this would happen. After OpenClaw 2026, it's going to be all about agents. Paul Thurat was talking about a. An agent called Clairvoyant. That Clairvoyance, sorry, that comes from a company that is mostly known for letting you move the Windows menu around. Object Windows.

Jeff Jarvis (95:42)

Do you need a company to let you.

Leo Laporte (95:44)

Yeah. And they call it AI. It's the same thing, by the way, as Computer for Enterprise, AI staff for everyone. What these are, are harnesses to the AI. So when you use CLAUDE code. Claude codes is an interface. We technical term for it is a harness to the CLAUDE model, which is then going to do the coding for you. But it's done in such a way that it kind of fits the way the coders think or something like that. And the harness also allows you to add skills and memory and hooks and all sorts of stuff like that. And so we're going to see more and more of these harnesses. Clairvoyance is AI model agnostic. You can bring your. You bring your own model. Perplexity is sort of the same. And that's what Perplexity does, is they. They orchestrate multiple models. So I think we're going to. This is just the tip of the iceberg. We're going to see so many of these. Amazon's Health AI is now open to the public. Public.

Jeff Jarvis (96:40)

Did you use this when you were part of the Amazon Health thing?

Leo Laporte (96:43)

Paris, you had Medical one or one?

Jeff Jarvis (96:45)

It was. It was available for medical One patients only, but now it's.

Paris Martineau (96:49)

Is that a thing?

Jeff Jarvis (96:51)

Apparently.

Paris Martineau (96:51)

I didn't know. I didn't read this. How was it used?

Leo Laporte (96:55)

Dear Amazon. I have covered the flu and a stuffy nose. What are my symptoms?

Paris Martineau (97:03)

What

Jeff Jarvis (97:05)

reversed?

Leo Laporte (97:06)

Yeah, just see if I can confuse it. So you never used anything like this?

Paris Martineau (97:15)

No, but I have noticed all the time in my One Medical app, something pops up and it's like chat to One Medical chat. That's what's in every app now.

Jeff Jarvis (97:24)

Are you still One Medical?

Paris Martineau (97:26)

I am, yes.

Jeff Jarvis (97:27)

Oh, I didn't know you kept that.

Leo Laporte (97:29)

Oh, open. AI.

Paris Martineau (97:31)

I just, I haven't been able to find another. I guess I should find a different provider, maybe through Mount Sinai, because that's where I'm seeing.

Jeff Jarvis (97:38)

What was it Dr. Bruce? What was your doctor you lost Dr. Dan?

Paris Martineau (97:42)

That. My issue is that Dr. Dan is in D.C. now and legitimately I, Whenever I come across someone, I will. I, I really. One thing I really like about one medical is if you're a member, you can access like 24 hour on call video appointments, the nurses, no insurance, no co parent.

Leo Laporte (97:59)

Dr. Dan. Sounds like a dream.

Paris Martineau (98:01)

And often I will get emotionally. Yes. I will get on there with somebody and they'll be like, oh my God, you saw Dr. You saw Dan. We love Dan. Dan is one of the best doctors we have. I'm just like, yeah, everybody loves Dan. He wore little glasses. He would always compliment my. My boots and then ask me about the fun bags I have to get his wife. He claimed to have once been Obama's doctor and had a note from Obama on his wall.

Leo Laporte (98:32)

Well, that's pretty good.

Paris Martineau (98:33)

Oh, he, he's great. Honestly. Just a phenomenal doctor.

Jeff Jarvis (98:38)

Is he like Dr. Robbie?

Paris Martineau (98:40)

I don't know.

Leo Laporte (98:40)

Sorry. That he went to watch is pretty. That's where Obama is. Maybe he went back to being.

Paris Martineau (98:46)

Maybe he did.

Jeff Jarvis (98:47)

New York.

Leo Laporte (98:49)

Oh, I thought he had a place in Washington.

Jeff Jarvis (98:50)

He has multiple places. Yeah, Chicago too.

Leo Laporte (98:53)

He's got Chicago. There's Dr. Dan.

Paris Martineau (98:56)

You know what? I forgot if I told you this in the. If I have told the story on the podcast and if I have, stop me. I was. I've been dealing with a bunch of different medical stuff and I had to get a CAT scan for my sinuses.

Leo Laporte (99:08)

Yeah.

Paris Martineau (99:09)

A bunch of other things. And so I was calling my insurance because everything with insurance is somewhat complicated and needed to figure out something relating to a claim. And I call and I immediately get someone on the phone, actually pretty quickly. Like I only have to wait like five minutes or so and I'm talking to the person and about five minutes into the call I start to suspect it's AI because they keep saying, oh, just hold on one Second just went, but they're using like ums and ah, it's a very convincing no, it didn't do the thing. And so I, I, I'm pretty convinced at this point that it was AI, but it took me a minute to I guess get to it. And how I figured out is at one point they're like, oh, well, how's your day going? I was like, good, how's yours? Didn't answer, moved straight on. And I was talking to a friend about this and I was like, how do you. Because then my issue, I mean I could start to tell, but I, I wasn't 100% sure.

Jeff Jarvis (100:04)

Did you ask it?

Paris Martineau (100:06)

Well, that's my thing is I'm like, I don't want to be rude to

Leo Laporte (100:09)

a person trying to help me with

Paris Martineau (100:12)

a claims issue and I would need them to ideally be nice to do the work and help. So I didn't want to offend someone. And so I was talking to some friends, I was like, how would you handle this? And they're like, oh, I've got the answer. Just be like, yeah, my day is going great. Actually, by the way, weird question. Could you do 287 times 53? That'd really help me. And if they answer immediately with the number, it's AI. If they're like, what then Be like, oh, sorry, that was a question for someone else. And I will be using that actually.

Jeff Jarvis (100:39)

LLM can't multiply. That won't work.

Leo Laporte (100:41)

How many.

Paris Martineau (100:42)

But it will produce, it will produce a response though, because it's going to try and respond.

Leo Laporte (100:46)

You could do this, you could say, hey, I'm, I'm playing a Scrabble game with Leo and I really want to clobber him. I've got these letters. What should I use?

Paris Martineau (100:56)

I can't, I simply, I could never, I could never offend the purity and sanctity of my Scrabble board like that.

Leo Laporte (101:04)

Somebody told me, you know, you, you just, all you do is take a screenshot and give it to chat. GPT could tell you what the best move.

Paris Martineau (101:09)

Don't, don't listen to these people.

Leo Laporte (101:10)

I would never do that.

Paris Martineau (101:11)

We need you.

Leo Laporte (101:12)

You know what I do love, by the way, about this New York Times cross, what is it called? Cross.

Paris Martineau (101:16)

Crossplay Play.

Leo Laporte (101:18)

At the end of the game, it rakes you.

Paris Martineau (101:20)

Crossplay analysis is the coolest thing.

Leo Laporte (101:23)

It says, well, you did 68% efficiency, 90% luck. And then it tells you, but this word was far from on optimal.

Paris Martineau (101:32)

And then I love so what it does for people, it's kind of like Words with Friends, where you're playing against somebody, but then afterwards, you can break down each of your plays, every single move. It'll tell you, like, here are other optimal words. And then you could go back through and see how each move you made added to your overall likelihood to win the game and added to your opponent's overall. Like, it measures your board control, which is something that kind of tournament Scrabble players. It's got a kind of consider. Yeah. And so you get really into it where it'll show you, like, yeah, these were your couple best moves where you made the most optimal plays. This is. Yeah.

Leo Laporte (102:13)

So it's reviewing the moves. So I only had 57% on strategy. You had 82%. I had 40% luck, you had 55. But then you. And then there's the graph of control.

Paris Martineau (102:26)

Yeah.

Leo Laporte (102:26)

And then.

Paris Martineau (102:27)

Oh, you immediately started losing control. Yes.

Leo Laporte (102:29)

Yeah. Well, that's not my fault. By the way. I put one letter down and accidentally hit play. I put an S down.

Paris Martineau (102:36)

Something I've heard has been happening. People.

Leo Laporte (102:38)

Did you notice that I added an S to a word? And that was my move. That's a dopey move. I made a point or two. And. And that was. That was when.

Paris Martineau (102:50)

And you've got to hold on to your S's. I have some of the highest value.

Leo Laporte (102:53)

I know I have no excuse for it. So here's the key turns in this. And when you have a chance to learn. That means stupid move. Earrings. I thought earrings was a good move, but. Okay, well, I thought it was too.

Paris Martineau (103:09)

It was pretty good.

Leo Laporte (103:09)

Yeah, it was a nice long word. You're really a killer.

Paris Martineau (103:13)

And then you can go into that. Simulations, detail. Click that. View additional details.

Leo Laporte (103:17)

Oh, advanced details here. Oh.

Paris Martineau (103:20)

So then it shows you based on what word you did, how much it contributes to your overhead. Likelihood, Likelihood of winning. And so this is kind of interesting in the first couple moves because I think it. Part of what it shows you in the first move or two is who's most likely to get the double word bonus first.

Leo Laporte (103:40)

Right.

Paris Martineau (103:40)

And so a big part of, like

Leo Laporte (103:43)

Scrabble strategy play defensively.

Paris Martineau (103:46)

I do. Yeah.

Leo Laporte (103:47)

Yeah, I thought you did.

Paris Martineau (103:48)

Depending on who I'm playing with.

Leo Laporte (103:50)

You keep killing me. I like, I'm already. I got a great word. It's ready.

Paris Martineau (103:54)

Well, Leo, I'm gonna forever be salty because the one time I wasn't playing defensively and I was like, listen, this is my only. The one Leo is saying all of this like I'm a killer. My only Game I've lost has been about against him. One of our games that we.

Leo Laporte (104:09)

It's the only game you've ever lost.

Paris Martineau (104:12)

The only game I've lost on cross play so far.

Leo Laporte (104:15)

Play? You say you play how many at a time?

Paris Martineau (104:17)

I've got like 15 games going right now.

Leo Laporte (104:19)

So you're just massacring your friends?

Paris Martineau (104:23)

Well, only one. Only you and one friend I'm playing. The others are randas.

Leo Laporte (104:27)

Well, I'm gonna learn how to hold on to my ass from now on with you.

Paris Martineau (104:30)

I'm going to. I keep trying to do better and better and then request to be paired with people so that they pair me with better and better players because they

Leo Laporte (104:39)

will also do that. They will match you to your.

Paris Martineau (104:40)

Yeah, they match you based on your strength.

Leo Laporte (104:42)

So I shoot where you're playing.

Jeff Jarvis (104:43)

1 months now 15.

Leo Laporte (104:45)

She said, Geez,

Paris Martineau (104:49)

let me see, let me see how many.

Leo Laporte (104:50)

So Amazon has told its engineers, you 80% of your work needs to be done by AI. But then they also told, oh, here we go, here we go. Look at all these crowns on her head. The crown means you won all the

Paris Martineau (105:06)

games I've got going on right now.

Leo Laporte (105:09)

Well, now I'm not going to feel so bad if I win, but I'm not going to feel so good if I lose.

Jeff Jarvis (105:15)

Is there a time limit on responding? Can you leave?

Paris Martineau (105:17)

Seven days.

Benito (105:19)

Yeah.

Leo Laporte (105:19)

You get a week. Yeah, okay. But people can lose pretty fast. I could tell when you're busy though because you won't respond for like a day.

Jeff Jarvis (105:26)

She's doing actual.

Paris Martineau (105:27)

Yeah, the weekend. I honestly is my slower time.

Leo Laporte (105:36)

So anyway, Amazon Senior engineers told 80% of your work needs to be done with AI. And then to their chagrin, told, you can't use Claude, you got to use our Amazon AI named Cairo. Now, after two incidents where AI Coder tools crashed Amazon's servers, they had an emergency all hands meeting yesterday. Free deep dive into a spate of outages. There had been a trend of incidents in recent months characterized by a high blast radius and gen assisted changes, among other factors. So they're basically saying, folks, as you likely know, the availability of the site and related infrastructure has not been good recently. We gotta really rethink our use of AI.

Jeff Jarvis (106:28)

That's a quote.

Leo Laporte (106:29)

Yeah. Junior and mid level engineers will now require more senior engineers to sign off on any AI assisted change.

Jeff Jarvis (106:36)

Senior engineer has to check things and look for booby traps.

Paris Martineau (106:39)

Yeah, I mean and from what I've heard, it's hard to check five coded stuff, especially when you didn't write it or didn't prompt it.

Leo Laporte (106:47)

I'm not real happy because they call their weekly meeting this week in stores tech or twist. That's awful familiar sounding. They had a 13 hour interruption at one point due to a cost calculator after engineers allowed the group's Cairo AI coding tool to make certain changes and the AI tool opted to delete and recreate the environment. Whoops.

Benito (107:17)

If a person did this, they're fired right away, right? If a person did this, they're fired right away.

Leo Laporte (107:21)

Oh yeah, but you can't fire an AI, especially if it's your own AI. Can we please use quad? We don't like Cairo. We don't like it one bit. Another reason Amazon's struggling a little bit is because it has a lot of data centers in the Middle east, two of which now have been hit. Amazon said buy a projectile, but in fact it was an Iranian missile.

Jeff Jarvis (107:43)

Another word for missile.

Leo Laporte (107:46)

Somebody dropped a rock out of an airplane and hit us exactly the wrong way. And this is the problem. There are a lot there. There has been a huge build out of AI data centers in the Middle East. Turns out not the best place as it turns out. Amazon said some objects data centers in the heat. I know why. In the heat. Oh, you know why? Because that's where the money is.

Paris Martineau (108:10)

It's because we're the.

Jeff Jarvis (108:11)

Why do they build them in Arizona?

Leo Laporte (108:12)

That's where the tax breaks are. Oh. So Amazon's cloud center services are down in some of the Middle east after, as Amazon called them, objects hit data centers in the UAE causing sparks and fire.

Benito (108:30)

1984.

Leo Laporte (108:32)

Yeah. Customers in Bahrain and the UAE began to report outages on Amazon's ME Central 1 region. So yeah, it's not a good place to be right now. Now around 4:30 p am PST. I don't know why they're doing PST, but they are One of our availability zones was impacted by objects that struck the data center creating sparks and fire. The fire department shut off power to the facility and generators as they worked to put out the fire. We're still waiting permission to turn the power back on. Millions and millions of dollars poured into these data centers. They're maybe now having some second thoughts. We are going to be covering the GTC keynote. You're going to join me, right Jeff?

Jeff Jarvis (109:21)

You bet your bippy.

Leo Laporte (109:22)

I'm very excited.

Jeff Jarvis (109:23)

Connoisseur of not your bippy Wong.

Paris Martineau (109:26)

He needs that.

Leo Laporte (109:27)

Jensen's one's leather jacket is available on Teemo for $8.53.

Jeff Jarvis (109:34)

How much do you think those things Cost.

Leo Laporte (109:36)

Oh, thousands, right?

Benito (109:37)

15,000.

Leo Laporte (109:38)

Yeah. He's not going to buy a cheap. Yeah. Not going to buy a cheap leather jacket. I was looking at motorcycle jackets. I thought it'd be kind of fun to have a, you know, kind of nice leather. I know. Because it's so me, isn't it?

Paris Martineau (109:53)

You should go. Custom would be my.

Leo Laporte (109:56)

Well, exactly what I was going to do is good.

Paris Martineau (109:58)

A shot.

Leo Laporte (109:59)

There's a company that makes leather motorcycle jackets. It's in the classic styles, Fox Creek leather. And so you can get like a very, you know, like you could look like the Fonz, like Fonzarelli if you want, or whatever you want. So I was looking at these. These are actually nice.

Paris Martineau (110:21)

Those are pretty cheap in comparison to.

Leo Laporte (110:22)

Yeah. They're not too old. Yeah. And they're handmade.

Paris Martineau (110:25)

Go to shots or. I don't know how to pronounce it. Yeah, I think it is shots in New York. When you come here, you should go there.

Leo Laporte (110:31)

That was the other place they recommend. Recommended on. On Reddit.

Paris Martineau (110:34)

Yeah, Shot nyc. I'd really recommend that. It's probably the best quality leather jackets and you could try them on in store. I mean for a kind of motorcycle leather jacket. And it's good. The reason I know all this is I spent way too much time researching what leather jacket to get and came up.

Leo Laporte (110:52)

I did. Look at this.

Paris Martineau (110:54)

This is an expensive one.

Leo Laporte (110:55)

Yeah.

Paris Martineau (110:56)

I mean because if you're getting a leather jacket, it's going to be expensive because you need large scale pieces, pieces of leather that are not being mixed together and are actually cut for your size. It's even harder to get one if you're a woman because all of the shot ones are lovely, but they are not made for people who have hips.

Leo Laporte (111:17)

Oh, you can't get them to, to fit your hips.

Paris Martineau (111:22)

If you do, then the arms are too long.

Leo Laporte (111:26)

Oh, I see. Sexy. So they're not. It's not.

Paris Martineau (111:28)

They're not made. They're not tailored for women.

Leo Laporte (111:30)

Yeah.

Paris Martineau (111:30)

In. In a way that's very annoying. And I'm like, if I'm gonna spend over a thousand dollars on another jacket.

Leo Laporte (111:35)

They are. They're all correctly.

Jeff Jarvis (111:37)

He wore. He wears Tom Ford, which run around 10,000. I put it in the chat.

Leo Laporte (111:42)

Oh, there you go. That's a designer.

Jeff Jarvis (111:44)

Very nice.

Leo Laporte (111:46)

I guess I don't. I never thought they looked that good on him. Anyway, getting back to the. The topic of at hand, we will be covering the GTC keynote which is going to be, I think, very interesting this year. There are at least two big Announcements that he's going to make. One is something called Nemo Claw, which is an open Claw style agent platform.

Jeff Jarvis (112:16)

Open. Open source, they say. Say. Or open access.

Leo Laporte (112:20)

Yeah, you know, and safer.

Jeff Jarvis (112:23)

That's the point.

Leo Laporte (112:24)

Yeah. And I, I would be very interested in that if, if NID is doing it. But of course It'll require a 5080 or some sort of. Or a Vera Rubin or some sort of fancy hardware.

Jeff Jarvis (112:35)

It's intended for companies, not for you. Yeah.

Leo Laporte (112:37)

Right. So, yeah, they're going to do their own agent, which I think we already know is going to be called Nemo Clock.

Jeff Jarvis (112:45)

Right. Yeah. Because it's associated with Nemo. What's their other Nemo product? Nematron.

Leo Laporte (112:54)

Nematron, okay.

Jeff Jarvis (112:55)

Which is their AI infrastructure platform. No, that's Nebius. I'm so confused.

Leo Laporte (113:02)

So according.

Jeff Jarvis (113:03)

That's their foundation model, right?

Leo Laporte (113:05)

According to. Yeah. So it'll be running on their model. Okay, that's interesting. Although they say it's just going to be. It's going to embrace open source AI model models.

Jeff Jarvis (113:14)

Well, as an agent, I wonder whether you build it there and then it runs elsewhere, that kind of stuff. I mean, that's where you don't need to. Once you build something with the agent, you don't necessarily. With the model, you don't necessarily need to use the model ongoing.

Leo Laporte (113:26)

Well, this will be. So Jeff, we will watch this with interest on Monday. We'll be streaming it because. Are you going to require cuda? Are you going to require Nvidia hardware or is it going to be open models? On open hardware that I could run, for instance, on my framework work? That would be fantastic. They also are expected to announce new hardware as well.

Jeff Jarvis (113:48)

New chip.

Leo Laporte (113:50)

Yeah, well, I'm trying to find the story on that, but I don't. I didn't put it in here.

Jeff Jarvis (113:54)

I guess the great thing about, about a Jensen Wong keynote, and they go more than two hours as a rule, is that he also has subtly an educational theme to it. He's explaining tokens, he's explaining scale, he's explaining digital twins.

Leo Laporte (114:10)

Right.

Jeff Jarvis (114:10)

That comes along for the ride and he's just superb at explaining it. At the same time, he gets across all these, you know, brands and specs that, you know, five people understand but that matter.

Leo Laporte (114:22)

Darren says they've said it will not require Cuda, which would be a big deal.

Jeff Jarvis (114:26)

Yes.

Leo Laporte (114:27)

Because that's their, that's their proprietary law.

Jeff Jarvis (114:29)

I think this is them kind of branching into.

Leo Laporte (114:31)

Yeah.

Jeff Jarvis (114:32)

Did you see the Kickstarter, Leo? I put it up. I'm Curious how it compares with your thing.

Leo Laporte (114:37)

Did I put it down here on chain Agentic management for Solana? No, this can't be it.

Jeff Jarvis (114:45)

Did you see the tiny AI pocket lab line 176?

Leo Laporte (114:51)

If it's tiny, it can't be that good.

Jeff Jarvis (114:53)

Well spelled. T I I N Y. Oh, tinny. Well, those two eyes. Not two.

Leo Laporte (114:59)

Tiny. Tiny. The world's first pocket AI supercomputer. What? It says it can run 120 gigabyte LLM local.

Jeff Jarvis (115:10)

I wanted to hear what you thought about this.

Leo Laporte (115:12)

Huh. This is for. Everybody's jumping on this openclaw, trying to figure out how they can make some money. What is the processor in here that. It must have a lot of ram.

Jeff Jarvis (115:21)

You got to go way down to see this.

Leo Laporte (115:23)

Doesn't seem very WikiLeaks likely. Probably. Yeah. So it's $1,399. Does it even include RAM that talks?

Benito (115:35)

Yeah, that just talks to a server. Right.

Jeff Jarvis (115:38)

The CPU is an ARM version 9.2 plus an NPU 30 int 8 tops. I don't know what that means.

Leo Laporte (115:46)

It comes with a terabyte SSD 80 tops memory. 30 tops is not super fast. Oh yeah. It looks like 32 gigs or 48 gigs of RAM. It's not going to be as good as my framework. I already got a framework. I'm half.

Jeff Jarvis (116:05)

Yeah. I'm just curious for.

Leo Laporte (116:07)

What this is for is it's half

Benito (116:09)

the cost of your framework though.

Leo Laporte (116:11)

Well, yeah. And it's, you know, smaller than a Mac Studio, but it's not cheaper than a Mac Mini. If you're using a server model, it doesn't matter how much processor you have. The same. This is. This sounds like they're pushing it for 80 gigs of RAM. Oh yeah.

Jeff Jarvis (116:27)

No, the video that they do promoting it, they picture the person going to the client and says, well, this isn't going to go up on the cloud, is it? No, it's all right here.

Leo Laporte (116:37)

Yeah. You can run GPT120B at 21 tokens per second, which is not, you know, that's tolerable QN3 at almost 30 tokens per second.

Benito (116:50)

Second.

Jeff Jarvis (116:51)

Huh?

Leo Laporte (116:52)

And you keep it in your pocket?

Jeff Jarvis (116:53)

Yeah, I've got a super computer computer in my pocket. There's a.

Leo Laporte (116:58)

And it works with open claw. Well, I think we're gonna. Like I said, this is gonna be the year. The year of the AI agent of the. Of the loop. So it'd be very interesting to see. There'll be a lot of companies. I think Nvidia's product will be very interesting. We'll look forward to seeing that. You're watching intelligent machines. Let's take a little tiny time out so that Jeff can ice his butt.

Jeff Jarvis (117:25)

I did that the last time around.

Leo Laporte (117:26)

Yes, well, you can never ice your butt too many times. Paris can trash me and Scrabble and we'll continue.

Paris Martineau (117:34)

You know, someone's hopped in the chat and asked me about my Scrabble play style and different things like that, and I'm, like, locked in.

Leo Laporte (117:42)

Do you have a style?

Paris Martineau (117:44)

Well, we're. We're more talking about the fact that someone asked me if I play against the computer in crossplay and how often I beat it. The answer is every time I beat it and I play it on a hard. But I think that the cross computer is not hard. Like, it's. It's. I would say it's like a very unoptimized player. I. I think I'm also particularly good at this because before Go to, the only game I had on my phone for the last couple of years was this thing called Classic Words that is just a. You play against a computer and you can do it offline. So I'd play it in the subway whenever I had no cell phone service. And I'd play on the extremely hard mode, which, despite what the name says, not that hard. But it's certainly harder than hard you

Leo Laporte (118:28)

to whoop my butt.

Paris Martineau (118:30)

It did.

Leo Laporte (118:32)

I'm probably imagining this, but. But I feel like I can through the screen, feel your either joy at crushing me.

Paris Martineau (118:47)

It's true. I do. I do feel whenever I get a really big one, I'm like, oh, it's going to go mad.

Leo Laporte (118:53)

I thought I could sense that, and then I could also. I feel like if I play a good word that, like, I could sense a little bit of frustration.

Paris Martineau (119:01)

I was. I screamed. I. I, like, went out and went, no, whenever, whatever move it was of our last game because I had set myself up to play a Z or an S there. Where you did. And you stole mine and got my 50 points that I needed to win the last move. And I was furious. But then I was like, paris, you must internalize this. Never set up a move unless you've counted the tiles and realize that Leo doesn't have the tile.

Leo Laporte (119:29)

You are.

Paris Martineau (119:30)

I must. I must get.

Leo Laporte (119:31)

I have no idea that you. You were such a shark. I've just been kind of playing this.

Paris Martineau (119:34)

I frankly didn't realize either.

Leo Laporte (119:37)

I'm gonna get serious now. Watch out.

Paris Martineau (119:39)

I was never really into. I mean, I played with words with Friends casually when it came out, but it was not a big deal. I've gotten really into scramble over the last couple of years because I play it with my mom whenever I go home to visit her.

Leo Laporte (119:49)

And then this is the thing I don't like about these computerized scramble me.

Paris Martineau (119:53)

My mom are vicious.

Jeff Jarvis (119:54)

She will be.

Paris Martineau (119:55)

She will be. I, she will like, we'll sit there, there for like five minutes looking at the board and then all place and then she'd be like, how dare you. That was my time. I was gonna play that.

Jeff Jarvis (120:05)

Are you playing this version? This version with her?

Leo Laporte (120:08)

No, they're playing real.

Paris Martineau (120:09)

I don't think she. I don't think she's in. Maybe I could get her to subscribe to New York Times games.

Leo Laporte (120:13)

Not as good. I'll tell you why all computer Scrabble has the same problem, which is you can try a word to see if it's a real word and then you. So you get to test stuff. Well, in real Scrabble you either play it or you don't play it. And you can. In real Scrabble you can bluff. You could put down a word that's plausible and you're playing.

Paris Martineau (120:32)

Oh, not in my household. If someone thinks it's wrong, they'll look

Leo Laporte (120:37)

it up and it's no, no, they, no. But they have to challenge you and they'll lose a turn if they're wrong. Oh yeah, play the real

Paris Martineau (120:46)

rules.

Leo Laporte (120:46)

That's the real rules. And that's a different, very different game because you can block tough.

Jeff Jarvis (120:51)

Yeah.

Benito (120:52)

There's an extra strategic layer to that game.

Leo Laporte (120:54)

And it's.

Jeff Jarvis (120:55)

Look at Paris right now.

Paris Martineau (120:58)

Because I know the big part of me getting into. I know a lot of words. And a big part of my crossplay addiction lately has been I've been Scrabble maxing, which involves kind of memorizing all the two and three letter words to start and then starting to memorize like the first 100 most common bingos, which are like where you play your entire thing and so on and so forth and the beginnings and end.

Leo Laporte (121:22)

You're getting serious. And you know the going to be tournamenting soon. We have at home a rotating Scrabble board, one of the deluxe Scrabble boards that rotates so you can.

Paris Martineau (121:31)

Oh, that would be so convenient because I often put it upside down. I often play my mom at a handicap because I'm like it, you know, I need to be on the, on the side where you can't look at it straight up.

Leo Laporte (121:41)

You need the rotating board, real wood tiles and all that's really the thing

Paris Martineau (121:46)

that's crazy about Crossplay, which will make a my mother upset, is if you go out, if you finish the bag, the other player isn't deducted the points, the points of the tiles, which is crazy.

Leo Laporte (122:01)

That's another thing that's different. I think the biggest one though is that you can try words and figure out if that's a real word.

Paris Martineau (122:06)

And there's a dictionary in the app.

Leo Laporte (122:08)

There is.

Paris Martineau (122:11)

And you can look up words.

Leo Laporte (122:12)

I didn't know that.

Paris Martineau (122:13)

I'm going to say you can only.

Leo Laporte (122:16)

Yeah, I didn't realize you. You were undefeated.

Paris Martineau (122:20)

Well, until the. I have one loss and 10 wins

Leo Laporte (122:24)

and then you clobbered me. The next game you beat me by like 100 points. That was the one where I accidentally pressed play on one letter. All right, we really got to do this. This episode of Intelligent Machines brought to you by Melissa, the trusted data quality expert since 1985. Bad data could be costing your business. Melissa specializes in making data clean, accurate and of peak quality. They've been doing it for 41 years now, combining expertise with cutting edge AI. Here's what that means for your business. You get address verification, of course, they've been doing this for 41 years. They validate addresses globally. Of course now they can do it in real time time in every country practically of the world in a form that is appropriate for that country, which really helps reducing failed deliveries. They also can do mobile identity verification. As we've gone more and more to using our phones for financial transactions, this becomes really important for know your customer regulations, things like that. Melissa will match customers to mobile numbers, which really cuts down fraud. But it also has another very good benefit. It opens SMS channels. Yeah. So you could talk back and forth with your customer. Change of address information is automatically kept up to date. Your customers will automatically be updated when they move. How about that? And something classically difficult to do, AI powered deduplication. On average, a database contains 8 to 10% duplicate records. Melissa's powerful matchup technology can identify non exact matching duplicates. Duplicate records, merge them, get rid of them. So you only have one record for every customer. Customer data intelligence too will let you enrich every record with demographic, geographic, psychographic, firmographic and property info. For more intelligent targeting, the new Melissa Alert service will monitor and automatically update your customer data. Not just for moves, but address changes, property transactions, hazard risks and more clean data leads to better marketing, roi, higher customer lifetime value and AI that works as intended. If you're using AI for personalization, or predictive analytics. Garbage data, as you know, means garbage results. Melissa ensures your AI learns from accurate information. When it comes to data, Melissa's got you covered. Look at Etoro, the social investing platform. 23 million users users. They use Melissa for identity verification. Their business analyst said, quote, we find electronic verification is the way to go because it makes the user's life easier. Users register faster and they can start using our platform right away, end quote. Whether you're a small business just getting started or an enterprise managing millions of records, Melissa scales with you easy to use apps for Salesforce, Dynamics, CRM, Shopify, Stripe, Microsoft Office, Google Docs, and on and on and on. Melissa's APIs integrate seamlessly into your existing workflows or your custom builds. And Melissa's solutions, of course, are GDPR and CCPA compliant. They're FedRamp and ISO 27001 certified. They meet SOC2 and HIPAA high trust standards for information security management. They do it right. Get started today with 1000 records cleaned for free at melissa.com twit that's Melissa. We thank them so much for their support of intelligent machines. So just a reminder, we will be in the club doing the event, the GTC keynote. Jensen Huang's going to do that. Actually, we got a lot of club stuff coming up. GTC's keynote is 11am Pacific, 2pm Eastern this Monday.

Jeff Jarvis (126:19)

What is that? UTC, Lee?

Leo Laporte (126:21)

The UTC, it's 1800 and that's important because we are now on daylight saving time and much of the world is not.

Jeff Jarvis (126:28)

Mike Elgin joined us on AI Inside today, but about 20 minutes late because he's in Mexico and he.

Leo Laporte (126:34)

Yeah, they don't change.

Paris Martineau (126:36)

Yeah, it was really beautiful because this change meant that today we started the show and it was so bright outside that I had to put my blinds down.

Leo Laporte (126:46)

It's dark when we start the show time.

Paris Martineau (126:48)

It is pretty nice. This is also coincided with a weird week in New York City where It's been the 70s, weather wise. And then next week it should go back to 30. We just had, we just had 18 inches of snow a coup like two weeks ago. Now it's 75 degrees and then not all the snow has melted yet though, which is very funny. And next week it's gonna be back to 30.

Jeff Jarvis (127:11)

Kardian has a story about the snow mountain in Toronto from plows, all kinds. Well, no, they make these huge, these mountains of snow and they're obviously they're, they're very off white because they contain oil and, and salt.

Paris Martineau (127:28)

There was a perfect Story in Gothamist a couple weeks ago called. We tested those gross piles of snow on New York City sidewalks. Here's what's in them.

Leo Laporte (127:39)

Probably pizza rats, right?

Paris Martineau (127:41)

It's all pizza. It's pizza Red all the way down, actually.

Leo Laporte (127:44)

Lead.

Paris Martineau (127:45)

Turns out snow underneath tracks has led at a whopping 279 parts per billion, which is quite high.

Leo Laporte (127:57)

Lead. Where's that coming from?

Paris Martineau (128:00)

Industrial runoff as well as you know.

Leo Laporte (128:03)

So do not eat the toxic. Don't eat the.

Paris Martineau (128:05)

The snow.

Leo Laporte (128:06)

No, no.

Jeff Jarvis (128:07)

Don't eat yellow snow or snowman or

Leo Laporte (128:10)

gray snow or brown snow or it

Jeff Jarvis (128:12)

melts and all the salt goes into the water supply.

Leo Laporte (128:16)

I wonder if the lead's in the salt. Like, the salt isn't exactly pure, you know,

Paris Martineau (128:22)

I mean, I think part of it is they shed from the elevated rail lines, which are kind of notorious for shedding lead paint. And then the water. Water flows down that.

Benito (128:33)

The snow plots are probably diesel, too.

Leo Laporte (128:36)

Yeah. This. This is. This is why I love it when the snow first falls in New York City and for about 10 minutes. For 10 minutes. And then it's a hellscape. After that, it's slushy.

Paris Martineau (128:47)

Especially right now. You haven't experienced the New York City snow post Covid, where everybody and their mother and their brother got a dog, and their dog has crippling anxiety, and so it's covered in dog crap.

Leo Laporte (129:03)

Oh, God.

Jeff Jarvis (129:05)

Then there's this. Putting it in the chat. Some people took the. As is New York's one. People take the snow and then they build beautiful things out of it. Oh, I just put it in Instagram.

Leo Laporte (129:15)

How beautiful is it?

Jeff Jarvis (129:16)

It's nice. It's nice.

Paris Martineau (129:18)

Wait a second. I've got a beautiful snow thing going on.

Leo Laporte (129:22)

New Yorkers create realistic snow art after blizzard. Oh, look at the.

Jeff Jarvis (129:26)

That.

Leo Laporte (129:27)

That's really good now. I hope they wore gloves when they did it.

Jeff Jarvis (129:31)

Yeah.

Leo Laporte (129:32)

That's a man's head created by a local architect, Jeff Zaborski.

Paris Martineau (129:35)

Got some links.

Leo Laporte (129:36)

There's some bears. Nice. That's much better than a snow bear. Those are architectural masterpieces. Polar bears.

Jeff Jarvis (129:44)

Yeah.

Leo Laporte (129:45)

There's a person sitting on a bench.

Jeff Jarvis (129:48)

Well, that looks like somebody who came out of.

Leo Laporte (129:52)

Somebody was sitting there. That actually is a human inside the snow.

Jeff Jarvis (129:55)

Yes.

Leo Laporte (129:56)

Yeah. Well, all right. Okay. Let's see. Back to the AI. How do you talk to somebody experiencing AI psychosis? I thought this would be useful for you guys talking to me. To be honest.

Paris Martineau (130:17)

We. We use our separate side chat for that.

Jeff Jarvis (130:20)

Yeah.

Leo Laporte (130:22)

When David saw. This is before of 4 Media. When David saw his Friend Michael's social media post asking for a second opinion on a programming project. He offered to take a look. He sent me some of the code and none of it made sense. None of it ran correctly, or if it did run, it didn't do anything. David told me. So I'm like, what is this? Can you give me some more context about this? And Michael's like, oh yeah, I've been messing around with chatgpt a lot. Michael then sent David thousands of pages of ChatGPT conversations, much of it lines of code that didn't work, interspersed in the code musings about spirituality and quantum physics, tetrahedral structures, base particles, not tetrahedral structures, tetrahedral structures and multi dimensional interactions. It's very like, woo, woo. David told me and we ended up having this interesting conversation about how do you know that Chat GPT isn't, isn't lying? Okay, unfortunately this post is for paid members only, so I can only tell you the problem, not the solution.

Jeff Jarvis (131:32)

I'm curious what Paris thinks of the Grammarly kerfuffle this week.

Paris Martineau (131:37)

I mean, it's crazy and I'm somewhat unsurprised. Do you want to give everybody the contest context? And have you seen the recent resolution that has happened since we've started recording?

Leo Laporte (131:46)

Yeah. So Cory Doctorow was on Sunday and he's one of the authors that Grammarly pretends to use as an authority. When you use Grammarly, you can ask it to rate your writing based on writing from people like Lauren Good and Cashmere Hill, Nilay Patel from the Verge and Cory Doctorow. I said, corey, did they ask you at a time? He said, said, no. Now he had an interesting take. He wasn't really upset, he just said it's nonsense. He said, because he had the wonderful phrase and I can't duplicate it exactly. But he said, writing is a noumenal experience. You're drawing on some liminal thoughts and it's a creative human process. No AI can ever do it. He's not an AI fan.

Benito (132:36)

He basically means it's like lossy going from your thought to. To a prompt is there's a loss

Leo Laporte (132:41)

kind of garbage did you do? Anyway, Grammarly says we're not going to ask permission. We didn't need to ask permission, but I guess if you want us to take your name off the list, we can.

Jeff Jarvis (132:51)

Neiman Lab did a version and they've

Paris Martineau (132:53)

since been sued by a journalist who is involved in this. And I believe since we started recording this show, Grammarly has decided to take it down.

Leo Laporte (133:02)

Ah, they Were getting a lot of.

Jeff Jarvis (133:04)

But another level. If you say. If I say critique my piece and if it came along and said, well, famous reporter Paris Martineau has written this, then it's like, search, Right? It's just. It went off and searched and found something. It's. The problem with this is it made it seem as if the authors were.

Paris Martineau (133:24)

Yeah, a little pop up. Would say like it was a comment from Stephen King saying this.

Benito (133:29)

That's the problem.

Leo Laporte (133:30)

Patel would write this headline this way,

Jeff Jarvis (133:33)

and then it would link to an article as a source saying that. Which is kind of. Okay. Demon Lab did it with journalists too. And tons of journalists were in their journalism professors. I'm just pissed I wasn't included.

Leo Laporte (133:44)

Did you do the quiz on the New York Times? AI or human? No, I did. Okay. How did you do?

Paris Martineau (133:52)

I think I got almost all of them. Human.

Leo Laporte (133:55)

I got all but one. I missed one. One? Yeah. Partly because I kind of recognized some of the samples.

Paris Martineau (134:02)

I mean, they were classic samples.

Leo Laporte (134:04)

Yeah. Cormac McCarthy or Lake Le Guin. There was one from Wolf Hall. Actually, that's ironically the one I got wrong was the book I just finished. But I think they say that 57% of the New York Times readers prefer the AI version.

Paris Martineau (134:23)

That's deeply sad.

Leo Laporte (134:24)

Here, I'll give you. Is this AI or human? You must not change. Oh, you must not change. One thing, one pebble, one grain of sand, until you know what good and evil will follow on that act. AI Yeah, I think that's AI Obviously cliched.

Jeff Jarvis (134:40)

Trite.

Leo Laporte (134:41)

How about this? The healers teach that every remedy extracts its cost. A fever brought down will rise again. Somewhere. Somewhere. Which one of those is AI? Wait, this was an Ursula K. No, that one. The first one was Ursula K. First one.

Paris Martineau (135:01)

Yeah. Because the. If you read the full quote, it then says, the world is in balance. In equilibrium. Title case. A Wizard's Power of Changing. Changing capitalized. And if summoning can shake the balance of the world. Like that's classic sort of stuff. No.

Leo Laporte (135:17)

Okay. Yeah, yeah, yeah. Cover of Time magazine apparently. Just came out. We've got it from Scooter X. The war for AI Dario Emoti on the front, along with Pete Hegseth. Also the assistant undersecretary for the Department of Defense, the former uber negotiator. I can't remember his first name. Mr. Michael. Who has been browbeating Dario about all of this. I don't know who the two.

Jeff Jarvis (135:47)

Who's the other person there? Was this. Is that his sister? Is that Dario's Sister. I didn't know.

Leo Laporte (135:51)

Oh, yeah, they both started.

Jeff Jarvis (135:53)

But who's the woman? Who's the. With red hair?

Leo Laporte (135:55)

I don't know who the redhead is.

Jeff Jarvis (135:56)

That's weird to find yourself on the COVID of Time magazine and nobody knows who you are.

Leo Laporte (136:00)

Well, I'm sure if you open the magazine, you would know. By the way, just spotted this is kind of remarkable. In Central Park Park, a new snow sculpture you might recognize.

Jeff Jarvis (136:15)

Hey, I've lost £25 with this illness.

Paris Martineau (136:18)

You know something I think it's really incredible about this is that it has man spreading in the snow sculpture.

Leo Laporte (136:24)

Yes, the men. The men are. So this is three people on a bench. Paris, me and Jeff, the Snowcast trio. And yes, your man's.

Paris Martineau (136:34)

My legs are put close.

Leo Laporte (136:36)

Your legs are properly closed?

Paris Martineau (136:38)

Yeah, theirs are wide open.

Leo Laporte (136:40)

Yes. Well, we just need that extra room. I'm just saying, I'm just.

Paris Martineau (136:45)

For all of your snow junk. For all your snowballs.

Jeff Jarvis (136:51)

Shrinkage.

Leo Laporte (136:52)

I'm just saying, I'm just, just saying, you know, I'm very self conscious when I'm on the subway or a bus and I really try to sit like as small as possible, but it's not that easy for a guy. Sorry. Try. Oh my. There are so many stories. AI art cannot be copyrighted. The Supreme Court declined to overturn a lower court ruling that AI generated art, having no human hand involved, cannot be copyrighted.

Jeff Jarvis (137:26)

This case goes back pretty early on.

Leo Laporte (137:28)

Yeah, it's been going on for years. What do you think? I mean, I think that seems right.

Paris Martineau (137:33)

Yeah. Who would hold the copyright?

Leo Laporte (137:36)

Stephen Thaler, a computer scientist from Missouri, some years ago, using not modern models, generated this image, a recent entrance to paradise, and tried to copyright it. In 2019, the Copyright Office. Well, why? That's a good question. Probably because if you copyright it, then you can sell it or something. I don't know.

Benito (137:58)

I think it's slippery. What is the definition of AI generated?

Leo Laporte (138:05)

Well, the prompt can be copyrighted.

Benito (138:08)

So if it's something holy made prompted that piece that you prompt out, that can't be copyrighted. But if you do something to that, can you?

Leo Laporte (138:18)

Yeah, I think if you modify. If the human hand is involved in the actual image, I think you can.

Jeff Jarvis (138:23)

No, it has to be created by the human was the issue. Because the argument was that copyright was there to protect humans, even though that's not true.

Benito (138:29)

So you can't sample AI then that means you can't copyright anything that you use. If you sample anything from AI, you can't use it. That you can't Copyright in.

Leo Laporte (138:39)

No, I believe you can't, because now you have a human hand touching it. I think that's okay.

Benito (138:43)

So that goes back to my original question then. Like, if I just apply one filter onto this image, have I done enough to copyright it?

Leo Laporte (138:51)

Well, they'll. This is something the court would have to decide is what is sufficient. Right.

Benito (138:56)

And I don't think that is way too slippery. And you're never going to be able to define.

Jeff Jarvis (138:59)

Find that.

Leo Laporte (139:01)

Well, they do it all the time. That's what these guys do. They get paid, they get nice robes. Some of them get RVs.

Paris Martineau (139:08)

You're right.

Jeff Jarvis (139:09)

It's for the lawyers.

Leo Laporte (139:09)

It's for the lawyers. It's their job. It's what they do for a living. Can't knock that. Thaler actually also tried the case in the UK and they also rejected it. So I guess there is agreement.

Jeff Jarvis (139:31)

It was the Copyright Office, as I remember, who turned him down. And the court just upheld that right.

Leo Laporte (139:35)

They upheld. Well, but he appealed to the.

Jeff Jarvis (139:38)

Through a Supreme Court.

Leo Laporte (139:39)

He appealed it all the way up to the Supreme Court. Yeah. This story from Yahoo, I guess. No, I'm sorry. From the Los Angeles Times. Within the last month, two US Judges have declared AI bots are not human. You'll be glad to know they're not human. I guess this is kind of related to the copyright issue. Bradley Hepner was indicted for. For. By a federal grand jury for looting $150 million from a financial services company he chaired. He pleaded innocent, was released on bail. Knowing an indictment was in the offing, he had consulted Claude for help on a defense strategy. Strategy. His lawyers asserted that those exchanges, which were set forth in written memos, were tantamount to consultations with actual lawyers, therefore protected by attorney client privilege, and couldn't be used against him in court. It turned out that the information that he gave Claude was, in fact, very incriminating, shall we say. And the federal judge said. Said the AI documents were not communications between Hepner and his attorneys. Claude's not an attorney. He also said they're not confidential because in its terms of use, anthropic claims the right to collect the user's queries and responses for training. And you should make a note of this to disclose them to others. Finally, he wasn't asking Claude for legal advice, but for information he could give his lawyer lawyers, whether he could pass it on to his lawyers or not. And in fact, when prosecutors tested Claude, asking whether it could give legal advice, the bot said, no, you have to consult with A qualified attorney. So the judge said, yeah, no, it's not covered by client attorney privilege. So just. I think that's important for everybody to know that. Beware there to ask Claude some questions. They fired that Ars Technica reporter who used quotes fabricated by an AI. He's out of work. Ben Edwards, no longer working in Ars Technica. I think that's probably the right thing to do. Yes.

Paris Martineau (142:01)

Yeah. What sort of quotes did he fire?

Leo Laporte (142:03)

Oh, you don't remember this story?

Jeff Jarvis (142:05)

We talked about it here. I thought we did. You were gone this week. That week.

Leo Laporte (142:08)

Maybe you weren't here.

Jeff Jarvis (142:09)

Yeah, yeah, actually you weren't. You weren't.

Leo Laporte (142:12)

This was on this. This was on the story about the guy who ran an open source project who got a PR from an AI who rejected the PR because it was written by an AI. And the AI said, hey, you, you can't do that. And then wrote a blog post. The AI wrote a blog post. Slam. It actually was a good conversation. We read some of the back and forth from the. The GitHub repository, which I thought actually was pretty well done and judicious on both sides. The AI later apologized, but the reporter apparently fabricated or used an AI that was fabricating quotations attributed to a source that didn't say them and, and character. And ours pulled the article down, characterized it as a serious failure of our standards. Ours has said the error appeared to be an isolated incident, and I think that's probably the case. Ours is pretty good.

Paris Martineau (143:15)

Edwards said that he was sick at the time and quote, while working from bed with a fever and very little sleep, he unintentionally made a serious journalistic error as he attempted to use an, quote, experimental Claude code based AI tool to help him extract relevant verbatim source

Jeff Jarvis (143:33)

material and make my damn quota, even though I'm sick.

Leo Laporte (143:37)

Yeah, I should have taken a sick day because in the course of that interaction, I inadvertently ended up with a paraphrased version of Shambha's words rather than his actual words. Can I get sued for doing.

Jeff Jarvis (143:51)

Yeah, I think you might.

Leo Laporte (143:52)

Okay. He was not drunk. He was sick and feverish. I apologize, guys in advance, but he shouldn't have done it. And although, you know, I've read, I believe I've read a lot of Benj's work, and I think he's quite good.

Paris Martineau (144:06)

So the irony of an AI reporter being tripped up by AI's hallucination is not lost on me. Yeah, I take accuracy in my work very seriously, and this is a painful failure on my part.

Leo Laporte (144:17)

Yeah, I, I I think does not reflect poorly on bench. Well, it does, but, I mean, I think he took responsibility.

Jeff Jarvis (144:23)

The question is, would you hire Ben having.

Leo Laporte (144:26)

I would. He's not going to do that again.

Jeff Jarvis (144:28)

He's not gonna do it again. Exactly.

Leo Laporte (144:30)

So I think that's. I don't think that that's.

Jeff Jarvis (144:33)

I agree.

Leo Laporte (144:33)

But I understand why they felt like they had a farm. Now. Meanwhile, the Cleveland plane dealer, which has a best name ever, it is. Has decided. Huh.

Jeff Jarvis (144:43)

I used to work with them. I started Cleveland. I started cleveland.com with them.

Leo Laporte (144:46)

Oh. They have begun to feature a new byline on recent articles about an ice carving festival, a medical research discovery, and a roaming pack of chicken slaying dogs. A reporter's name is paired with the words advanced local express desk, which means apparently the article was drafted by AI

Jeff Jarvis (145:07)

So the argument is. I wrote about this. The argument that Quinn makes, the editor is that they can get more reporting done. And I was talking with a class at Montclair State this week, and I said. I explained to them my job when I worked at the Chicago Tribunal was called rewrite. Reporters would call in the notes, and I would look up the clips, and I would call sources, and then I would write a paragraph at a time on deadline, making a story. When I worked at Time, Inc. Correspondence would send in 40 pages of notes, and I would turn into a little tiny story. So, in essence, the AI is doing the job that we called rewrite. And the argument that Quinn, the editor, makes is that he gets more reporting as a result. And that's what we want, is more reporting. And so if it's checked properly and worked properly, if the reporter sends in reporting notes and it gets turned into a story, what's the problem? And my argument was, yeah, I think he has a point. The only thing I would argue is, why turn it all into an article? Why not think about other forms that the AI can turn it into that may be just as valuable now that we can create more things.

Leo Laporte (146:07)

Yeah. So what do you think then about the fresh out of college job applicant who withdrew Drew from a reporting fellowship with the Cleveland Plain Dealer when they found out the position didn't include writing, just filing notes to an AI writing tool. In other words, giving it the notes and then letting the AI write the article.

Jeff Jarvis (146:25)

Little snot.

Leo Laporte (146:29)

Professor Jarvis says you're a snot.

Jeff Jarvis (146:33)

Paris, what do you think about this?

Paris Martineau (146:36)

Can you repeat the question?

Leo Laporte (146:37)

So a kid out of college plane.

Paris Martineau (146:42)

I was reading.

Leo Laporte (146:43)

Did you just say a mean word on my a little as fresh out of college job Applicant had applied for a reporting fellowship but then found out that the position did not include writing but going out doing reportage and then giving your notes to an AI to write the final article. Kind of like a copy editor except it's an AI.

Paris Martineau (147:03)

I think that's messed up that.

Leo Laporte (147:06)

So you agree with the student withdrawing.

Paris Martineau (147:09)

Yeah, I think that you should make that very clear from the first job interview and it should not be surprising to the person taking that job.

Jeff Jarvis (147:18)

But back in the day, since I'm old guy here, back in the day there were many jobs on newspapers that were reported only.

Paris Martineau (147:24)

Yes, no, I, I know that. But one again like, like what we've talked before with Anthropic and even this Edwards story. We just talked about that when the people doing those rewrites are human, you have someone you can blame. In this case, if this AI is going to publish something under this person's byline that ends up being wrong, they're the one on the hook probably for it or their reputation.

Jeff Jarvis (147:51)

They should get a chance to look at it and.

Paris Martineau (147:54)

But I also think if you're hiring an early career journalist the clear they're clearly literally taking this job because they think it's going to help them advance their career and in as an early career fresher to college journalist that means getting bylines and getting experience writing. And if you are in a position where that's what you thought you're going to be doing and on day one of the job you realize it's not that something has gone terribly wrong.

Jeff Jarvis (148:17)

Well, I'll devils advocate you that we argue all the time that you shouldn't come into journalism to write. You should come journalism to report. The reporting is the most important part and valuable part of journalism. And so that's where the value is, the writing.

Leo Laporte (148:30)

It was a fellowship. It wasn't like a job, it was like a fellowship.

Paris Martineau (148:34)

I'd argue it's even more problematic because that's supposed to be like an internship. You're supposed to be giving someone experience.

Leo Laporte (148:42)

Right. Doing a job experience, collecting information for an AI which may be more likely the kind of job you might be

Jeff Jarvis (148:49)

getting in every correspondent attempt. Time Inc. Did not write the story. Never wrote a story. They sent really. They said in 30, 40 page. Literally 3040 pages.

Leo Laporte (148:56)

So it's a rewrite type. It's an AI based writer.

Jeff Jarvis (149:00)

Time Inc. I was, I was, I was writer when I was at Chicago Today, the paper that had no tomorrow.

Leo Laporte (149:04)

You get a byline when you do that?

Jeff Jarvis (149:06)

No, they didn't used to but they added them.

Leo Laporte (149:09)

Okay.

Jeff Jarvis (149:09)

Used to be time had no bylines. It was very, very much like I remember those days.

Leo Laporte (149:15)

Yeah.

Jeff Jarvis (149:15)

We had a guy at Chicago to today we call Bullet Bob Glass. He was a really good reporter. He had great contacts in the city government. He spoke out of the side of his mouth like you expect in Chicago. He would send in notes and notes and notes and, and, and the reason he was named Bullet Bob Glass is because you write up his stuff and he had good stuff. But at some point you couldn't figure out how to make a story out of it. So they would just say in other reporting, colon, bullet, bullet, bullet, Bullet. Because you couldn't make any sense sense of it. So we called him Bullet Bob Glass.

Leo Laporte (149:45)

A the Journal of the Canadian Pediatric Society, Pediatrics and Child Health has issued corrections on 138 cases that it's published over the last 25 years. To add the disclaimer, the cases described are fictional.

Paris Martineau (150:04)

Oh boy.

Leo Laporte (150:05)

The corrections come following a January article in a New Yorker magazine magazine that mentions that one of the stories was made up. The New Yorker article made public at admission by one of the co authors. Yeah, we just made that up. This doesn't have anything to do with AI. So I guess it's just another example of the pressure to fabricate. Doesn't require AI to do this.

Paris Martineau (150:34)

This.

Leo Laporte (150:39)

There is a really interesting debate going on in the open source community right now. We talked about this with Cory Doctorow and Joey de Villa on Twit as well. There is a Python library called Care Debt that you made a sound. Do you know that library? Are you a fan?

Jeff Jarvis (150:57)

I burped.

Leo Laporte (151:01)

She's passing number notes. You're burping. I don't know. Maybe we should just do a final ad and get the hell out of dodge. It is 2 hours and 30 seconds.

Paris Martineau (151:10)

It is three hours into the show.

Leo Laporte (151:12)

Not three hours. It's 8pm well, yeah, we. If. Yeah, okay. I guess you're right. All right, let me, let me take a break and then we'll do our picks and we'll. We'll get the heck out of Dodge. Now I did get an email from somebody last week saying don't make this a charge show. So grim. Leo, do you feel that this was a grim show? Honestly?

Jeff Jarvis (151:37)

Yeah, I think it was beamed at us because you're. You're Dr. Optimism.

Leo Laporte (151:42)

I think you're Dr. If I did more voices, it would be less grim. So I'm gonna do the. The rest of this show.

Paris Martineau (151:51)

I think that we should all talk in a funny Voice.

Leo Laporte (151:54)

Yeah, that sounds good.

Jeff Jarvis (151:56)

So let me ask you a question. I'm listening to a book right now which is really good, but I think you'd really, really like, which is called the Last Kings of Hollywood, about Spielberg and Lucas and Coppola. And it's really well written. It's really good. But the. The narrator, whenever it's a quote from Steven Spielberg, it does a. A wimpy voice like that, which is the.

Leo Laporte (152:21)

Not how Steven Spielberg talks.

Jeff Jarvis (152:23)

No. It's really irritating. Both Lucas and Spielberg get whidby voices.

Paris Martineau (152:32)

Yeah, I'm just imagining. Be like, oh, now I do the Steven Spielberg voice.

Leo Laporte (152:39)

We're going to use a puppet for the. For the E.T. what do you think?

Jeff Jarvis (152:43)

I listen to another book called the typewriter and the guillotine. And the. And. And the narrator in that would read something that was from a French person and would do it in a French accent. No, that's the way I read the book.

Leo Laporte (152:55)

I hate.

Jeff Jarvis (152:55)

No, I hate that.

Leo Laporte (152:56)

That's all right. Let me hear. This is a little bit of the narration. Let's see if he does it.

Jeff Jarvis (153:00)

This is audible.

Leo Laporte (153:02)

Oh, yeah. They're not going.

Jeff Jarvis (153:03)

You got to skip all of the

Paris Martineau (153:04)

first things read by Sean Taylor Corbett.

Leo Laporte (153:07)

Well, Sean sounds.

Paris Martineau (153:08)

You got to get to chapter one. You're in the preview right now.

Leo Laporte (153:12)

I know. That's all I'm getting is the music at the beginning.

Jeff Jarvis (153:15)

You got to skip at the end. And then it'll skip to the next.

Paris Martineau (153:16)

Click, pause, and then go to the table of contents of non fiction.

Jeff Jarvis (153:20)

There we go.

Paris Martineau (153:21)

Any passage presented.

Leo Laporte (153:22)

It sounds like this anyway.

Paris Martineau (153:23)

Is a direct quote from an interview written a third party. We got to click the table of contents, open that up, and go to a chat.

Leo Laporte (153:32)

You can't because it's a preview. They're not. Oh, maybe they. No. Yeah. See, I can only.

Paris Martineau (153:36)

Oh, oh, oh. It was right there.

Leo Laporte (153:41)

Oh, no. But it does look like a good book. I just like to say that. And audible. Probably be the best way to listen to it.

Jeff Jarvis (153:48)

Yeah, it's good. And I'm fine.

Leo Laporte (153:50)

I like Hollywood stories.

Jeff Jarvis (153:51)

And if I were Spielberg, what would you think? Oh, you gotta make.

Paris Martineau (153:56)

I would think that they clearly must not like movies.

Leo Laporte (154:02)

I want Liza Minnelli's new book. I hope she read reads. It probably doesn't, though. But she's this week released a biography of her life, which has been quite wild.

Paris Martineau (154:19)

And can you do Eliza Minnelli?

Leo Laporte (154:22)

No.

Jeff Jarvis (154:23)

Oh, she's narrator. She's narrator. Get a preview.

Leo Laporte (154:26)

Liza Melli, 79, lets loose on lovers, drugs and what Happened with Lady Gaga and the Oscars. In bombshell memoir, it's called Kids. Wait till you hear this. And it just came out yesterday.

Jeff Jarvis (154:39)

Yeah, you can. You can go to it on. On Amazon.

Leo Laporte (154:42)

I think I'm not supposed to do this, but Bonino just cut it out, okay? Just put a big boop over it or something like that.

Benito (154:52)

Well, then I'm gonna have to cut out all the context too, then.

Leo Laporte (154:54)

Ways people become emotional at the mere

Jeff Jarvis (154:57)

mention of her name.

Leo Laporte (154:58)

Well, that's not her.

Jeff Jarvis (154:59)

That's the intro. Probably hit the. Hit the end of the bar. Just. Just. Yeah, just. What do you call it? Scrub it.

Leo Laporte (155:06)

You've reached the end of this preview.

Jeff Jarvis (155:15)

You reached the end.

Leo Laporte (155:16)

Just like Steven Spielberg. You reached the end of this movie. To see more. All right.

Jeff Jarvis (155:25)

To buy it.

Leo Laporte (155:26)

Well, good. We just know this was just the Ungrim Seg segment. I just want to do one Ungrim sequence segment with funny voices and laughter.

Jeff Jarvis (155:34)

Oh, music. No, that's no good. You can't play. You shouldn't have played it because it has music in it.

Leo Laporte (155:37)

I don't think that was real music. I don't think we have to worry about. I don't know. I'm more worried about Audible getting mad at me, to be honest.

Jeff Jarvis (155:46)

Promoting their books. Audible. Come on.

Leo Laporte (155:48)

I keep saying that, but that doesn't stop them.

Benito (155:50)

That doesn't matter.

Leo Laporte (155:51)

Yeah, doesn't seem to matter. That's why I said, oh, it's a pretty good book that you'd probably want to listen to on Audible. Audible.

Jeff Jarvis (155:59)

Can I have one fit?

Leo Laporte (156:01)

Yes, I see this. And that's why I highlighted it in red.

Jeff Jarvis (156:05)

So you can avoid it. Or so you could come to it.

Leo Laporte (156:07)

So I could come to it. Because it says Jeff's effing had it.

Paris Martineau (156:11)

It's orange. Are you colorblind?

Leo Laporte (156:13)

No, I just missed.

Paris Martineau (156:15)

I pushed, you know, I just wanted to make sure.

Leo Laporte (156:17)

No, I know it's orange. I was painting a picture for the people at home who can't see it. What is the problem, Jeff? Google rolls Gemini and more apps, okay? Meanwhile, I still cannot get Google Gemini embedded in my Google Chrome. You didn't do a good read.

Jeff Jarvis (156:34)

No. You got Chrome. You capitalize Google.

Paris Martineau (156:36)

I still cannot get Google Gemini in more apps. Okay? Meanwhile, I still cannot get Google Gemini embedded in my Google Word Chrome on my Google Chromebook. You can do voices, basically account. Wtf?

Leo Laporte (156:52)

You didn't. You said you couldn't do voices. You can do voices.

Paris Martineau (156:57)

I can, but I don't know what that was.

Leo Laporte (157:00)

It doesn't. You don't have to know what it is. Definitely don't.

Jeff Jarvis (157:04)

I'm pissed.

Leo Laporte (157:05)

So what are you upset about?

Jeff Jarvis (157:07)

In my browser, in my Google Chrome, on my Google Chromebook, under my Google Workspace account, I can't get Google Gemini. All everybody else uses using Chrome has Gemini in there, so.

Leo Laporte (157:18)

And everybody else using Chrome doesn't want it.

Jeff Jarvis (157:20)

Well, but I'm the one person who does and I can't get it.

Leo Laporte (157:23)

So here's our Jeff.

Paris Martineau (157:24)

I'll let you know what you're missing. Every time I write an email now, somewhere between, oh, I got it 20 and 44 of the words have a little angry purple.

Jeff Jarvis (157:35)

I got a cheerio.

Paris Martineau (157:36)

That's just like, what if you change this to be something worse?

Leo Laporte (157:39)

That's not how Neela Patel would say it.

Guy Kawasaki (157:43)

All right?

Benito (157:43)

And the redon doesn't do auto. It's not like spelling anymore.

Leo Laporte (157:47)

Here's the rundown. Show my screen. This is the rundown, right? This is the thing. Do you not have this little guy here that you rotate? You click it. I don't ask Gemini.

Jeff Jarvis (157:58)

No, I do. I do. In sheets. I do. In Gmail. I don't In Chrome browser.

Leo Laporte (158:03)

Oh, in the.

Jeff Jarvis (158:04)

I want to use it to see what happens. Just, I want it to analyze, you know, tabs and things like that. Ah, that's what I want.

Leo Laporte (158:11)

Okay, I have it.

Jeff Jarvis (158:11)

I have it in Gmail, I have it in sheets, I have it in docs, but I don't have it for some reason in a Chrome browser on a Chromebook.

Leo Laporte (158:19)

That is odd.

Jeff Jarvis (158:20)

It's really odd. Google. Come on, man. Come on. I'm not joking now.

Paris Martineau (158:25)

You should get computer.

Leo Laporte (158:27)

Yeah, you should get. Really honestly, just get it. You gotta get rid of your computer.

Jeff Jarvis (158:32)

Get computer.

Paris Martineau (158:33)

Yeah, toss that computer. Computer and get computers.

Jeff Jarvis (158:36)

The truth is, on my computer, I can't have computer.

Leo Laporte (158:40)

Oh, no, Perplexity computer on your computer.

Jeff Jarvis (158:42)

Because my computer is a Chrome computer.

Leo Laporte (158:45)

Do we have any insight on my computer?

Paris Martineau (158:48)

I can't have computer as a perfect. I just like, we can't have another computer based title after last week.

Leo Laporte (158:56)

But what was last week's title? I didn't. I didn't hear it.

Paris Martineau (159:00)

You got to download computer, I think.

Leo Laporte (159:03)

Really?

Paris Martineau (159:03)

Yeah.

Leo Laporte (159:04)

Second week.

Benito (159:06)

No, it was you got to get. You got to get computer. It was last week.

Paris Martineau (159:10)

You got to get computer.

Leo Laporte (159:11)

All right, this week I'm going to title it. We have computer at home.

Paris Martineau (159:18)

We can't get computer at home though. Jeff can't have computer at home.

Leo Laporte (159:22)

No, come on.

Jeff Jarvis (159:23)

You can't have computer on my Computer,

Leo Laporte (159:26)

do we have anything to say about Jay Graber stepping down at blue sky? CEO? Blue sky stepping up.

Paris Martineau (159:31)

I frankly think more CEOs should step down when they're like, I shouldn't be the CEO of a company of this size. I think it's great.

Leo Laporte (159:39)

She was there from the beginning when it was just a research project work

Jeff Jarvis (159:43)

and made it into something, especially as X disintegrated. She did wonderful work. And Tony Schneider, who's stepping in as the interim CEO, is great. He, he's a. A nice investor and has worked very closely with Matt Mellon and he's just temporary.

Leo Laporte (159:59)

They're going to bring in a real CEO.

Jeff Jarvis (160:01)

He's act. He acted as CEO of WordPress, of automatic for a while. Tony's great. So we could. We should find out what Mazdick has to say because he's on the board.

Leo Laporte (160:13)

Oh, yes, we should. Are we trying to get Mike on? I would like to.

Paris Martineau (160:17)

You think? You have to find a four and a half minute window that works for his schedule.

Leo Laporte (160:21)

I could tell you one thing. William Shatner did not get $42 million. $42. Sorry, $42 from Jay Graber. He did get $42 from Elon Musk. So there. This is the new so X is now doing X money. You, I guess you have to be a, you know, fading movie star to get the access. But last, last month, Shatner said Musk sent him $42 through X money, which helped Shatner raise $200,000 for charity. Good for him.

Jeff Jarvis (160:57)

Have you seen the Will Shatner Raisin Bran commercial?

Leo Laporte (161:00)

Yes, it's quite funny. It was on the super bowl

Jeff Jarvis (161:05)

here, bringing you fiber to the masses with Kellogg's Raisin Prep.

Leo Laporte (161:09)

It was a little. It was a little.

Jeff Jarvis (161:11)

It was a little edgy.

Leo Laporte (161:12)

Edgy, yeah. I love Shatner and I wish him the best. He says, one day, Musk writes to me, well, I don't have my fortune tied up in cash. It's all in ownership of the varieties of companies, Shatner said. So I jokingly say, well, if you need a few dollars, I can lend you a few dollars. Instead, Musk sent him $42, a nod to the Hitchhiker's Guide of the Galaxy. And now I'm not sure exactly how he turned this into a big charitable contribution, I guess. Oh, I get it. He says, I got $42 from Elon. I'll sell each of these dollars. He offered donors access to the beta in exchange for contributions. For $1,000, donors will receive $1 from the X Money account. He said he thought he would get $42,000 or $1,000 for Everybody Buck. But the first round sold out so fast, Musk sent another $42. And so basically they're in this cooperative to promote X Money, which if you ask me, I wouldn't go near, but okay, fine.

Benito (162:24)

Do they believe that money is not fungible? Like, I don't understand this at all.

Leo Laporte (162:29)

Yeah, because you're not getting a physical dollar bill signed by Elon Musk. You're just getting a dollar in your X Money account. Account. But that. Oh, that did give you access to it. Because if somebody who has an X Money account sends you money, now you have an X Money account.

Jeff Jarvis (162:44)

Oh, okay.

Leo Laporte (162:44)

So that's whatever you do, folks. Do not send me any X money. I do not want it. Don't send me last ad and then we're going to do our picks of the week. If you don't mind, boys and girls, you're watching Intelligent Machines with a groggy Paris Martineau and a fading Jeff Jarvis. I have worn them down. Down. I have worn them down. I did. Did you miss me from last week? I bet Jason Hiner didn't wear you down, did he? No, he kept the show moving.

Jeff Jarvis (163:18)

Oh, Paris, you're muted.

Paris Martineau (163:20)

He actually had a very hard out because the podcast studio he was in was going to kick him out if we did a show this long.

Leo Laporte (163:28)

So he would have been gone eight minutes ago.

Paris Martineau (163:30)

A large cane would have come from stage right and pulled him off.

Leo Laporte (163:35)

This episode of Intelligent Machines brought to you by Stash. Have you dabbled in investing here and there, but haven't been happy with how things are going? Stash helps turn good intentions into consistent progress. Stash isn't just another investing app. It's a registered investment advisor that combines automated investing with expert personalized guidance so you don't have to worry about gambling or figuring it out on your own. Stash is simple, smart and stress free. Choose from personalized investments. Let Stash's award winning smart portfolio do the work for you. Or pick a combo of both. Stash is there to help guide you every step of the way. Join over 1,000,000 Active Stash subscribers and finally, let your money work as hard as you do. Don't let your money sit around. Put it to work with stash. Go to get.stash.com im to see how you can receive $25 towards your stock purchase and to view important disclosures. That's get.stash.com IM get.stash.com IM paid non client endorsement. Not a guarantee nor representative of all clients. Smart portfolios are discretionary managed accounts and subject to additional fees. See the advisory agreement and deposit account agreement for details. Investment advisory services offered by Stash Investments LLC and SEC Registered Investment Advisory Advisor. Investing involves risk. Now it's time for our picks of the week. That should liven things up. Who wants to start?

Jeff Jarvis (165:11)

Let's.

Leo Laporte (165:11)

Let's let Paris start.

Paris Martineau (165:14)

I'm trying to decide whether I want to invite anybody. Okay, I won't.

Leo Laporte (165:19)

You're playing too many games already. They. I know. They want to know.

Jeff Jarvis (165:22)

No, this is becoming a.

Leo Laporte (165:23)

Did you pick from handle by the way, or the that?

Paris Martineau (165:26)

Yeah, I did pick my handle. Apparently you can just set your handle to anything. So I could just change it to the letter A. It's not unique, but I just kill. I thought it was kind of funny.

Leo Laporte (165:35)

Stone cold killer.

Paris Martineau (165:36)

Stone cold killer. So I guess my pick will be. It's labeled. I'm going crazy under my picks because I've descended into coffee madness. Oh, I have recently been optimizing my pour over setup as I've talked about in the show. I've got a brew logger now going on manual. I vibe coded. This photo that I'm showing is. I realized that I think my issues with coffee consistency might be because my manual grinder isn't producing an even distribution of particles. So I found a website that I guess could actually be my pick of the week. I'll find it once.

Leo Laporte (166:17)

Oh, you have a 292.27 micrometer standard.

Paris Martineau (166:20)

I know. And my particle size was way too high to begin with. And see if you look, you have one that's up there that's 1,226 microns and another from the exact same batch that's 950 microns.

Jeff Jarvis (166:36)

But how does it taste, Paris?

Paris Martineau (166:39)

It tastes good. But sometimes. The thing is I. I started doing this because I had dialed in a great coffee recipe. I had like 92 clicks on my onesie Presso J manual grinder and this wonderful coffee from perk that I'd really recommend called the Ethiopia chelcha which is an anaerobic natural process that has phenomenal blueberry strawberry.

Leo Laporte (167:06)

This is deeply afraid of.

Paris Martineau (167:09)

But I was getting some of my best cups.

Jeff Jarvis (167:12)

Why? She's a domestic reporter. She's red string up.

Paris Martineau (167:16)

But then I. I changed nothing about my setup. I've got my hario switch. I'm doing a or six pore method. I've got ranging from 205 degrees to 210. But sometimes I would get a 9 out of 10 cup and the next day I'd get a 5 out of 10 cup. And I was like, what is the issue? And the issue is that my particle size is trash. And so now, now what I've realized I'm going to do is go to. I'm going to put this also in the rundown so you can click on it. I found a place in Greenpoint called Hyonia Club, which is a coffee club for freaks where you can pay $15 to get a day pass and try out. They have every fancy coffee equipment you could imagine and you can try them all out and determine whether or not it's actually worth spending 200 plus dollars on a grinder.

Leo Laporte (168:11)

200? Oh, that's not a good enough grinder. You need to spend $2,500 on this espresso.

Paris Martineau (168:18)

Dude, I'm gonna try, I'm gonna try the maser fillers.

Leo Laporte (168:22)

Well, this is what you need.

Paris Martineau (168:23)

That's an espresso grinder. I don't make espresso.

Leo Laporte (168:25)

Oh, that's right. You need a regular grinder.

Paris Martineau (168:28)

I could, I don't know necessarily that I want to even pay for an electric grinder because we have a barrage more cost to it and I'm only making 20 grams of coffee at a time. However, I might pay like $400 for a manual grinder, which is.

Benito (168:43)

Doesn't the CR lab have like every grinder available?

Paris Martineau (168:47)

Yeah, well, that's a great point. We don't have. We haven't tested manual and specialty coffee grinders and I think that could be a great avenue for us to explore.

Leo Laporte (169:00)

Oh, maybe this is what she's working on, huh?

Benito (169:02)

You gotta pitch that. You gotta pitch that's.

Paris Martineau (169:03)

Listen it. I, I'm going to. As I've gotten to this, I downloaded this great book called the Physics of filter Coffee. I, I'm.

Leo Laporte (169:12)

I am like you watch James Hoffman's reviews.

Paris Martineau (169:15)

Although I've. Because I don't like YouTube videos, but I have read a lot of. Yes, I know, but I'm like trying to get more into the physics of it rather than specifically coffee influencers because their vibes change.

Leo Laporte (169:25)

He's not an influence. He's really a coffee expert. I would.

Paris Martineau (169:28)

Yes, he's an expert. I follow the recipe I'm using is by another expert like this. And I was using a Hoffman recipe for a while.

Leo Laporte (169:36)

But we gotta. I gotta get you on with a coffee geek. We've done coffee shows on the club and we'll do a special coffee show with you. And we should the coffee geek because I. I think Mark Prince is fantastic and he is an expert on all this stuff. He does like his espresso. The other thing I would recommend. I wanted to buy this. I almost did. This is a brewer that does pour over automatically for a more consistent.

Paris Martineau (170:00)

I mean I don't know.

Leo Laporte (170:03)

And it has a pit and everything. So you can.

Paris Martineau (170:06)

Based on the thing that says ratio four makes me suggest. Suggest to me that it might be a 4, 6 brewing method. But I think part of it is I do. I'm making coffee for just me and it really.

Leo Laporte (170:16)

This makes one cup.

Paris Martineau (170:18)

It takes five minutes literally in the morning, morning from grinding.

Leo Laporte (170:22)

I think the handwriting's gotta stop.

Paris Martineau (170:24)

I don't think the hand grinding. So Leo, I was right there with you two months ago. I was over at a friend's house.

Leo Laporte (170:30)

Don't use the Ozo though. That's terrible. That's a horrible.

Paris Martineau (170:33)

I. I was at. I mean, listen, I do. Oxo was a problem, but I was at a friend's house and he was manually coffee grinding in the morning and I was like, what is this, the stone Age? Are you an idiot? And then I did more research and I was like, no, honestly, you can. Can get. You can get some good stuff. But I think what I've realized is I need to. It all depends on your particle size distribution.

Leo Laporte (170:59)

And this is who has size analyzer. Is this an app?

Paris Martineau (171:03)

Yes, I need to actually find where it is, hon.

Leo Laporte (171:05)

I got it right here, the grind size analyzer. So you use your phone, you take a picture, you upload it and it will take. Tell you, all right, I'm going to do this.

Paris Martineau (171:13)

It's just a web app. It's a website. Right.

Leo Laporte (171:17)

I want to do this. Yeah, it's not a. You're right. Unspecialty Edu. So I will do this with our Barat grinder and I'll let you know.

Paris Martineau (171:26)

You know, there's certainly some problems with it. Like obviously you're just taking a photo. You can't, despite what the instructions say, have the grinds go over little boxes. But I would suggest doing it a couple times with samples from all from the top and bottom of your grind cup because it'll. It'll differentiate because you got boulders up top and fines below, as I well know now.

Leo Laporte (171:53)

But you are hysterical. So actually the whiskey is unspecified low

Paris Martineau (171:58)

key, but it's also resulting in delightful coffee.

Leo Laporte (172:01)

And so how many cups of day do you drink?

Paris Martineau (172:05)

One in the morning.

Leo Laporte (172:06)

Yeah, see that's my problem is I Can only drink one in the morning,

Paris Martineau (172:09)

but it's when it's.

Leo Laporte (172:11)

But if it's a really good Ethiopia

Paris Martineau (172:13)

chelcha, when I'm making it right, it tastes like. Like strong blueberry and strawberry notes in the city. It's.

Leo Laporte (172:21)

So I would also recommend medium, a light.

Paris Martineau (172:24)

It's like light medium.

Leo Laporte (172:25)

Yeah.

Paris Martineau (172:25)

Over there. I also recently just got a bag in today from Hydrangea Roasters. It's going to bit be little a lighter one. I'm trying but I would really recommend people check out this Ethiopia Chel chip from Perk Coffee. And if you're listening to this later this week on the 13th, perk, this is not an ad. I just really like this one thing of Perk is kind of obsessed with the number 13. So on the 13th of every month, this roaster gives you 31 off any order of their site. So I don't know if you want to try a fancy coffee, you can get it for a third up stuff. It's kind of fun.

Leo Laporte (172:56)

I might, I might try this. I usually just make espresso, but this

Paris Martineau (173:01)

chelchi, I don't know how this works.

Leo Laporte (173:02)

Is it like. How is it like? Is it compared to yerkachefe, which is what we've been.

Paris Martineau (173:06)

It's an Ethiopian yerikacheffe.

Leo Laporte (173:08)

Oh, it is. It's just.

Paris Martineau (173:09)

They call it from the ghetto zone. I believe Chelche might be the producer.

Leo Laporte (173:14)

Oh, okay. I really, really like it. Yeah.

Paris Martineau (173:18)

I'm a big.

Jeff Jarvis (173:19)

They're doing this just to the asked for this.

Paris Martineau (173:21)

And so on their website they have Revel espresso brewing instructions as well.

Leo Laporte (173:28)

And this is an anaerobic natural process. Jeff.

Paris Martineau (173:31)

This is, you know, some of.

Leo Laporte (173:33)

Is it washed? It's not washed. Is it?

Paris Martineau (173:35)

It's not washed. And. But you know, one of the things I've been getting really into is co ferments. Have you heard about this? They're a bit controversial in the coffee world, but I think that's dumb. It's where the fermentation process, they introduce additional fruits in there and see what's going.

Leo Laporte (173:51)

You're doing this just to annoy Jeff, aren't you?

Paris Martineau (173:53)

I am. He looks like he's being tortured.

Leo Laporte (173:56)

Oh, we're torturing him. We're just torturing him. We do aeropress around here with a baratza grinder and it produces a pretty good cup of coffee. But I, you know, I, I heard about.

Paris Martineau (174:09)

All the Aeropress boys are doing soup now. Look it up.

Leo Laporte (174:11)

Soup in an aeropr press.

Paris Martineau (174:13)

That's. It's not. It's not actual soup. It's a method of brewing coffee that's called soup.

Leo Laporte (174:19)

Soup. Okay.

Paris Martineau (174:20)

And apparently it's the hot thing in more ways than one.

Leo Laporte (174:24)

So your recommendation also is the. Which subreddit? The pour over subreddit.

Paris Martineau (174:29)

I reckon it's a blessing and a curse. The. Our slash, pour over has caused all these things that just. That word vom.

Leo Laporte (174:36)

I've never done pour over. I never got pour over because when I was a kid, we had a cat Chemex, and we would pour the coffee. Well, Chemex, but this is a little different.

Paris Martineau (174:44)

No, no, no. That's what I've realized. So before all of this, me a fool. Three months ago, I was doing the Chemex. Little did I realize the Chemex filters, they're stripping all the flavors I want out of that. Plus the crappy grinder I was using, I was getting. I was paying for all this fancy coffee. I was getting trash. It's like a whole. I've got a Hario switch now, which allows me to do both immersion and percolation. But it's glass because I got the glass one because I'm worried about microplastics. So I've got to preheat it. It's a. It's a diet.

Jeff Jarvis (175:16)

What happens when you add caffeine to this mania? Room displaying right now.

Leo Laporte (175:20)

Here's my question. How long does it take you to make your coffee in the morning?

Paris Martineau (175:26)

From stepping over to my kitchen to drinking coffee, less than five minutes.

Leo Laporte (175:31)

Oh, that's all right.

Paris Martineau (175:33)

The actual brew time is. Is 250 to 350.

Leo Laporte (175:38)

It feels like you're doing a lot of stuff.

Paris Martineau (175:40)

No, this. This is all the thinking that I do before, after, around the brewing.

Leo Laporte (175:49)

I have just joined R. Pour Over. So, uh. Oh, look at this.

Paris Martineau (175:55)

Yeah, everybody's making fun of that guy.

Leo Laporte (175:57)

That's a portable kit ready to go. Got everything you need for your pour over. Is that the kind of manual grinder you have there? Is that?

Paris Martineau (176:05)

Yes, yes, that is the kind of manual grinder.

Leo Laporte (176:07)

Yeah. I can't believe that.

Paris Martineau (176:09)

Seems there was a video someone posted of someone on an airplane brewing coffee because they had a horror stag, fellow stag, ekg, kettle, all the stuff.

Leo Laporte (176:22)

This is a great group because it's people really who are not embarrassed. They're among friends and they can share their obsessions.

Paris Martineau (176:29)

All right, so if people out there who've listened to all this and understood have any thoughts as to whether I should get a Z P6 or a K Ultra grind, please Let me know. Or if I should spring for the Pietro. But that might be crazy.

Leo Laporte (176:45)

You're crazy.

Paris Martineau (176:46)

What I'm. What I value is fruitiness, and that's covered by.

Leo Laporte (176:51)

And you don't get that in an

Paris Martineau (176:52)

espresso because the ZP6 is clarity focused but at the sacrifice of body. And the K Ultra might be too body focused that I wouldn't get my flavor separation. It's maddening, I tell you. I'm so sorry.

Leo Laporte (177:08)

Poco. I have the Ninja Lux, actually. It's quite a good coffee maker, but probably not in the category that Paris thinks. But I actually love for $500, it's a really good coffee maker from Shark. All right. Right now you make me want a cup of coffee. The problem is the sun's gone over the yard arm. It's not time for coffee. I have to drink it in the morning or I won't sleep well.

Paris Martineau (177:33)

Yeah, I was just saying. I don't even think I could do a decaf or something about the idea of just drinking coffee. Like beverage that would make me feel awake too.

Jeff Jarvis (177:40)

So that's all I drank, is decaf.

Leo Laporte (177:42)

What you really need is a percolator. Get one of those Hamilton beach percolators to have.

Jeff Jarvis (177:47)

My house was on all day long.

Leo Laporte (177:49)

And you leave it running. Yep.

Jeff Jarvis (177:51)

And you have.

Leo Laporte (177:51)

You could drink all day, but you

Jeff Jarvis (177:53)

make would come over to visit my mother and sit down with him.

Leo Laporte (177:56)

Coffee's on. Come on over. Have a coffee. Clatch. Jeff Jarvis, your pick of the week.

Jeff Jarvis (178:04)

Well, so we can talk about how much culture has changed. Two things. One, Hollywood Reporter reports that YouTube is now the world's largest media company.

Leo Laporte (178:15)

Company. Yeah. Isn't that amazing? I actually bookmarked that as a story. That's unbelievable.

Jeff Jarvis (178:21)

YouTube had more than $62 billion revenue, which beats Disney.

Leo Laporte (178:26)

$62 billion bigger than Disney.

Jeff Jarvis (178:29)

So who was it here who did this? Moffett Nathanson. Analysts estimate its value on its own at 500 to 560 billion, versus Netflix, which has a market cap of 409 billion.

Leo Laporte (178:43)

Wow.

Jeff Jarvis (178:45)

Total revenue of 40 billion. But that's pl. That's ad revenue plus subscription business.

Leo Laporte (178:50)

Yeah, because I, I. I subscribe to YouTube TV. I give them my 80 bucks a month.

Jeff Jarvis (178:54)

YouTube has around 10 million subscribers now, which will soon overtake eventually Charter and Comcast.

Leo Laporte (179:02)

Yeah.

Jeff Jarvis (179:02)

So that's our present world. But I want to take you children to the past.

Paris Martineau (179:07)

What's that?

Jeff Jarvis (179:08)

Is 1994, Grandpa. Yeah, the year is 1994. And in that time, it was so exciting. Every fall that we would have an entire one hour special. In addition to having the guide special, we had a one hour special of.

Leo Laporte (179:26)

Now you can see why he was San Francisco's most eligible bachelor, can't you?

Jeff Jarvis (179:30)

Well, one of us ET Tease Entertainment tonight's t fall TV special. So I, I, I, I swept in the link below. You can keep the sound off you

Leo Laporte (179:42)

can we watch it?

Jeff Jarvis (179:44)

Yes, you can. There it is. There it is. You want you for Mary Hart's hair.

Leo Laporte (179:52)

Pretend you can press any button you want. Television season is upon us and it is truly out of this world.

Jeff Jarvis (179:58)

I can do wonders.

Leo Laporte (180:02)

Such big ties. Where do you show up?

Jeff Jarvis (180:05)

So if you go to the link,

Paris Martineau (180:07)

that's yeah, we're in the 45 minutes.

Jeff Jarvis (180:09)

Oh, yeah. In 207. I'm there if you go to the link.

Leo Laporte (180:12)

When I was in my early radio days, right around this time maybe a few years earlier, I would record and watch e T every night looking for gossip I could use on my radio show the next day. Yeah.

Jeff Jarvis (180:26)

So 207 I, I, I, I swept

Benito (180:29)

across to just find out whose birthday it is. Is watching ET that used to be

Leo Laporte (180:33)

the only we had to do this. Yes, we had to do that. All right, let's jump to Jeff's appearance on Entertainment tonight with. Was it Mary Hart?

Jeff Jarvis (180:43)

Yeah, and John Tesh.

Leo Laporte (180:45)

And John Tesh. Wait a minute. No, it's the beginning. I don't, no, it should, it should

Jeff Jarvis (180:49)

be at the, at the time code.

Leo Laporte (180:50)

Oh, I got to look at the link again.

Jeff Jarvis (180:53)

Start at 42 minutes.

Leo Laporte (180:54)

All right. I can't zip my pants up.

Paris Martineau (180:57)

I'm Mary Hart, the OG Mary har comes to the small screen.

Leo Laporte (181:01)

Does that mean I'm under arrest?

Paris Martineau (181:03)

And the good guys get the bad guys.

Leo Laporte (181:05)

This is chasing crooks.

Jeff Jarvis (181:07)

It's so et up and a beauty earns her. It's wonderfully written.

Paris Martineau (181:12)

That's a crazy. Oh, he's in a little floating box. You missed it. Judges Tom Shales of the Washington Post, TV guides. Jeff Jarvis and Jonathan Storm.

Leo Laporte (181:25)

B roll,

Jeff Jarvis (181:28)

I think is my so called life. It's wonderfully written.

Leo Laporte (181:30)

That was a good show.

Benito (181:32)

One season.

Jeff Jarvis (181:32)

By far my pick for the best show. Yeah, well, there's that. I also panned friends and I'm sure

Leo Laporte (181:38)

some critic will say there's no business like I like Tom Shields. Is he still around?

Jeff Jarvis (181:42)

I don't know. He's dying.

Leo Laporte (181:43)

Of course it's I don't know this guy.

Paris Martineau (181:46)

I watched that recently.

Jeff Jarvis (181:49)

Three white guys tell you about tv.

Leo Laporte (181:51)

So these were your picks, huh? Of, of all three of them. You were the one that picked the show that. Oh, what did you hate?

Guy Kawasaki (181:57)

One of the shows to be first canceled with McKenna. Chad Everett comes back and boy, is he old Short.

Leo Laporte (182:03)

It's in a little bit of trouble. I think this year with his show, it's possible that Martin Short will not get to air.

Jeff Jarvis (182:10)

Why is Daddy's Girls the absolute worst this season?

Leo Laporte (182:12)

Daddy's Girl there really ought to be. Oh, Dudley Moore. No, no, no. Yeah, I was looking at all these shows. All right, ladies and gentlemen, that's a little glimpse of 1994 seen through the lens of one Jeffrey T. Jarvis. Thank you very much. I have so many things I could show you. This one I think is good for Paris. It's called Payphone Go. It's like Pokemon Go.

Paris Martineau (182:34)

Oh, I've seen this and I want to play.

Leo Laporte (182:36)

It's only with payphones.

Jeff Jarvis (182:38)

Oh, no, another game.

Paris Martineau (182:39)

Where's the nearest payphone go to? You don't.

Leo Laporte (182:41)

Oh, that's a good question. I have to get an ID and all that, but you find a payphone and you and you and you go to it and you get points for going to it. Wow. There don't look like there are so

Paris Martineau (182:53)

many pay phones in San Francisco.

Leo Laporte (182:55)

Yeah, but out here in the boonies, I don't see.

Benito (183:00)

What do the different colors mean?

Leo Laporte (183:03)

I don't know. So it says, use the map to locate a pay for phone, pick up the receiver, dial 888-683-6697 and then enter your player ID. The first person to call from a play Payphone payphone gets 20 points. Second person gets 10, third gets five. Oh, look, there's a payphone. There are two in Petaluma.

Jeff Jarvis (183:25)

So I wonder what.

Leo Laporte (183:26)

There's one, there's two at the fairgrounds and there's one here. And each has one claim just around the corner from our old studio. Studio.

Benito (183:34)

I love this game. This is awesome.

Leo Laporte (183:36)

Yeah, so you go, you make a call and you get points.

Jeff Jarvis (183:39)

Brooklyn. Go to Brooklyn.

Leo Laporte (183:40)

Should we go to Brooklyn? All right, let's zoom out.

Benito (183:43)

Yeah.

Paris Martineau (183:43)

Wait, no, it's only in California.

Leo Laporte (183:45)

No, really.

Paris Martineau (183:46)

It says California still has only 2002.

Leo Laporte (183:49)

That's no fun. Why? Gotta expand it. Oh, man. Well, I guess you can't play Pay Fun Go.

Benito (183:58)

I mean, it's probably just one dude sitting at homemade this. So you know.

Leo Laporte (184:02)

Oh, of course it is.

Paris Martineau (184:03)

How did you find out the pay phones?

Leo Laporte (184:05)

You look. Oh, there must be some.

Benito (184:07)

There must be some public record, right?

Leo Laporte (184:09)

Public records.

Jeff Jarvis (184:10)

Well, these aren't aren't phone company phones anymore. They're they're no.

Leo Laporte (184:14)

In fact, Henry's best friend's father made a killing because he bought all of AT&T's pay phones. They wanted to get out of the business. It was a real pain to go around each phone, collect all the dots times. So he put cell phones in the pay phones and he already had a business selling restock. He would sell ice cream in freezers. You know how you go to a bodega and there's a freezer and there's ice cream in it? He had a really good business. It turns out it's not an ice cream business, it's a logistic business. He made sure that the freezer cases had scales in the feet so they wouldn't bother restocking them until the weight of the freezer was low enough to justify it. And he realized I could do the same thing with a payphone coin case. So I will save time and money only collecting when the payphone is full.

Benito (185:07)

Brilliant.

Leo Laporte (185:08)

He's very wealthy. Very wealthy. Got all those payphones for a song. Here's a company that wants to pay you $800 to bully AI. This might be a good job for you, Paris. The company wants candidates with extensive personal history of being let down by technology. The eight hour session is intended to promote the startup's AI memory tool. If that's your job, go to professional AI bully and get paid to bully AI. Oh, only one person gets the job. Memvid.com. this sounds a little bit like a promotional view vehicle. One person.

Benito (185:52)

Oh, sounds like we're making a pariah for all the AI to kill at the end when it all.

Leo Laporte (185:56)

And then I did get you something. This is for you, Benito. It's called Tweak Bench. Your favorite producer's favorite plugins. These are all the VST plugins used by FA by famous producers. I know you want to write your own with Claude code.

Benito (186:13)

This is awesome though.

Paris Martineau (186:14)

That.

Benito (186:14)

That is great.

Leo Laporte (186:15)

I thought you might like this. Your favorite producer's favorite plugins. Lol. I'm not sure what the LOL is all about, but. And they're, you know, five bucks. They're not expensive.

Benito (186:29)

Yeah, I mean there's a whole. There's a whole range of free ones too that are like.

Leo Laporte (186:32)

That's true. Yeah, No, I know. Yeah. I don't know what the deal is with this. These are Tweak Bench plugins. So I guess this is a company that does plugins, but five bucks, nice price. Get the whole Tweak Bench bumble bundle for 60 bucks. Ladies and gentlemen, we have now filled our

Jeff Jarvis (186:53)

order.

Leo Laporte (186:54)

Cornucopia to the brim with good.

Jeff Jarvis (186:58)

Don't make it.

Paris Martineau (186:58)

When's the last time we did a three and a half hour pod? It's been a minute.

Leo Laporte (187:03)

Three hours and 12 minutes. You do you think we started?

Paris Martineau (187:06)

I logged in in to the Zoom call. 3 hours and 33 minutes.

Jeff Jarvis (187:11)

It's like flight attendants don't get paid until the door is closed.

Leo Laporte (187:13)

The door doesn't start counting. The doors don't close until I push the button.

Paris Martineau (187:17)

And Leo spent those first 30 minutes trying to figure out how Zoom browser works.

Leo Laporte (187:22)

I'm gonna figure that out before the next episode.

Paris Martineau (187:25)

You've said that so many times.

Leo Laporte (187:27)

One of these days. Thank you so much to Guy Kawasaki, a wonderful guest. A wonderful, wonderful guy. Really enjoyed talking to him. It's probably not too late to email him and get his book for free on in a kindle form. Just email. Everybody has something to hidemail.com and Guy will send you a copy. Thank you. Paris Martineau. She is doing her coffee research for Consumer Reports where she's an investigative journalist. So nice to see you. You got the Hawaiian shirt on. When'd you do that? I, I, I missed that at the

Paris Martineau (188:02)

beginning when we all changed into match.

Leo Laporte (188:06)

You trying to get to work? Thanks to Jeff Jarvis. Put on his Hawaiian black hoodie.

Jeff Jarvis (188:11)

Hawaiian hoodie.

Leo Laporte (188:12)

Yeah, yeah. Jeff, of course, the author of the Gutenberg parenthesis magazine. And soon hot type.

Jeff Jarvis (188:18)

Not so soon now. It was going to come out June 11, hot type. But with a production delay, they were going to push it into July. And I kind of had a little fit and I said, that's death time. So now it's coming out at the end of August, but you can pre order it now still.

Leo Laporte (188:31)

So July is not a good time for a book to come out.

Jeff Jarvis (188:34)

No, because everyone's gone and they're in the hot type.

Leo Laporte (188:37)

The magnificent machine that gave birth to mass media and drove Mark twain mad. Available July 23rd. Amazon has.

Jeff Jarvis (188:46)

That's August now.

Leo Laporte (188:47)

Have you read the, did you do the Audible?

Jeff Jarvis (188:50)

I haven't done it yet.

Leo Laporte (188:51)

No.

Jeff Jarvis (188:51)

No.

Leo Laporte (188:51)

But it is going to be available.

Jeff Jarvis (188:53)

Yes.

Leo Laporte (188:54)

Okay, good.

Jeff Jarvis (188:55)

Yep.

Leo Laporte (188:56)

So do me a favor. Whenever you do Mark Twain's voice, can you do it like Mark Twain here?

Jeff Jarvis (189:02)

Well, who, who was it who played Mark Twain?

Leo Laporte (189:05)

Hal Holbrooke.

Jeff Jarvis (189:06)

Hal Holbrook. I think I now have to do my Hal Hobr.

Leo Laporte (189:09)

Mark Twain. Good night. Do you do tonight, do you do a Hal Holbrook?

Jeff Jarvis (189:12)

Well, should we play a little bit for Paris on the way out?

Leo Laporte (189:15)

Here of Mark Twain tonight?

Jeff Jarvis (189:18)

Yeah, sure.

Leo Laporte (189:18)

Why not? If we haven't been taken down yet. Yet. I wish I'd seen it. I think there were broadcast versions of it. Wait, if you're going to go around

Benito (189:27)

the country, they're not going to let us out on. Even post this on YouTube.

Leo Laporte (189:31)

It's not music. It's not music. It's just Hal Holbrook on stage talking like Mark Twain from 1967. So it's probably not even in copyright anymore. He did a pretty good, credible job with this. The most interesting jackass there is.

Paris Martineau (189:50)

Why is he dressed like Colonel sand

Jeff Jarvis (189:54)

at the end of his life? That's what Twain wore.

Guy Kawasaki (189:55)

You admire him?

Leo Laporte (189:56)

He used to. Twain himself used to go around and do lectures. This is all, in fact, he had

Jeff Jarvis (190:01)

to because the machine that he backed to compete with the Linotype bankrupted him. And late in life he had to go around the world giving talks to make money, to make his money back.

Leo Laporte (190:10)

Would have been fun to see him, though. I bet he was. Oh, God, yeah.

Jeff Jarvis (190:15)

I should do that voice though, don't you think?

Paris Martineau (190:17)

In the.

Leo Laporte (190:17)

Yeah, you talk like there's a lot of.

Jeff Jarvis (190:19)

There's a lot of Twain quotes. I think I could do that.

Paris Martineau (190:23)

Yeah.

Guy Kawasaki (190:24)

See,

Leo Laporte (190:28)

I'm glad we got some voices in so this wasn't so grim. Thank you for joining us. We do Intelligent Machines every Wednesday, right?

Jeff Jarvis (190:36)

We finish by Thursday.

Leo Laporte (190:37)

It's usually done by Thursday. We start at 2pm Pacific, 5pm Eastern, 2100 UTC. Twitch, YouTube, TikTok. No, not TikTok. X, Facebook, LinkedIn and Kik, of course, in the Club Twit Discord. And thank you, Club Twit members for making this show possible. We really appreciate your support. Without that, I don't know if we'd be here. I really don't. So we thank you for keeping us on. If you're not a member yet, quick, go to Twit TV club, Twitter, Twit and join after the fact on demand versions of the show available at TWiT TV IM. There's a YouTube video channel dedicated to the show. Great for sharing clips.

Paris Martineau (191:18)

Can you use the rest to do a little voice? Can you do a little voice for the rest of the thing?

Leo Laporte (191:22)

Yeah. What voice should we do?

Jeff Jarvis (191:23)

Dr.

Paris Martineau (191:24)

Twain or a Transatlantic.

Leo Laporte (191:28)

A Mid Atlantic.

Paris Martineau (191:29)

Mid Atlantic, yeah.

Leo Laporte (191:32)

Well, thank you for joining us on this fine show and we'll be back again next week for another thrilling, gripping edition of Intelligent Machines. Good night, Martha, and save a beer

Paris Martineau (191:45)

for me on the radio.

Leo Laporte (191:48)

See you on the radio.

Jeff Jarvis (191:49)

Download computer.

Paris Martineau (191:53)

You can't get computer on his computer.

Guy Kawasaki (191:56)

See?

Leo Laporte (191:58)

Thank you, everybody. Bye bye. Hi there. Leo laporte here. I just wanted to let you know about some of the other shows we do on this network you probably already know about. This Week in Tech. Every Sunday, I bring together some of the top journalists in the tech field to talk about the tech stories. It's a wonderful chance for you to keep up on what's going on with tech, plus be entertaining by some very bright and fun minds. I hope you'll tune in every Sunday for this Week in Tech. Just go to your favorite podcast client and subscribe. This Week in Tech from the Twit Network. Thank you.

Paris Martineau (192:34)

I'm not a human being, not into this animal scene. I'm an intelligent machine.