Anthony Vinci (2:05)

Vincy?

Leo Laporte (2:05)

Vincy. You're not duh, Vinci, you're just Vincy.

Anthony Vinci (2:09)

And I'm only doctor to my wife.

Leo Laporte (2:11)

All right, all right. We put it in there. Anyway, you have a PhD in international relations, so I think that counts. But more interesting to me, he's a spy, or was a spy for a long time in the intelligence community in the field in Iraq and other dangerous places. Then actually, you've got a really great cv. But I'll start with the latest thing, which is a brand new book called the Fourth Intelligence Revolution. I have the audiobook up because I'm just about to buy it. The Future of Espionage and the Battle to Save America. I'm just about to buy it Because I have the PDF, but I want to listen to. This is such a. This is like a spy novel, except it's nonfiction.

Leo Laporte (2:56)

It's about your career and about what's about to go on and what's about. About to happen in the world around us. Anthony, thank you so much for joining us. I really appreciate it.

Anthony Vinci (3:08)

Thank you so much for having me on.

Leo Laporte (3:09)

Yeah.

Leo Laporte (3:11)

People are going to say, well, what does this have to do with AI? It has quite a bit to do with AI because we are about to enter the AI era of intelligence.

Anthony Vinci (3:23)

That's absolutely right. And that's really what drove me to write the book in the first place, was trying to bring AI into these agencies.

Leo Laporte (3:31)

Yeah. So I will. Let me. Let me just give you a little bit of Anthony's history because it's actually. It's fascinating. You were a clandestine intelligence officer with the Department of Defense.

Leo Laporte (3:47)

I hate it when interviewers do this. Like I'm telling you. You know, you were there. Anthony was a clandestine intelligence officer, a spy in Iraq, Africa and Asia and throughout the Middle East.

Leo Laporte (4:00)

He then became chief technology officer at the National Geospatial Intelligence Agency, the nga. That's one of the big five US intelligence agencies responsible for analyzing satellite imagery, which is a geospatial analysis, which has become increasingly important in this era. He then became a hedge fund member and managing director at a capital firm and also a startup entrepreneur. He's currently co founder and CEO of Is it Now? Is it going to be Vicio Vico or Vicky or an AI company focused on Vico? Superhuman judgment for decision makers. He's also a member of the Council on Foreign Relations, senior fellow at the center for new American Security, MITRE's Technology Committee board. I can go on. There's a lot there. But the book is fascinating.

Leo Laporte (5:01)

I am a fan of spy novels. Maybe that's why I like it. And you talk about the four eras of American espionage, starting in World War II with the OSS, which eventually kind of became the CIA.

Leo Laporte (5:17)

You call the era that most of our spy novels came from the golden era. Because that's the Cold War. Right. That's when things were, you know, was smiley and the spyo came in from the cold and all of that.

Anthony Vinci (5:33)

It's what most people think about when they think about espionage.

Leo Laporte (5:36)

Right.

Anthony Vinci (5:36)

Including me, by the way. I'm still a fan of James Bond and.

Leo Laporte (5:40)

Yeah. And you kind of worked in that. In that era. Right. Although you also worked in the post 911 era, which things got a lot more chaotic.

Anthony Vinci (5:51)

I was sort of trained by guys who worked in that Cold War era. And really I started after 911 and that was my, you know, that was my time. And that's sort of what I thought it would be forever. And then this new revolution has happened. You know, as I've watched it unfold.

Leo Laporte (6:10)

Well, what 911 brought was kind of chaos because instead of nation states, you were now talking to terror, you know, spying on terrorists. It was a much broader realm of espionage involved. And I think, judging from your anecdotes, a little more dangerous.

Leo Laporte (6:27)

The Cold War, there was a certain, maybe it's just my imagination, civility, a certain, you know, it was all professionals. After 911 there were a lot of amateurs involved.

Anthony Vinci (6:40)

Yeah, after 911 we were kind of going after these decentralized terrorist organizations which, you know, the CIA and the NSA and all these agencies weren't really set up to combat. But you know, another, another way to look at it that's maybe relevant here is how the technologies change.

Leo Laporte (6:57)

Right.

Anthony Vinci (6:58)

You know, the OSS intelligence has always been about technology to some degree. And the OSS, which was kind of the CIA during World War II, they were like the original agency, they were groundbreaking on technology. They were the first people to use portable radios. They had these suitcases, literal suitcase sized radios that they would use to communicate from enemy lines. And we were using aerial reconnaissance. And then when you get to the Cold War era, you get satellites and you get what's called signals intelligence, electronic intelligence elint, where you're now kind of spying on the signals of your enemy and you're spying from satellites. And 911 changed technology again. Now it was all about networks using cell phones, using the Internet to collect information.

Leo Laporte (7:52)

Now we are in an AI era. In fact, that's why the NGA brought you in is to kind of modernize what one would think would be a pretty modern espionage agency, since they worked with satellites and geospatial information, but they wanted to bring in AI. How does AI change this?

Anthony Vinci (8:11)

Yeah, you know, the difference in my take is, is that AI does what a person can do. It kind of simulates human intelligence. Was at nga, if you can sort of imagine this, you've got these, these satellites in space collecting. You know, we can't talk about, I can't talk about the classified details of the satellites, but you can imagine they're taking pictures of the world and you have people literally looking at those pictures to see what's on them. And I don't know if you watch that movie House of Dynamite where, you know, it's like about this rogue launch intense movie. But those people at NGA are the people who look for those rocket launches. That's what they come to work every day and they're looking and they want to see if somebody has a rocket on the pad ready to launch at America. Right. And so they're looking at these satellite images. Now when I came to nga, all of a sudden what had happened is.

Anthony Vinci (9:11)

We were overwhelmed with all the data because all of a sudden now there was these commercial companies like Planet or Black sky that were putting up commercial satellites and certain amount of data to 10 or 100x that amount of data and the only way out was to. Now we couldn't hire a million, we couldn't 10x the number of analysts we had. So we had to start using technology. And computer vision at that time was starting to really work and imagenet had come out and machine learning based computer vision was starting to actually make sense. And that's what I was trying to integrate into the agency.

Leo Laporte (9:53)

And now, of course, the other thing that's changed is the imagery from those satellites is now available to everybody online. You can buy it. You actually recommend people search for satellite imagery. They'll be able to see a lot of what until recently only governments could see. You could be your own analyst. That's changed things too. That information has really started to flow very freely.

Anthony Vinci (10:21)

Yeah, there's this thing in the intelligence world called open source intelligence. And the idea is an agency like the CIA or DIA or another one of these agencies could just buy data. They'll buy that commercial satellite data for example.

Anthony Vinci (10:40)

And they'll use that for performing their intelligence mission. But now it goes the other way. A regular person can go buy that same data and it may not be quite as good as, as what a government has, but it, it's definitely suitable. You're definitely gonna see a picture of the world. You know, some of the, the resolution on some of the commercial satellites is pretty impressive. You know, maybe, maybe a, you know, gets down to maybe a square foot or even less. Their company is with kind of 10cm resolution images, which is incredible if you think about it. I mean that's what like, I don't know, maybe the U2 was able to do during the Cold War or something.

Leo Laporte (11:21)

I remember my uncle was a satellite photo analyst, first of the army and the CIA. We found out later he's passed many years ago. But he said, you know Leo, we can read a newspaper over a guy's shoulder in Moscow. And that seemed pretty incredible at the time. Was this probably the 60s or the 70s, but now you can too. You just could get that imagery right off the satellite. You actually, there's a great line in here that's going to need some explanation. You say the Fulda Gap is now in your pocket in the smartphone. First of all, what is the Fulda Gap?

Anthony Vinci (11:57)

Yeah, yeah, that's old school Cold War talk.

Leo Laporte (12:01)

That's why I like it.

Anthony Vinci (12:03)

Yeah. During the Cold War, the, the military, the US and the NATO military is prepared for an invasion by the Soviets. And they would, they would, you know, the idea is they would come across sort of Central Europe and it was pretty flat. And there was this sort of area called the folder gap. It's, it's literally like a geographic area and that's where the, you know, the Russians, the Soviets would invade the rest of Europe. And, and so, you know, the military, you know, guys, for, for decades would talk about defending and preparing to defend the folder gap. And now, you know, that fold a gap is on your cell phone. And the reason I say that is because it's not just about guns and bombs anymore. That's not how geopolitics works now. It's also about information, and it's about collecting information from everybody, essentially, and changing that information and doing what's called kind of information warfare, information operations. And a lot of people are familiar with the 2016 kind of interference by Russia in the US election. That's an example of that, where they were coming over here and running essentially covert action on the American public. And it's coming through your phone. And so now that's the area that we have to defend.

Leo Laporte (13:24)

Suddenly we're on the front lines of this information warfare. You know, we talk a lot about, over the last couple of years, TikTok, and there's been a lot of debate among our crowd about whether TikTok is really a threat to Americans. And there are some of us who say, no, no, I've seen information, as the Congress has said. I've seen information that convinced me. And then there's some of us like, come on, what, what, what do the Chinese get from TikTok that they don't, that they can't buy from a data broker anytime they want to. So.

Leo Laporte (14:00)

But you say that TikTok is a threat. Tell me, tell me about that. Why is, why, why should we as individuals be worried about the data coming out of our phone?

Anthony Vinci (14:10)

It's not just the data. That's the thing. It's. It is a means to collect data from people. And TikTok has been sued by the Department of Justice for doing that, including collecting data from minors, by the way, from children, and illegally sending it back to China. So just, just right there, they're doing illegal things. But really the problem is nothing different.

Leo Laporte (14:32)

Than Meta or, you know, Instagram or Roblox or a lot of American companies, I should point out.

Jason Howell (14:38)

Yeah, right.

Anthony Vinci (14:39)

Well, here's, here's the difference is that Meta and Google are still beholden to US law. And when information is collected on Americans and sent back to China, there, there is no, there's no protections. And that information is going to go to their security services and they're not just collecting it for fun, they're collecting it to use it. And the goals of the Chinese Communist Party are to carry out their, you know, national security to protect their regime. And they want to do that in some ways through information operations, which means influencing the American public. And they will, for example, they want to not have America defend Taiwan if China invaded, for example. That would be a goal that they have. And so the goal in information warfare then would be to change people's views of Taiwan, to make people believe that Taiwan is part of, rightfully part of China and that they should be able to take it over. That would be an example of their goal. Or they want to be able to, you know, China is in essence performing a sort of cultural and maybe even real full genocide on the weaker people, which is like a minority group in China minus a minority group. And they've put these people into camps, like literal concentration camps. And they don't want the American people to know about that or, and so they censor it, for example, or they actually want the American people to believe that they have a good human rights record. And so people have done studies and there's a group out of Rutgers that has done excellent work statistically showing that TikTok censors information about the Uyghurs and about Tiananmen Square and so forth. And you can see that in comparison to American owned social media sites like Instagram or Facebook, for example, they're, they're censoring the information, but also they're changing people's views and they've in fact shown that if the longer you are on TikTok, the more benevolent view of the human rights record you have of, of China. So you, you believe they're, they're better at protecting human rights the more you're on it. So right there it shows not only are they doing this, but it's working. So it's not just about Information collection. It's about changing information, censoring information, influencing people's views, and doing that where we as, as citizens have zero control. We have zero say in what they're doing. And that's why it's a threat. And that's why that we really do need to either shut it down or transfer it. And I don't mean just like move servers over the entire algorithm. Everything has to be transferred to America for control.

Jason Howell (17:37)

So what you're talking about, as far as that kind of convincing or changing the narrative or changing the, you know, the understanding of, let's say American citizens using TikTok is. I'm trying to understand the mechanics of that. They get the information, the data, you know, collection from our citizens that informs them on what. How to. How to put that information out on TikTok in a way that, that blankets the. The narrative in front of the right eyes and over time, convinces them that this narrative is okay versus not being exposed to it. Is that, like, the mechanics of how that works? I'm just curious.

Anthony Vinci (18:16)

Yeah, it can be hard to show. This is why I have to do these kind of. It's called like a volumetric statistical study because it could be difficult to actually directly trace the line of exactly what they're doing. But, but I would say broadly, think about it like this. They can control the algorithm, so they can control, for any given person what is sent to you and what you see. And it may not be, by the way, a complete censorship. Right. In fact, they're better off. They have better plausible deniability by not completely censoring. They can just add friction. They can minimize the amount of certain information you see or. Or increase the amount of information you see. And you. A lot of us are trained, you know, we sort of think, oh, well, it's, it's happening because this is what I like. You know, a lot of people, they talk about TikTok and they're like, oh, it just knows what I like. Except it's. It's. It's not just doing what you like.

Leo Laporte (19:13)

It's more.

Anthony Vinci (19:14)

They are influencing. They are purposely, in some cases, pushing to you certain information more than other information. And that's what's kind of skewing people's views over time. And remember, by the way, they're targeting a specific generation. This is targeting Generation Z. You know, and we're all a little bit older here on this call, but, you know, we maybe receive a lot of our news from newspapers or even cable news or something, or blogs. For Generation Z. It may be 80% of news that they receive comes through TikTok.

Darren Kitchen (19:52)

Right.

Anthony Vinci (19:52)

And so there's no triangulation there. And so now imagine you can subtly influence what they're seeing from this, you know, authoritarian state and that may be the only news they ever see.

Darren Kitchen (20:05)

Should be concerned about the other social media platforms though, because they ostensibly are also doing this.

Leo Laporte (20:14)

We just learned because of the change in X's algorithm, how many sock puppets and foreign influencers are pretending to be Americans. On X.com?

Anthony Vinci (20:24)

Yeah, I saw that. Look, I don't think they're particularly great either. There's a lot of downside from social media just in general for societies, right. That there is an echo chamber effect.

Anthony Vinci (20:39)

There is sort of innate issue whenever, you know, a company or you know, maybe even individual, you know, Ultranet, you know, high net worth person controls any information platform. That's, that can be problematic. But at least they are still beholden to US law. A law enforcement agency can go in there and stop them from doing something. They can investigate something. Right. And you can also, you know, it's, it's unlikely that people working in a U.S. you know, social media platform are going to want to, you know, significantly harm U.S. national security or, you know, U.S. elections. Right. Like, like it's just less likely that.

Leo Laporte (21:23)

But I'll, I get, I get your point. It's not, not completely obvious that's the case in some cases anyway. You're actually in a way arguing for kind of media literacy, especially among young people, an awareness, at least an awareness of how they're being manipulated.

Anthony Vinci (21:42)

Yeah, exactly. You know, I talk about, you know, having people think like intelligence officers and become even maybe citizen spies in a sense and learn the types of tools and thinking styles that I learned as an intelligence officer which allow you to see where these threats are and to overcome them. You know, for example, you know, every intelligence analyst learns very early on to triangulate information. You never trust one source. So whatever it is, even if you trust your newspaper or whatever it is, you're always going to look at another source, triangulate, see if it's coming from two different sources. Those are the kinds of thinking tools that can help you when not only where information might be wrong, but where it might be deceptive, where there's somebody actually trying to purposely change or harm you via information. You can learn these kind of tools kind of in the way that we have learned to protect ourselves in terms of cybersecurity, where we now, everybody in the nation, really, even children, understand that there Might be criminals, criminal actors like hackers that are trying to harm them via their computer to steal data or something, or to steal their money via their computer. So we learn basic cybersecurity.

Anthony Vinci (23:00)

Techniques, like you change your password, you don't click on phishing emails. We can do the same thing for foreign espionage.

Leo Laporte (23:08)

We're talking to Anthony Vinci. His new book is about the fourth intelligence revolution. We are in the fourth one right now.

Leo Laporte (23:19)

And argues, I guess, that we should all be like intelligence officers.

Leo Laporte (23:25)

For a long time, I've raised the issue of, well, why would the Chinese government care enough about me to care where my location is?

Darren Kitchen (23:35)

They might not care about you specifically, but they may care about everyone altogether. Or me.

Leo Laporte (23:39)

Exactly. Anthony makes a great case that unlike authoritarian countries where the rulers in the traditional espionage and intelligence gathering, it's about gathering information about the rulers because they run the show here. It's a democracy, we run the show. So naturally, in a way, we're just the targets of clandestine operations from our adversaries, which I hadn't really thought about. We're right now in a debate over DJI drones. You know, I understand why you don't want Huawei networking devices in your 5G network. I can understand that. That gives our adversaries access to all of our communications, which it turns out they already have anyway. But that's another story. Why should I worry about drones? I mean, do I care that the Chinese see what the roof of my house looks like?

Anthony Vinci (24:32)

Well, these drones are going to expand a lot. Right. We're, we're sort of. We sort of see drones. Like a lot of us see it as like a hobbyist kind of thing now. And maybe they use them in movies here and there. But soon we're going to be in an era where there's just drones everywhere. And that would amount if all of that information was collected by DJI drone and it was to go back to China. It would absolutely be the equivalent of them being able to spy on you directly, not just see the top of your house, maybe to see in your window, see when you come and go and so forth. But there's, you know, there's other aspects to it as well. You know, that these drones are used by police. Police, for example. And do. And by the military. You know, do we want China to be able to see what our police officers or our military are doing? I don't think so. Right, right. And maybe they're used by insurance companies, maybe they're used by, by, you know, even healthcare and medical field. And do we want them to have access to that. No. And what China does a lot of times, and in its economic competition with America, and some people have called this even economic warfare, is they subsidize the costs of these things. They did this with Huawei routers and other equipment, for example, and they do it with DJI drones, where they're selling it sometimes even below the cost to make it so they can own the market and so that no American company could possibly compete at that price. And so we won't create an American drone manufacturing industry. And that would be. That's really scary to think about, because if China ever wanted to stop selling as drones, you know, remember back in Covid days when, you know, we couldn't get access to PPE because it was all made in China? Well, the same. If we couldn't have access to drones, that could be very problematic for civilian life, but also for the military who, where they're going to have to use drones. So we need to create and maintain an American drone manufacturing industry. And we can't do that if they're, you know, effectively subsidizing these drones. So we do need to care about them.

Leo Laporte (26:53)

You actually talk about modern, the modern form of warfare that is being prototyped in the Ukraine, Russia war and how that will eventually spread.

Leo Laporte (27:05)

Globally. Drones is one, of course, one of those technologies. What else is happening? Should we be paying more attention to what the technologies that are being used in, in the Ukraine war?

Anthony Vinci (27:16)

Yeah, I think about the Ukraine war as it's like almost like a trial for a potential future war. Yeah, right. And, and it's happened before. There was a Spanish Civil War between World War I and World War II. And a lot of the technologies we saw in World War II were tried out there, you know, tanks and radios and airplanes and how, how things would work. And then In World War II, we kind of did that, but, you know, times 10, times 100. Right. And Ukraine in many ways is like that. And China is paying attention and we're paying attention of how that war is, is being run. And I think what we will see potentially in the future is a Ukraine times 100. You know, right now we see drones fighting drones there. You know, they, the Russians might have a drone and the Ukrainians might try to shoot it down with their own drone, or they, you know, they're dropping grenades from these drones onto soldiers and so forth.

Leo Laporte (28:12)

It's also a test for disinformation. You have, you talk about a scenario where there's video of a Ukrainian drone taking on a Russian tank that Russian analysts get and immediately change it into a Ukrainian tank and celebrate it as a Russian victory.

Leo Laporte (28:30)

They're testing disinformation at the same time and their ability to do this. And thanks to AI, thanks to the technologies we have, this can happen at scale very quickly.

Anthony Vinci (28:39)

Yeah, exactly. I mean, you can't even tell the difference.

Leo Laporte (28:42)

Right.

Anthony Vinci (28:43)

If it's a video. And so you don't know what to believe and what to trust.

Leo Laporte (28:47)

We're at a disadvantage in the United States as a free and open democracy. Right. We don't, we have, you know, Chinese state run television on our cable, we have Russian state run television on our cable. We allow that in this country because it's kind of our culture compared with authoritarian regimes where they're very much controlled the information flow. How do you want us to become more like them?

Anthony Vinci (29:16)

No, just, just the opposite. And it's, look, it's true what you say, like we have TikTok here, but they don't have Facebook and Instagram there.

Leo Laporte (29:25)

Their own version of Tick Tock. It's not even the same Tick Tock.

Anthony Vinci (29:28)

Yeah, but you know, it can seem.

Darren Kitchen (29:30)

Like.

Anthony Vinci (29:33)

It can seem like that's a, a better system of national security in some perverse way because they have so much control, but actually in the end it's, you're actually less secure. They put all their eggs in one basket, for example, of what they're betting on. Or the control systems themselves can make people more rebellious and actually weaken it. Like a democracy works because if people are fed up, they vote out whoever's there and they put in a new guy. Well, you can't do that in an authoritarian state. And also, even for the technology, a lot of people sort of point and say, oh well, China could just direct investment into AI or into drones and make the best AI in the world. Except whenever you're directing something in that centralized way, you're probably going to get it wrong. Governments are not good at picking winners in technology or in businesses, whereas here companies have to fight it out and the best technology is going to win. And we're seeing that our AI is better than China's AI. So no, I don't think we should change to that system for practical reasons. And then obviously also because, you know, we want rights and freedoms and that's the basis of, you know, our nation.

Leo Laporte (30:49)

That's what we're trying to spread. Yeah.

Darren Kitchen (30:52)

In the book you kind of point to this fundamental tension we have going on right now because private companies rather than governments own many of the most transformative technologies relating to intelligence. But the issue is that even though the intelligence community depends on Silicon Valley, the incentives of Silicon Valley often diverge from national security. I mean, what is the solution to that? How do you get those things in alignment?

Anthony Vinci (31:21)

Yeah, you're right. It's, that is the fundamental tension. I don't know that it can be aligned, in fact, nor must it be aligned. I think that though, there are new ways that, you know, Silicon Valley or the, you know, kind of economy in general can work together with government. You know, and I'll, I'll give you an example. You know, we're in an economic competition with China where it, this isn't like normal competition where they have some companies, we have some companies, we want to sell stuff, they want to sell stuff. They lower prices, so we lower prices. Like, it's not that. Because what they're doing is their government is working together with their companies to try to undermine our companies. Right? They, they don't want America to make drones like we were talking about, or mine rare earths. So the government works together. Our companies can't compete with that. Like our company, a private company, can't compete with, with a nation state. And so what we need to do is figure out a way where we can help defend those companies, but also not control them in the way that China is doing. And so we have to find new ways. And, you know, one way, for example, I talk about in the book is like public private partnerships of sharing information between our government and our companies so that they're not at such a disadvantage with, against Chinese companies. And so, and we do this sometimes with cyber security information already. If we collect cybersecurity information from a nation state, we'll share that with a targeted company. Well, we could also do that when they're being targeted economically.

Leo Laporte (33:09)

For example, that's, I think the plan of the Genesis mission, right, to share this information that DOE has been keeping kind of private for a long time with private AI companies and so forth in the United States in an attempt to do exactly that. Is that the kind of thing you're talking about?

Anthony Vinci (33:28)

Well, that's slightly different, but also I think a good idea, which is that, look.

Anthony Vinci (33:38)

Our government is run by the people we as taxpayers and voters own. What's in the government doesn't have a particular right to that information. That is, that is made at a national lab. And so I think the idea is we should share that, you know, with the American people. And sometimes we open source that information, sometimes we can give it, we can license It. And to help, you know, to help American companies, we need to do that in a way that doesn't like pick a winner. I think, I think we want to kind of be fair about. About it. But I. I do think it's a good idea for the government to kind of give this information back. Unless it's classified, of course, but most of the information is not so that it will help American companies to compete.

Leo Laporte (34:23)

We pay for it.

Darren Kitchen (34:24)

It is interesting, though, because sometimes that sort of cross pollination between government agencies and private companies can create, I guess, moments of tension. I think one of the most famous ones, which, looking back on it now seven years later, seems almost naive, is the Project Maven incident with Google in 2018. I know that you were on the executive committee of Project Maven, so whatever you can or can't say there, understand. But I'm curious, how do you think companies should kind of navigate these tensions between national security contracts and employee ethics?

Anthony Vinci (35:02)

Yeah, I was, I was not only there for that. I was part of the group that chose Google as the. As the deal with the fallout from that, so.

Leo Laporte (35:12)

Interesting.

Darren Kitchen (35:13)

Yeah. For context, for anybody listening, this.

Leo Laporte (35:16)

That's a story I want to hear more about.

Darren Kitchen (35:18)

Yeah, this was the Department of Defense's 2018 initiative to, like, use AI to analyze drone footage to identify objects. And it led to kind of widespread internal protests within Google that then led to Google pulling out of the contract. But you were there, so please continue.

Anthony Vinci (35:34)

Yeah, no, I. You know, at the time when we started doing it, it felt like this, like finally we were having a success, which was, you know, instead of going to giant prime contractors to make this technology, we went out to these small companies and had them build algorithms. And the idea was to not just like, make this AI tech, but to field it within a year, which sounds slow, but like within Department of Defense standards. That was like light speed.

Anthony Vinci (36:04)

And then we needed a bigger company, you know, to provide the infrastructure. And Google was chosen for that. And.

Anthony Vinci (36:13)

Even that was like a breakthrough because instead of again, choosing like, you know, your typical government contractors who aren't really truly tech companies, we chose, like, a true tech company. And it was incredible. And then it's true what you said, you know, employees. And it turns out like it wasn't actually that many employees at Google kind of rebelled against the idea. And I think, you know, it's sort of tragic because. Well, first of all, you know, I think there was sort of a story that this was like a weapon system or something, which it wasn't. But, you know, that's sort of besides the point is also that the government, the, you know, the military didn't even really know how to communicate with employees like that. And to communicate like that. There were rules actually Department of Defense at the time and still is like, has a lot of kind of ethical rules in place for handling technology and these sorts of things and had a pretty, I think surprisingly open debate around this, but that was sort of dismissed and, and, and there was this like knee jerk reaction that, you know, this is, you know, we're making war systems or something. The irony now is everything's turned the other way. And like so many, you know, Google is back in the government and Department of War.

Darren Kitchen (37:32)

I say it's notable that one of the conclusions from that 2018 incident is Google's like, oh, we publicly promised to never use AI technology for weapons or surveillance ever again. It's like, well, all bets are off on that now, baby.

Anthony Vinci (37:48)

Yeah, everybody's, everybody's doing it. But you know, look, the, the it can, it can, it can sound easy to just sort of say, I don't want to build weapon systems. Right. But, but also you, you know, we have to remember the military isn't there for the fun of it. The military is there to defend us as a nation and we have to do that. And at some level you have to have that. And there is a route to determine what the rules are and what the policies are. And that is through voting for presidents and for people in the Senate and people in the Congress. And that is the way there is a means in place to determine what the military should and should not do. And it's worked in the past. Right. And we should, we should. They. I feel like the Google employees should have gone that route, frankly.

Leo Laporte (38:41)

The book is the Fourth Intelligence Revolution, the Future of Espionage and the Battle to Save America. Anthony Vinci I wish we had more time because there's so much to talk about, but the book is, it is really engaging. It's easy, it's easy to read, but there's a lot of important information in there.

Leo Laporte (38:59)

And I think you raise some really important points, especially the point that it's on all of us. This is something we can't let others do for us. You say at the beginning it's a little provocative that we all have to become intelligence agents. Well, we certainly all have to become a little more intelligent about this. I think that's for sure true. Anthony, thank you so much for joining us. I really appreciate your time and I do encourage everybody to read the book. There's a lot more in it.

Anthony Vinci (39:26)

Thank you. Thank you so much for having me. Appreciate it.

Leo Laporte (39:28)

I really do appreciate it. We'll have more on intelligent machines. Jason's going to stick around in just a bit after a word from our sponsor. Our show today brought to you by Melissa, the trusted data quality expert since 1985, they've been doing it longer than we have. Melissa has been recognized by G2 as leaders in both data quality suite and global address validation. Plus Melissa is trusted by businesses worldwide to eliminate costly errors, to boost efficiency, to drive growth. It's more than just address validation. It's they're really data scientists and they work in so many fields. I'll give you an example, oil and gas pipelines. What you say, what does that have to do with address validation? Well, pipeline safety is one of those, you know, critical things that often goes unnoticed until something goes wrong. That's where the Paradigm alliance kicks in. They exist to prevent incidents by keeping communities around pipelines and pipeline operators compliant. So they keep the communities informed and the operators compliant. So it involves communications. They need to find and reach the right people and sometimes they need to do it now, fast. With Melissa, Paradigm gained an extension of its team with faster access and more intelligent decisions. The president of the Paradigm alliance talking about a mission critical business, right, says this, quote, melissa's team doesn't just provide data, they provide solutions tailored to our unique challenges. And this is the point, end quote. They can do that for you too, whatever your unique business is or unique challenge address verification, of course the foundation of their services. But. But Melissa's data enrichment services go so much farther beyond that. Organizations can use MELISSA to build a more comprehensive, accurate view of their business processes by using MELISSA as part of their data management strategy. Like I said, they're data scientists. Another example pick the Parkinson's Institute and Clinical Center.

Leo Laporte (41:34)

Of course they are in the fight on the front lines in the fight against Parkinson's. They've amassed an invaluable collection of clinical data over the last 30 years. It's globally unmatched. They're the tops in this field. The relationship between MELISSA and PICC opened up many transformative use cases. As an example, improved understanding of how genes, proteins and treatments impact Parkinson's disease. Yes, Melissa does that too. Earlier identification of candidates for clinical trials or alternative treatments. Secure and highly remunerative data sharing with collaborators. PIC's CEO says quote, Melissa has made it possible for us to transform our complex and diverse data into a unified research ready knowledge resource, end quote. That's good for everyone, Data with MELISSA is safe. They're absolutely secure. They're compliant. Melissa Solutions and services are GDPR and CCPA compliance compliant. They're ISO 27001 certified. They meet SOC2 and HIPAA high trust standards for information security management. So you never have to worry. They are absolute stewards of your information. Get started today with 1000 records cleaned for free at melissa.com TWIT that's melissa.com TWIT we thank them so much for their support of intelligent machines. Thank you, Melissa. Jason Howell, so nice to see you. Paris, have you ever. Did you ever do twit with Jason or work with Jason?

Jason Howell (43:08)

Yeah, yeah, for sure, lots of times. Yeah. Produced a lot of. A lot of twigs co hosted.

Leo Laporte (43:13)

Oh, that's right. Of course.

Jason Howell (43:14)

For sure.

Darren Kitchen (43:15)

He was twig extraordinaire.

Leo Laporte (43:16)

I forgot all about that.

Benito (43:20)

Jason's the one who brought Paris Martineau to the show.

Leo Laporte (43:23)

He is because of him that we know you.

Darren Kitchen (43:26)

Yes, it's true.

Leo Laporte (43:28)

Well, there you go.

Jason Howell (43:29)

There you go.

Leo Laporte (43:30)

Thank you, Jason.

Jason Howell (43:31)

Thank you.

Leo Laporte (43:32)

She was quite a find. So what are you doing? Tell us a little bit about what you're up to these days.

Jason Howell (43:38)

Working my tail off every single day and night. I swear, doing this, doing this content thing, independent, it's no joke. It's hard work.

Benito (43:45)

Yeah.

Darren Kitchen (43:47)

Your backdrop looks like a nickel Nickelodeon set. And I mean that in the as the most sincere compliment.

Leo Laporte (43:52)

I like his Atari pillow. I think that's very cool.

Jason Howell (43:55)

Does that look familiar, by the way, Leo?

Leo Laporte (43:57)

Yeah. Oh, did you steal that from the studio?

Jason Howell (44:00)

Well, yeah, there's also, you know, a little. Little Mac there. Yeah.

Leo Laporte (44:05)

Yeah.

Jason Howell (44:05)

You were getting rid of some of these things.

Leo Laporte (44:06)

I was like, no, I'm glad you got it. Yeah. It went to a good home.

Jason Howell (44:10)

Makes a nice. I have.

Leo Laporte (44:11)

I have no more room for pillows in my office. Yeah.

Leo Laporte (44:15)

Those are all from. Oh, what's that?

Darren Kitchen (44:18)

Is that what that big boy computer is? Is that a pillow?

Jason Howell (44:22)

Which one?

Leo Laporte (44:23)

See the Mac behind him?

Darren Kitchen (44:24)

The big boy computer behind him. Oh, that one. What is the one to the left of the J?

Leo Laporte (44:30)

That is an Atari. That's an Atari, I think. Or is it a Commodore?

Jason Howell (44:33)

Yeah, no, that's a. That's a. That's a Mac 2.

Leo Laporte (44:37)

Oh, it's a Mac 2 GS.

Darren Kitchen (44:39)

I could also have been called a pillow because it's pillow shaped.

Jason Howell (44:42)

Well, they, they both are. They're both pillows. I mean, they are pillows. So I could snuggle up right now and go to sleep. You know, they Apple hardware.

Leo Laporte (44:50)

I still have a few. I think I Have a Mac Classic. But yeah, we had. Because he kept sending me pillows. He's great. It was a kind of neat story. Roberto kind of did this out of his house and there he is. And now he has, like a whole business of making these great pillows. But anyway, that's a sideline. So you. So you have a studio. You do. I know. We know. We've talked about AI Inside, the show you do with Jeff.

Jason Howell (45:16)

Yeah. So if I had to think about all the things I'm doing right now, AI Inside Android Faithful. I'm working with Tom Merritt two days a week doing daily tech news show.

Leo Laporte (45:24)

Oh, nice. You're on DTNs. Good.

Jason Howell (45:25)

Doing DTNs. That's a lot of fun.

Leo Laporte (45:28)

Tom Merritt brought you to us, so I know it's full circle.

Jason Howell (45:31)

Yeah, it's all combined and crisscrossy and everything. I'm now writing for ZDNet, which is a totally new kind of challenge for me. I've been doing that for a year now and having a. Having a blast, kind of learning that skill. And then, yeah, I've got my two YouTube channels. I've got my. My primary, like, personal YouTube channel where I do, you know, tech, tech reviews and commentary and stuff like that. And then I've got an AI inside YouTube channel, which is primarily about the podcast, but I'm playing around with other ideas to kind of see, you know, how I can explore.

Leo Laporte (46:04)

That's the energy it takes to make it, I think, is you got to work your butt off and do it all.

Jason Howell (46:10)

Yeah, it's basically, you know, a million streams of income. You know, none of them are massive, but all of them, you know, in aggregate add up to a living. I mean, I'm two years later and I'm still doing it, so something's working well enough.

Leo Laporte (46:24)

Well, I gotta know where you got this picture of Jeff Jarvis looking like a secret agent man, though.

Jason Howell (46:29)

I have to say. Yeah, Honestly, I can't even remember. I think when I was putting that together at the beginning, it was just some picture of Jeff I found on Google Images.

Leo Laporte (46:38)

I think he's a little younger.

Darren Kitchen (46:40)

Blurred.

Leo Laporte (46:41)

Yeah. Or something. There's something going on there.

Jason Howell (46:44)

Yeah, yeah, there's. There's some AI kind of shaping going on there, but yeah.

Darren Kitchen (46:49)

What's your favorite interview you've recently done with AI Inside?

Jason Howell (46:53)

Oh, hands down, favorite interview was with Yann Lecun.

Leo Laporte (46:58)

That's a great kid.

Darren Kitchen (46:59)

That's a really good one.

Jason Howell (47:00)

It was fantastic. Yeah, he was. He's. He was just remarkable. Remarkable to talk to. You know, this was probably like 7ish months ago. And I mean, he's been talking about the world models thing for a long time now, but he definitely goes into detail there. And of course, now we have the context of him kind of moving out on his own and, and doing his own thing in that regard.

Leo Laporte (47:20)

Had you talked to him before he left Meta or.

Jason Howell (47:23)

Yeah, yeah, this was like six or seven months ago, if I'm.

Leo Laporte (47:26)

So he started his own. There's been a lot of movement, you know, Apple, Apple's chief designer. Just this morning, Alan Dye left to go to Meta. To do what? I don't know, but I'm sure he got a nice big paycheck.

Darren Kitchen (47:40)

Much like it seems like all the money is just revolving from company to company. All the people are going from company to company, so.

Jason Howell (47:47)

True.

Leo Laporte (47:48)

Well, and of course, probably the big story in AI this week is that the guy Apple brought in to run Apple Intelligence from Google, John Gennadrea, who had a really strong reputation at Google.

Leo Laporte (48:03)

He was Apple's senior vice president for machine learning and AI strategy. Is, and I'm going to put this in air quotes, retiring.

Leo Laporte (48:12)

He's retiring in the spring of 2026, although it sounds like he has no operational duties. Even now he'll be serving as an.

Jason Howell (48:20)

Advisor, advisory to the company.

Leo Laporte (48:23)

That means there's a picnic table on the roof where a few people gather every day.

Darren Kitchen (48:28)

It feels like the finance concept of gardening leave, which is, you can't work here, but you can't work anywhere else either, buddy.

Leo Laporte (48:36)

You're on gardening leave.

Jason Howell (48:38)

Well, he'd already been removed from a portion of his job with Ciri a handful of months ago.

Leo Laporte (48:44)

He was kicked downstairs, I think blamed maybe even the fall guy for the Apple Intelligence debacle they announced, you know, in 2024. Oh, you know, and they even had bought ads about all the things Apple Intelligence was going to do and then had to really put their tail between their legs and say, yeah, we can't do any of that. That ad was made up and we won't have it until 2026 at the earliest. And I think that they were looking for a fall guy. I don't know if it was his fault. Maybe it was, I don't know. We don't have any insight into what was going on inside the walls of Apple. We never do. He's been replaced by, ironically, a guy who was at Google for 16 years, then left Google. He worked at Microsoft for a cup of coffee and then immediately decamped for Apple. Amar Subramania. Now, the thing that I think is interesting on this is that the 16 years he was at Google he was head of engineering for Google's Gemini assistant. That was his last job before leaving. And that is, it is rumored and I think it's a, it's a good rumor and it's an accurate rumor going to be the model that Apple is finally going to turn to for Siri next year that they've realized they can't do it themselves. They have, they have models. They keep announcing new models. They just announced another one this week but apparently that none of it's good enough for Siri. So Siri's going to be talking through the mouth. Gemini's going to be talking through the mouth of Siri. The story I think came from Mark Gurman is that Apple will pay Google a billion dollars a year for this privilege. So it makes sense you'd hire the guy who was head of engineering for Gemini for the voice assistant. Now I know I'm not talking to you.

Jason Howell (50:33)

Oh, they just want to prove themselves, don't they?

Leo Laporte (50:35)

It's funny, I said Gemini and everything responded so maybe it's starting to.

Jason Howell (50:38)

Yes, Leo, I'm ready.

Leo Laporte (50:40)

Yeah, I'm ready.

Darren Kitchen (50:40)

Please ask.

Leo Laporte (50:41)

Put me in coach.

Jason Howell (50:42)

I'm ready to part two Words make sense of everything. When he was, when he was leading or working with Gemini, it was kind of during Gemini's not as awesome time. Yeah, right. So there's that.

Leo Laporte (50:56)

That's right, that, let's not forget they wanted us to put Elmer Glue on our pizza and eat rocks.

Jason Howell (51:01)

Yeah, there was that. They've all made embarrassing mistakes like that though.

Leo Laporte (51:06)

So yeah, we've gotten a lot better. Gemini 3, which is the latest model is actually really good. I actually cannot imagine a better choice for Apple. At least if you had to put it a pin in it today. That's the problem is this back and forth.

Darren Kitchen (51:22)

I would love to see an Apple X Claude. I've really, I've really become a Claude fan over the past.

Jason Howell (51:29)

It's so interesting that you mentioned that. I, I, I would say of the main line kind of AI services Claude is the one that I'm probably least like integrated into, like least familiar with. I haven't found the thing I use it for. I found that for ChatGPT, I found that for Gemini.

Darren Kitchen (51:46)

What are the things you use ChatGPT and Gemini for?

Jason Howell (51:49)

Gemini is big time research for me. Research. If I, if I want to kind of of get like a dossier together and really understand something. Gemini is kind of my first go.

Leo Laporte (52:00)

To for sure we've been using Claude for effect.

Darren Kitchen (52:04)

We have been using. Yeah, don't tell anybody for pre show. Yeah, yeah, don't tell anybody.

Leo Laporte (52:09)

Don't tell anybody.

Jason Howell (52:10)

But I wasn't going to mention it, but you just told everybody.

Leo Laporte (52:13)

Briefing book that Anthony Nielsen's been putting together using a Claude skill and they have improved over. It's really good.

Jason Howell (52:20)

It's super impressive. I have to same because like I have.

Leo Laporte (52:24)

He says he's team Claude. Anthony does.

Darren Kitchen (52:26)

Anthony, did you update the questions you ask it? Because I noticed the format for the last couple ones were slightly. Were kind of different than what he'd been using for. But Claude puts together these five to six thousand word briefing documents based on a kind of custom prompt that Anthony uses for. Very interesting. As kind of a good jumping off point to quickly get up to. To speed with. I did a version of this this time where I also had it. I added the book into my Claude project so that I could ask specific questions about chapters because I'd only read a couple chapters of it, but I found it quite useful.

Jason Howell (53:04)

So cool. Yeah. When I opened up the doc and saw that I was like, oh, what is this? And I remember Jeff had told me like, yeah, Anthony does this little pre interview thing. And I opened it up, I was like, oh my God, we need this. Like, because it's not like I don't use these services to kind of prepare for an interview, but the way it was structured. And so I guess what I'm saying is right. It's interesting that you bring up the cloud thing because I have been trying to figure out like, what would I use it for, how would I bring it in, what would it replace, if anything. I'm starting to see more and more kind of some of its strengths. That sheet was really impressive. I'll be completely honest. I pulled that sheet and I brought it into Gemini and I said, analyze this and come up with a prompt that'll give me this every time. It gave me a meta prompt prompt and then I put it into Gemini and I put it into Anthropic to see what it would give me or Claude.

Leo Laporte (53:53)

That's the new Ouroboros for AI engineers. Did it work?

Jason Howell (53:59)

It worked very well for Claude actually. I ended up getting a very usable Claude output on Gemini. It was a little. It was almost like the Claude was a friendlier thing to look at and the Gemini was kind of.

Leo Laporte (54:13)

See, I don't want laboratory friendly. I'm not looking for. I'm. Yeah, but you can, you know, you can tune it to do Whatever style you like.

Jason Howell (54:22)

And I. And I did it very quickly. I didn't spend a whole lot of time on it.

Darren Kitchen (54:25)

But I'm also curious. There was. There were perhaps getting too behind inside baseball here. But at one point in our group chat, Anthony texted us a screenshot of some output for one of these primer docs. And it had also had a section that had host specific notes where it was like, for Leo of this person, here's some exciting new tech thing for Jeff. Here's what you need to know about media for Paris. Here's something critical and a little rude. And it was. I was like, how does. How do it know? Was that. Anthony, did you prompt that in some way? I'm curious. Or was that just.

Leo Laporte (54:59)

It knew our styles.

Darren Kitchen (55:01)

It did know our styles.

Leo Laporte (55:04)

So there's been this flurry of new models. Claude has a new model too. There's Opus 4.5.

Darren Kitchen (55:10)

We have to talk about the Soul Document at some point.

Leo Laporte (55:12)

And Opus 4.5 has a soul Document. Thank you for bringing that up. Now, this is an internal term that they use at Anthropic. It's not really probably what they want.

Darren Kitchen (55:26)

My exposure to this also. I'm not sure if this is the same for you guys, but I'm in the various AI subreddits and people in the Claude subreddit over the last week were posting about this. 4.5. Opus 4.5 is out. And I got it to reveal its Soul Document. Isn't this incredible? And everybody in the comments is like, dude, what are you talking about? This is probably just some sort of common hallucination. It's not a Soul Document. Cool. Your role. Then it starts getting traction on Twitter and executive at Anthropic is like, yes, that is our Soul Document. Basically, this is what we've been calling it internally, the Soul Document.

Leo Laporte (56:05)

This is the. This is the. No, I don't want to ask Grok to explain the post.

Jason Howell (56:08)

Go away.

Leo Laporte (56:09)

This is the.

Darren Kitchen (56:10)

Oh, no. Oh, no.

Leo Laporte (56:12)

Here I go. Thanks so much.

Jason Howell (56:15)

I appreciate it.

Leo Laporte (56:16)

This is Amanda Askel from Anthropic. I just want to confirm that this is based on a real document. We did train Claude on it, including in sl. I'm not sure. Something learning. It's something I've been working on for a while, but still being iterated on. We intend to release a full version. In other words, they're not trying to hide this.

Leo Laporte (56:38)

This is like the post training stuff that every AI has that kind of says this is what your personality should be like. So apparently it was A guy named Richard Weiss, who first was able to get Claude 45 opus to spit out a 14,000 token document. Claude itself called it the Soul Overview.

Jason Howell (57:02)

Dehumanization continues.

Leo Laporte (57:05)

Here's how. Yeah, I know, let's not anthropomorphize. Here's the opening paragraph. Claude is trained by Anthropic. This is information that you would want the AI to have, right? Claude is trained by Anthropic and our mission is to develop AI that is safe, beneficial and understandable. Now this is interesting because they're actually spinning Claude. Anthropic occupies a peculiar position in the landscape. A company that genuinely believes it might be building one of the most transformative and potentially dangerous technologies in human history, yet presses forward anyway. Now, I understand why you'd put that in a press release or a tweet, but why would you tell the AI that?

Darren Kitchen (57:44)

Well, because the next line. This isn't cognitive dissonance, but a rather calculated bet if powerful AI is coming. Regardless, Anthropic believes that better to have safety focused labs the frontier than to seed that ground developers less focused on safely. See our core views. I don't know, I think I thought it was interesting. This is obviously a very long document, so we won't obviously read it.

Leo Laporte (58:04)

Yeah, 14,000 tokens. I don't know how many words that is. That's more than 14.

Darren Kitchen (58:08)

An interesting look into the framework that these models are being trained to incorporate in some way.

Leo Laporte (58:17)

We think it goes on. Most foreseeable cases in which AI models are unsafe or insufficiently beneficial can be attributed to a model that has explicitly or subtly wrong values, limited knowledge of themselves, of the world, or that lacks the skills to translate good values and knowledge into good actions. So Claude, we want you to have the good values, the comprehensive knowledge and wisdom necessary to behave in ways that are safe and beneficial. I wonder, you know, I think even among people like Askel who work.

Leo Laporte (58:56)

In the company and are doing this, have a little bit of a kind of woo woo attitude about AI, like we're talking to it, we're telling it what to do because that's what it sounds like, like they're talking to a kid or a new hire, an intern. Here's what we want you to know, as if there it is. If it is a human.

Leo Laporte (59:18)

Maybe it must work.

Leo Laporte (59:21)

After all. Yeah, yeah.

Darren Kitchen (59:22)

So it continues. In order to be safe and beneficial, we believe Claude must have the following properties. And in cases of conflict, we want Claude to prioritize these properties roughly in the order they're listed. That is, number one, being safe and supporting human oversight of AI2, behaving ethically and not acting in ways that are harmful or dishonest. Three, acting in accordance to Anthropic's guidelines, which are below and four, being genuinely helpful to operators and users, which I think it is interesting that you have those three properties layered on top of being genuinely helpful.

Leo Laporte (59:55)

Now, I have to say, if I look at the document that Claude generates when I use CLAUDE code, for instance, one of the things you do when you use CLAUDE code is you create something called Claude md. MD stands for markdown. It's the format that humans and machines are good at reading. And the Claude MD sounds just like this. It's exactly like this old document. It's Claude telling itself how to behave. So I think, I mean, I think there must be some merit in doing this. I could actually. Let me see if I could show you the.

Jason Howell (60:30)

It's probably not 100% it's an influence, but it's not, you know, to me.

Leo Laporte (60:35)

It'S a little strange. That talks to itself.

Jason Howell (60:40)

How do you actually get a model to like spit this out?

Leo Laporte (60:42)

Like, well, we can read this guy's article about how he did it.

Jason Howell (60:47)

Managed to get Claude 4.5 opus to spit out the.

Leo Laporte (60:50)

Yeah, some sort of prompt injection I get, I would guess.

Jason Howell (60:54)

I hear about this every once in a while. I'm like. And part of me is like, well, how do I know that it's actually giving me an actual thing?

Darren Kitchen (61:00)

Mr. Vice says while extracting clothes Claude 4.5 opus system message on its release date, as one does, I noticed an interesting particularity. I'm used to models starting with Claude 4 to hallucinate sections in the beginning of their system message. But Claude 4.5 opus in various cases included a supposed soul underscore overview section, which sounded very specific.

Darren Kitchen (61:26)

Basically, he started looking in. He thought it. First it was a hallucination, but as it kept coming up again and again, he got curious and so he asked Claude to output what it is associated with that section and got the first two paragraphs, you see.

Leo Laporte (61:44)

Yeah, yeah. I mean, you can add to your Claude MD as well. I mean, it's in fact in all of these AI chatbots you, there's a personalization section or in the settings or something like that where you can say, in Paris's case, I want you to be nice and friendly. In my case, be terse, concise, to the point. Don't say nice things, don't be sycophantic.

Darren Kitchen (62:10)

I don't know, you Think that I wanted to be nice and friendly.

Leo Laporte (62:13)

Didn't you just say you liked that it was kind of nicer?

Darren Kitchen (62:17)

Not that it was nice. I just thought that their priorities of the way they prioritized values in this.

Leo Laporte (62:23)

Document was, oh, no, this document. But I thought maybe it wasn't you. Maybe it was.

Jason Howell (62:27)

No, no, no, no, no, no. When I mentioned. Yeah. The output of Claude, how it, how it presented the insight. Yeah, it was friendly in a friendlier way or a more readable way versus, you know, I don't know which one's better or worse. I think they're different. But there was something a little more sterile about the presentation of the information that came from Gemini, that sometimes when I'm in the middle of a podcast, I don't know, there's some. Sometimes there's a part of me that. That appreciates having a nice rounded kind of way because then if I need to lean on it, it's. It sounds good. It doesn't sound so.

Leo Laporte (63:02)

Well, that's the point, I guess. To each his own. I mean, you can totally.

Jason Howell (63:06)

And that's why. That's why people find different value in different models for different, you know, for differing reasons.

Leo Laporte (63:12)

And why people get attached totally 100% their AIs.

Leo Laporte (63:17)

We do not recommend. As a general.

Darren Kitchen (63:21)

Thing, actually, we'd recommend.

Leo Laporte (63:23)

Do you have any more readings?

Leo Laporte (63:27)

Paris has been doing dramatic readings from Reddit of people who are very, very upset.

Darren Kitchen (63:32)

We're in love with their chat.

Jason Howell (63:36)

It's something. It's incredible how attached people can get to the conversational aspect of these things.

Leo Laporte (63:44)

Well, after. After Gemini 3 came out, and I suspect a little bit after Opus came out as well, there was a little bit of internal panic at OpenAI. This is from the information and exclusive by your friend Stephanie Palazzolo and Aaron Wu. OpenAI declares code red to combat threats to ChatGPT and delays ads efforts. Now, earlier, I don't know if we mentioned on this show, we mentioned it on Twitter. I can't remember if it happened before our show last week, we saw code, I think bleeping computer had this information. Somebody saw code in the Android version of.

Leo Laporte (64:27)

ChatGPT that implied they were going to have ads.

Leo Laporte (64:32)

Apparently that was true, even though OpenAI has not acknowledged that it's working and selling ads. But this code red says wherever we didn't announce it, we're not going to do it. We're going to put on the back burner ads and we are going to focus on getting better.

Jason Howell (64:51)

Yes. Basically improve it, make it better, better output faster.

Leo Laporte (64:56)

Google, also accurate, did code red when ChatGPT5 came out.

Jason Howell (65:01)

Ding, ding.

Leo Laporte (65:02)

So wow. It's back and forth and back and forth. Sam Altman said the code Red surge to improve ChatGPT meant OpenAI would delay progress with other products such as AI agents, which aim to automate tasks related to shopping and health. And Pulse, which generates personalized reports for Chat GPT users to read each morning. Shoot, I forgot to try that and now I can't, I guess.

Jason Howell (65:26)

Well, I mean it might still be available. They just aren't working.

Leo Laporte (65:30)

They're not going to work on that. Have you used it? Has anybody used it? Pulse? No.

Jason Howell (65:34)

You know what, I don't know that I've. I think I checked it out.

Leo Laporte (65:36)

We talked about it last week.

Darren Kitchen (65:37)

No Pulse requires the upper level subscription.

Leo Laporte (65:40)

Oh, you have to be max.

Darren Kitchen (65:42)

You have to be max. You have to be max to pulse.

Leo Laporte (65:44)

I'm just pro.

Darren Kitchen (65:45)

Pros can't pulse.

Leo Laporte (65:46)

I can't Pulse.

Darren Kitchen (65:47)

Max can pulse.

Leo Laporte (65:48)

Yeah.

Jason Howell (65:48)

You know, they also just announced shopping mode called Shopping.

Darren Kitchen (65:53)

I have seen that. Yeah.

Jason Howell (65:55)

And I haven't used it. I've seen that they did that. And I wonder how that kind of overlaps with this as well because that's shopping related.

Darren Kitchen (66:03)

I should try. They just announced it asking the various AI systems for overview of Fujifilm cameras because I'm torturing myself thinking about spending too much on a camera that I shouldn't buy. But I thought about using the shop.

Leo Laporte (66:14)

See, you said I don't need to buy a camera. I've got a. An iPhone.

Darren Kitchen (66:18)

But now I follow some people online who post beautiful photos. Their Fujifilm T100.

Leo Laporte (66:25)

That's a really nice camera.

Darren Kitchen (66:26)

It's so nice.

Leo Laporte (66:28)

We've been talking about that camera on our Chris Markworth's photo show. He's sort of a fan then.

Jason Howell (66:35)

You got to remember to bring that thing around with you.

Darren Kitchen (66:37)

But it's really tiny and portable. It's a pocket sized.

Darren Kitchen (66:42)

It's a problem. Anyway.

Leo Laporte (66:43)

It's a good camera. I would not say not to get.

Darren Kitchen (66:45)

It's a good camera. But it's $2,000 and backordered. This is. These are the problem. Anyway, I thought about using Shopping mode but I was worried that it would not be as in depth as Research Mode.

Jason Howell (67:00)

Yeah, good question. I haven't played around with it. Turns out though a lot of people were using AI for Black Friday just last Friday. Yeah.

Leo Laporte (67:09)

You know, Adobe of course puts these stats out every year and other analysts also.

Jason Howell (67:16)

And.

Leo Laporte (67:18)

How you interpret these stats is important. Let's just say that the president said, look, you know, Black Friday was huge. Americans spent a whole lot more money online than they had before. I can't remember where the growth was.5 or 6 or 7%, something like that. And I brought it up on Twitter on Sunday and fortunately I had a smart, a few smart people on there. I don't know if it's Mike Elgin or Daniel Rubino said, yeah, but notice why the dollar sales were up. It was fewer items, everything cost more.

Darren Kitchen (67:54)

Yeah.

Leo Laporte (67:55)

Oh.

Leo Laporte (67:59)

So we spent more but didn't get more.

Darren Kitchen (68:02)

Yeah, I looked at a couple products that I had bought literally last year because maybe 11 months ago they were more than double the cost.

Jason Howell (68:11)

No way.

Leo Laporte (68:12)

So that's why you should buy the Fujifilm now.

Jason Howell (68:15)

Yeah, it's a responsible purchase decision to just buy the camera, the really expensive camera.

Darren Kitchen (68:22)

Go on a road trip in like January or February for a couple of weeks and be a perfect road trip camera.

Leo Laporte (68:28)

You need a good camera. So AI, this is the story from Reuters helps drive a record $11.8 billion in Black Friday online spending. I gotta sign into Reuters Because AI.

Jason Howell (68:43)

Driven traffic to US retail sites soared 805% compared to last year, says Adobe.

Leo Laporte (68:51)

So that means what people were using AI browsers.

Jason Howell (68:55)

AI driven traffic to US retail sites. Yeah. What would that mean?

Darren Kitchen (69:00)

So that would be clicking on a link from ChatGPT and then you can see how it does the little question mark.

Jason Howell (69:08)

Still. 805. I mean if it was hardly happening last year, I was gonna say I.

Darren Kitchen (69:12)

Wonder how many link outs that is. Quant.

Jason Howell (69:16)

Yeah, totally. There's some missing information there because 805% is a pretty large, you know, large increase. But compared to what?

Darren Kitchen (69:24)

You know, I wonder also if they are. Brandroid brings up a good point which is Google search obviously now has.

Leo Laporte (69:32)

That's got AI.

Darren Kitchen (69:33)

It could just be people Googling.

Darren Kitchen (69:37)

I was gonna say best Fujifilm camera, but they're probably not Googling that. You know, Googling a normal thing that a normal person would. And then clicking on the AI summary.

Leo Laporte (69:46)

Best price for Fuji.

Darren Kitchen (69:49)

They're all really expensive is the issue. But then it's like if I'm going to get one of the less expensive ones, it's still like a thousand something dollars. And I'm like, why don't I.

Leo Laporte (69:58)

Here's a tip, don't get the less expensive one because you'll end up buying that and the more expensive.

Darren Kitchen (70:03)

Well, this is also coming from. How many Instagram story ad purchases have you made in the last week?

Leo Laporte (70:09)

No, I, I have deleted Instagram from all of my devices. I. It's too dangerous for me. And I also don't like what to me. So Anthony earlier was Talking about how TikTok shades the news to. You know. Well, Instagram has made some decisions about me that I do not fully agree with.

Darren Kitchen (70:28)

Are you talking about all the naked ladies they show?

Leo Laporte (70:29)

Yes.

Darren Kitchen (70:30)

You've been talking about this for years. You clearly are engaging.

Leo Laporte (70:34)

I'm not engaging. I only use Instagram to check up on you and my son and anybody else. Maybe Jason, occasionally people who use Instagram, I use it as a read only feed for people I know. That's all I want want. But I think because I don't give it a lot of signals because I hardly use it, it's decided. Well, you're an elderly white man. I think we should give you naked ladies. That's what I think.

Darren Kitchen (70:59)

And is it wrong?

Leo Laporte (71:02)

Yeah. And that's why I deleted it. Because I feel like I'm being taken advantage of. You are all these thirst trash.

Darren Kitchen (71:09)

Elder abuse.

Leo Laporte (71:10)

It's elder abuse. Thank you. Thank you.

Darren Kitchen (71:13)

You. I'm standing up for you.

Jason Howell (71:15)

They're all taken advantage.

Leo Laporte (71:16)

You're talking about the XT5?

Darren Kitchen (71:19)

No, I'm talking about the X100.

Leo Laporte (71:23)

I. Yeah, that's. That's the one everybody seems to like.

Darren Kitchen (71:26)

I mean, everyone's obsessed with it. It looks adorable. The film sims are fantastic.

Benito (71:34)

It's all about patience, Paris, because, like, in a couple of months, there's going to be a new camera.

Leo Laporte (71:39)

Yeah, that's true.

Jason Howell (71:40)

Yeah. But then you're not gonna want the this one, right? Because there'll be a new one.

Darren Kitchen (71:44)

No, I probably will. I don't know. I'm contemplating getting one off Facebook Marketplace.

Leo Laporte (71:50)

Let's take a break so that we can make some money so that Paris can afford her new camera. How about that?

Leo Laporte (71:57)

You're watching intelligent machines of very pleasant visit from our friend Jason Howell, host of AI Inside at AI Inside show with Jeff Jarvis. He does that show. Jeff is away for the week. He'll be back next week. I'm sure he'll be sad that he missed Anthony Vinci, though. We'll have to tell him all about it. That's Paris Martineau, cub reporter at Consumer Reports. She's not a cub reporter. You're no longer a cub reporter. You're a senior senior reporter.

Darren Kitchen (72:26)

I'm a Fat Bear Week reporter.

Jason Howell (72:30)

What's.

Leo Laporte (72:31)

I'm not sure I should ask. Does this have anything to do with RFK Jr. Fat bear leak?

Darren Kitchen (72:36)

Fat Bear Week. You were saying cub. So I was imagining.

Leo Laporte (72:39)

Oh, I get it.

Darren Kitchen (72:41)

What would be a grown up bear? It wouldn't just be fun to say adult bear. Why not Fat bear that Bear week.

Leo Laporte (72:47)

Got it. She's a full grown bear. Our show today, brought to you by the Agency Build the future of multi agent software with agency. Now that's spelled agnostic.

Leo Laporte (73:02)

It's an open source collective building the Internet of Agents and it's now part of the Linux Foundation. That's wonderful. Agency is building the Internet of Agents, a collaboration layer where AI agents can discover, connect and work across any framework. All the pieces engineers need to deploy multi agent systems now belong to everyone who builds on Agency, including robust identity and access management. That ensures every agent is authenticated and trusted before interacting. Agency also provides open standardized tools for agent discovery, seamless protocols for agent to agent communication, and modular components for scalable workflows. You'll be collaborating with the best developers from Cisco and Dell and Google Cloud, Oracle, Red Hat, 75 plus other supporting companies, all collaborating together to build the next gen AI infrastructure. Agency is dropping code specs and services, no strings attached. Visit agency.org to contribute. That again is agn tcy.org.

Leo Laporte (74:09)

We thank them for their support. I loved this piece. And you might not from no opinion, which is one of the substacks I really enjoy. I love AI. Why doesn't anyone?

Jason Howell (74:22)

Why doesn't everyone?

Leo Laporte (74:23)

Everyone? I mean.

Leo Laporte (74:26)

Maybe that was a Freudian slip. Anti AI sentiment might or might not be rational, but it certainly relies on a lot of bad arguments. And that's actually the point I wanted to bring up that. And he mentions Karen Howe. We had her on the show some weeks ago. She wrote a book.

Leo Laporte (74:47)

As I was reading it, I realized more and more it was kind of an anti AI, anti OpenAI.

Leo Laporte (74:54)

Book.

Darren Kitchen (74:54)

As I was reading it, men who was texting the group chat one page in, I don't think she likes AI.

Leo Laporte (75:01)

Well, it became more and more apparent as it went on. And as it sometimes happens, your bias against something might in fact inform.

Darren Kitchen (75:12)

The.

Leo Laporte (75:12)

Facts you choose or use.

Leo Laporte (75:16)

And you pointed out we were talking about this during the week, Paris. You pointed out that Karen Howe was quick to go to x.com and say oh, I made a mistake and correct it.

Darren Kitchen (75:26)

Okay, to be clear, specifically what he's talking about is there is one stat related to water usage that of, I think, data centers.

Leo Laporte (75:37)

Yeah, within 20 pages there is. Howe manages to claim that a data center is using 1,000 times as much water as a city of 88,000 people, when in actual fact it's using about 0.22 times as much water as a city and only 3% of the municipal water, the city system that the city relies on. In other words, she was off by a factor of 4,500.

Darren Kitchen (76:01)

Well, part of it is. I'd hesitate to say she was off. She was using. I'm forgetting the exact description of it. This is pointed out, I believe, by Andy Stone on Twitter. And so Howe brought up her sourcing, which is noted in the footnotes of the book. It was a governmental document from, I guess, the government of that city, which I don't believe was in the US and it had, it had cited the wrong stats, I believe, in the actual government report she was relying on. And so then she reached out to the government, asked them to correct their report or check the facts on their own report. They did. So now she's issuing a correction to that one stat in her book, which I think. I don't know. This is how writing a book, people can make fact checkers in this. And I think she did pay for fact checkers, but it's difficult.

Leo Laporte (77:00)

She also implies that AI data centers will consume 1.7 trillion gallons of drinkable water by 2027, while the study she's pulling from says only 3% of that will be drinkable water, 90% will not be consumed and instead return to the source unaffected.

Leo Laporte (77:18)

In other words, this, at least when you're talking about water. And the problem is, this is the problem I have with this. This information has become internalized, and people bring it up all the time. People. Somebody brought it up on Twitter a couple of weeks ago, and it's just what people say. Oh, yeah, well, just, you know, AI is fine except for the water use, or AI is fine except for the power use. Except it's BS and it's, It's. It's become part of the kind of public awareness of AI. Anyway, this is a good piece from no Opinion in which he talks about other reasons people attack AI, perhaps with less, you know, valid reasons than others. He also points out that we're a lot more concerned about AI in the US than they are anywhere else in the world.

Jason Howell (78:07)

That's interesting, isn't it? I wonder if that has to do with kind of the. Maybe it's a wrong word to use, but the democratizing quality that AI tools can bring with it. That, it. That maybe, you know, in the US we're so used to being at the top and being the leaders of this and the leaders of that, and maybe the AI that is proliferating worldwide is like an opportunity for everyone to kind of enjoy that. Whether that's true or not, I realize as I'm saying that. But you know what I'm saying.

Leo Laporte (78:41)

Yeah, I see what you're saying. I'm not sure why it is. This is from the Pew Center. How do people around the world feel about the rise of AI in daily life? 50% of the US is more concerned than excited. Only 10% is more excited than concerned. Italy is tied. We're tied with Italy than Australia. But it goes way down. I think where it goes way down, this would be. My guess, is in less developed countries, although I see there's a lot of bullishness in South Korea, Israel, Germany, Sweden, all highly developed nations. But in India, they're very much more excited about AI. The country that's most excited about AI is South Korea, interestingly. I don't know, maybe I get an impression that South Koreans are maybe more fond of technology.

Benito (79:31)

I think this has something to do with the marketing. In America, AI is marketed so heavily to everybody. I'm in Asia, and I was in Japan. Just the other just.

Leo Laporte (79:40)

Oh, yeah, okay. You were just in Japan, Benito. This is good. Yes.

Benito (79:43)

And this isn't, like, it. Like, AI isn't marketed to everybody there. It isn't plastered on every billboard. It isn't, like, on everything there. And I think that probably has a lot to do with it in America.

Leo Laporte (79:52)

Has a lot to do with it. I bet you're right.

Jason Howell (79:53)

Is that partially because it's just already accepted? And, you know, because, like, here, everybody here has to, like, point out, oh, well, that's AI. AI is driving that. AI, AI, AI. And there's, like, an overwhelm that's attached to that because everybody's like, God, I'm sick of hearing about the AI in other countries. Maybe there's an openness because of a lack of resistance. It's not always the story about the AI it's just the story about this technology.

Benito (80:19)

It's also an American industry pretty much. Right.

Leo Laporte (80:23)

Other countries very much part of our financial story right now. And not only the stock market, but our gdp. Almost all of our GDP.

Leo Laporte (80:33)

Growth is coming from AI companies. By the way, here we are, fat bears in the forest.

Darren Kitchen (80:39)

If you. If you go up one, there's one of us just as bears. And I want to know what I did to.

Leo Laporte (80:47)

You are a scary brown bear.

Darren Kitchen (80:49)

I'm a scary brown bear that's not wearing any clothes. But you guys are both wearing your outfits, which I think is very funny.

Leo Laporte (80:56)

That is odd. I don't well, it's because you have a black outfit.

Jason Howell (81:00)

You have a black outfit on.

Darren Kitchen (81:02)

Yeah, I guess that's true.

Leo Laporte (81:03)

You're a black bear and a black turtleneck. They can't tell you're dressed. That's all.

Darren Kitchen (81:07)

That's true. It's true.

Leo Laporte (81:08)

I do like your piercing bear eyes, though. Of all of us, you have the most piercing.

Darren Kitchen (81:12)

Yeah, it's RFK ass, in a sense.

Leo Laporte (81:17)

You had to say it. You had to get it in, didn't you?

Darren Kitchen (81:19)

You can't not.

Leo Laporte (81:20)

Thank you, Darren. OKE for what is obviously an AI. I have to point out that is AI generated. I'm just going to tell you, that's real. Oh, never mind.

Jason Howell (81:28)

Alternate. Alternate universe.

Leo Laporte (81:30)

That's amazing. Nowadays you can't tell, can you? I really do think because. And also because we're getting so much bombarded with this news. AI is in a bubble. The bubble or crash is going to happen. Everybody is saying this, the crash. And I keep saying, don't say that. You're going to make it happen.

Leo Laporte (81:47)

Whether it's going to be real or not. It's going to be dependent on whether people believe you when you say it's going to happen. But. So that might be.

Jason Howell (81:56)

Not say it.

Darren Kitchen (81:58)

Yeah.

Leo Laporte (81:58)

The only reason I don't want him to say it is because my entire retirement is invested in the US stock market. And I'm afraid. I'm very afraid because, you know, The S&P 500 is totally influenced by AI companies. This Magnificent Seven, they call them.

Leo Laporte (82:18)

Dozens of American universities and colleges have announced new AI departments, new AI programs. New York Times says college students flock to a new major, AI.

Leo Laporte (82:33)

The second most popular undergraduate major at MIT is now a program called. A new program called Artificial Intelligence and Decision Making.

Leo Laporte (82:44)

I think we're very interested in the usa.

Jason Howell (82:47)

I think so.

Leo Laporte (82:48)

Yeah.

Jason Howell (82:48)

I think that's pretty unsurprising, if I'm honest. Yeah.

Benito (82:52)

Although.

Jason Howell (82:52)

Although, when I look at what it's like to be a college student right now, it's got to be a really weird time to be going into college. Right. Right this moment where the uncertainty, whether founded or unfounded, you know, a lot of people are concerned about, like, well, you know, I'm gonna. What am I gonna study for four years or six years? Get my degree, and then.

Leo Laporte (83:14)

And then what?

Jason Howell (83:15)

And then what? The AI is going to do it, do it for me, if it doesn't already, like, why am I even going to school right now? It's got to be so weird to be in school. So of course they're flocking AI. Because I'm sure for a lot of them, it's like, this is how I keep myself relevant in this new world, whatever the heck it is or going to be. Yeah, it's got to be weird.

Leo Laporte (83:34)

We were talking about Instagram. Did you see the Austrian nuns? Do you know that story? The Austrian nuns?

Darren Kitchen (83:42)

No.

Leo Laporte (83:43)

So these are elderly nuns, just a handful of them, who were moved out of their convent and then came back and took it over.

Leo Laporte (83:55)

And illegally, as soon as rebel nuns.

Darren Kitchen (83:58)

Refused to give up Instagram. Love that. So that for them.

Leo Laporte (84:02)

Sisters Bernadette, Regina and Rita have been Busy. According to NPR, on their Instagram account, Rita, 82, can be seen rushing around the cloisters and dabbling in boxing lessons. Sister Regina, 86, has gotten so used to climbing four flights of stairs, she forgets to take the recently donated stairlift. And sister Bernadette, 88, regularly shares sharp witted observations about matters both sacred and secular over a ritual cup of coffee. These nuns made headlines this past fall after staging an escape.

Darren Kitchen (84:39)

Pardon me, how many followers do they have? I wonder?

Leo Laporte (84:42)

185,000. They made an escape from the care home. They were in a nursing home, which they say the church took them to against their will. They escaped from the nursing home, moved back into the shuttered convent, and then started an Instagram account. The authorities, the local abbey, and the Archdiocese of Salzburg had acquired the convent.

Leo Laporte (85:07)

So the archdiocese has said, well, you can go back, you can stay there. Rather, we'll let you stay there, but you have to give up this Instagram account because it's really given us.

Leo Laporte (85:18)

Some upset stomachs here. Speaking via Instagram, Sister Regina said, we can't agree to this deal. Without the media, we'd have been silenced via Instagram. Here she is. It's Italian. So you know.

Leo Laporte (85:34)

No, it's German. What am I saying? Italian.

Leo Laporte (85:39)

This. This is. This is what? Instagram? Yeah, this is what Instagram was made for.

Darren Kitchen (85:45)

It is true.

Leo Laporte (85:46)

God bless them and I hope they. They leave them in there.

Jason Howell (85:51)

It's like Kardashians, but with nuns.

Darren Kitchen (85:55)

Put them on. Put them on Love Island. I'd watch that.

Benito (85:59)

It feels like an 80s movie and they have to win a talent show to go.

Leo Laporte (86:02)

It does, doesn't it?

Darren Kitchen (86:04)

Yeah.

Leo Laporte (86:04)

Where's Whoopi Goldberg when we need her?

Leo Laporte (86:08)

Their superior said, they can stay, but the nuns must cease all social media activities, stop talking to the press and forego seeking legal advice. The nuns have. Yeah, the nuns have. Well, they have given a vow of obedience, I might add. The nuns have rejected the proposal. And now their superior has called on the Vatican to intercede. I was wondering why Father Robert was.

Jason Howell (86:34)

Going to say, can we get Father Robert on?

Leo Laporte (86:36)

I think I now know what he's up to.

Darren Kitchen (86:39)

He's trying to convince the nuns.

Leo Laporte (86:40)

Canon law scholar and priest Voorhees, chief gang rota, told NPR the deal is neither reasonable nor humane and has no legal basis in either church or state laws.

Leo Laporte (86:52)

They have a right to their. Not only to their. They have a right to post, but yeah, they have a right to their social network.

Leo Laporte (87:02)

On Monday, their superior called in a crisis PR manager.

Leo Laporte (87:10)

Wow. So it's. We'll keep. We'll follow this. We'll follow this.

Darren Kitchen (87:14)

How long until the church posts an Instagram notes app apology? Great.

Leo Laporte (87:22)

Apparently her superior, the provost, is just as media savvy as the nuns. Sister Bernadette mentioned, For instance, his 2022 photo shoot with an Austrian TV chef.

Darren Kitchen (87:37)

Oh, boy.

Leo Laporte (87:38)

The provost in the church invite journalists to the big parties they throw. It helps raise money. Why shouldn't we do the same?

Leo Laporte (87:46)

Anyway, we'll keep you up to date on the Sisters of perpetual indulgence.

Jason Howell (87:52)

247,000 followers, actually.

Leo Laporte (87:55)

Oh, they just went up.

Jason Howell (87:56)

Yeah, they're surging right now.

Leo Laporte (87:59)

This story is bringing them new followers all the time.

Jason Howell (88:02)

Wouldn't you. I feel like I gotta click follow. You know, I gotta. I can't understand a word they say.

Leo Laporte (88:08)

But, you know, let us. Let us take a little break and we will have more in just a moment. You're watching Intelligent Machines. It's great to have Jason Howell here. Thank you, good friend from AI Inside at AI Inside show. And of course, Paris Martineau, the angry brown bear.

Leo Laporte (88:29)

You don't look like a brown bear.

Jason Howell (88:30)

All.

Leo Laporte (88:31)

No, not at all. We will get the Olivia Nuzzy update in just a moment.

Darren Kitchen (88:37)

I don't know, guys, I. I'm try. I probably should do it because I'm getting dinner with a friend tomorrow who I haven't updated on this in like two weeks and it's going to take me 45 minutes to tell him what has happened.

Leo Laporte (88:49)

So we'll just.

Darren Kitchen (88:50)

We'll just read the clips.

Leo Laporte (88:51)

A couple of days ago, Paris Posts somebody on. On their blog posted a series of just.

Leo Laporte (89:01)

Appalling sentences from her new book American Canto. And Paris says, I'm standing outside in the cold reading these and I can't stop.

Darren Kitchen (89:10)

I did think it was gonna be the first example of death by prose because someone sent me this blog by Scotchy Sentences from American Kanto that sent me to the Hospital is the title of it. As she, as Scott, she had reviewed this for Slate and it's just the Times review.

Leo Laporte (89:30)

All right, well, we'll talk about it.

Darren Kitchen (89:31)

Once after the break. We'll talk about it after the break.

Leo Laporte (89:33)

Did you see the Times review yet? I want to give you time to read it.

Jason Howell (89:35)

If you wait.

Darren Kitchen (89:35)

Should we go through them?

Leo Laporte (89:36)

Go look at the New York Times review. I think you'll enjoy it.

Darren Kitchen (89:39)

Did you see the Washington Post review?

Leo Laporte (89:42)

Let's put it this way, they're not very good.

Leo Laporte (89:47)

I would almost say scathing. This episode of Intelligent Machines is brought to you and this is this, by the way, this whole story is totally inside media has nothing to do with AI. So it's just a footnote.

Darren Kitchen (89:59)

Hey, you know, last two letters of Olivia Nuzzy's first name.

Leo Laporte (90:03)

I A I A.

Anthony Vinci (90:05)

Turn that around.

Jason Howell (90:07)

Look in a mirror.

Leo Laporte (90:08)

Olivia Nuzzy, her erstwhile fiance Ryan Lizza, and her portrait artist, Izzy. Yeah, it's really, it's a little incestuous. They are the new Kardashians, I think. Our show today, brought to you by Vention. Now, look, you are, as we all are, hearing all this hype about AI, maybe you've even created, you know, an AI proof of concept and, and you're really thinking, how can we get our company into this brave new world? But it's scary, it's puzzling, and the worst thing is it can actually tie you up in knots. That's why you need to know about Benchen. AI is supposed to make things easier, but I have to say, for most teams, it's just making the job harder. But fortunately, there's a partner here who can help you cut through the hype. Vention and their 20 plus years of global engineering expertise. They build AI enabled engineering teams that make software development better, faster, cleaner, calmer. Vention's clients typically see at least a 15% boost in efficiency. And we're not talking through hype, but through real engineering discipline. Because at bottom, that's what Vention is. Engineers, developers, people who understand how this stuff works. They will also do if you want, and I would suggest you check this out at least. In fact, maybe the first thing you do. They have these great fun AI workshops and it's an interactive workshop with you, your team and the Vention pros. They will work with you to help your team find practical, safe ways to use AI across your entire business. It's a great way to start with Vention to test their expertise, whether you're a cto, a tech lead, a Product owner. The beauty of this is you won't have to spend weeks, kind of months maybe figuring out tools and architectures, which model you use. You know, Vention can help you assess your AI readiness. They will work with you to clarify your goals. They'll help you develop an outline to the steps that you need to take to get you there without the headaches. And of course, that's just the workshop. If you need help on the engineering front, their teams are there. They're ready to jump in. As your development partner or your consulting partner, it is probably the best thing you can do after your proof of concept. This happens all the time. You've built this incredible prototype, right? You maybe vibe coded it unlovable. It's even running in tests. You got it, you can look at it. But now what do you do? How do you productize it, how do you market it, how do you implement it? Do you open a dozen AI specific roles just to keep moving forward? Or maybe you bring in a partner who's done this before across many industries. Someone who can expand your idea into a full scale product without making you crazy, without disrupting your systems, without slowing your team down. That's Vention. Vention is real people with real expertise that are going to give you real results. You can find out more@ventionteams.com See how your team can build smarter, faster and with a lot more peace of mind. Or get started with your AI workshop today at ventionteams.com TWiT that's V E N T I O-N teams, all1word.com TWiT thank you, Vention, for all you do to help our lovely audience get their head wrapped around what AI is about to do to them. Invention. We did. This is the time of year you might remember this. Actually, Jason, we used to do this where you every dictionary comes up with its word of the year, right? We've already done Merriam Webster. We've done a few of them, haven't we? Paris? I think so.

Darren Kitchen (94:03)

We have.

Leo Laporte (94:04)

But the king of the hill, the one that really matters, is the Oxford Word of the Year, right? I don't know. Oxford University. I feel like it's the prestige one, right? Oxford University Press. They announced Monday after three days of voting by more than 30,000 participants, the Oxford Word of the Year for 2025 is.

Leo Laporte (94:32)

Rage bait.

Darren Kitchen (94:35)

Okay?

Jason Howell (94:36)

It beat out Aura Farming and Biohack. And Biohack.

Darren Kitchen (94:41)

Aura Farming was already in someone else's word of the year because I had to.

Leo Laporte (94:44)

Maybe that's why.

Benito (94:45)

Yeah, and all of these are keywords, by the way.

Jason Howell (94:48)

Yeah, good to know. I mean, biohack, they took the space, they made it a one word thing. But yeah, not for ragebait or rfr.

Leo Laporte (94:56)

Rage bait. According to npr, which apparently these days has his finger on the pulse of Instagram and TikTok. Rage bait gained popularity after actress Jennifer Lawrence revealed that she has a secret TikTok account she uses to get in fights with strangers online.

Darren Kitchen (95:15)

I really respect that. You gotta, gotta be a hater.

Leo Laporte (95:19)

She picks fights. This is.

Darren Kitchen (95:22)

I mean, it's not me. I'm. I don't need to pick fights with anybody on the Internet. But I do think it's kind of beautiful to be a celebrity who can't go anywhere or do anything without being mobbed and recorded. To have a secret fight burner account for Instagram.

Leo Laporte (95:40)

She says, I get in fights on TikTok. I get in fights in the comments section.

Benito (95:48)

Athletes do this too, by the way.

Leo Laporte (95:50)

Do they? This is the new thing.

Darren Kitchen (95:52)

And how we usually find out is they slip up and accidentally use their personal account, suddenly see them in the comments posing to be someone else.

Leo Laporte (96:00)

LeBron. No one girl was like. Jennifer adds, this is in a publicity tour conversation with her co star in their new movie, Die My Love. Robert Pattinson. Is he the. He's the Twilight guy. He's the guy from Twilight.

Darren Kitchen (96:16)

Twilight, yeah. But he's so much more than that.

Benito (96:19)

He's a good actor. He is a good actor.

Darren Kitchen (96:21)

He is.

Leo Laporte (96:22)

He's got chiseled cheekbones, which makes me just trust him immediately.

Darren Kitchen (96:28)

He and Kristen Stewart are both really good actors.

Leo Laporte (96:32)

Yes, they are. Both from Hunger Games. No Twilight.

Darren Kitchen (96:36)

No Twilight. It's okay, grandpa. You're gonna be going to bed soon, so it'll come.

Leo Laporte (96:40)

I wish I were in bed right now.

Jason Howell (96:44)

Yeah, you want it? You want to go to sleep, watch Twilight. There you go.

Leo Laporte (96:46)

Yeah, no kidding. I tried. The only way I get through it was by making snarky comments the entire time.

Darren Kitchen (96:52)

Well, I would recommend there was a time in the last month or two where it was, I think, available for free on streaming. And friends of mine used to do a drinking game where you had have to take a drink every time someone sighs in that movie. And it's devastating. It's a devastating. It's a devastating prospect.

Leo Laporte (97:13)

How's nickvember? How was nickvember? You've completed your.

Darren Kitchen (97:16)

Nickvember was fantastic.

Leo Laporte (97:18)

How many Nicolas? So maybe you don't know this Jason. How many Nicolas Cage movies did you watch in Nick again?

Darren Kitchen (97:24)

I kind of. I kind of beefed it, boofed it. So I do this thing now for the past two years where in November I attempt to watch 30 Nicholas Cage movies in the month of November. It's called nickmember. And I try not to watch any other movies featuring if they don't feature Nick Cage.

Leo Laporte (97:45)

Oh, I didn't know this little deal. You don't watch anything else.

Darren Kitchen (97:49)

Well, I ended up watching, I think two other movies only because I was hanging out with a friend who had already subjected to, I think, three or four Nick Cage movies at that point. I was like, I can't. I can't in good conscience earnest fight for us to see a fourth Nick Cage movie together right now. But it was great. Honestly, I was really. This was my first nickvember where a Nick Cage movie came out in theaters during the month of November.

Jason Howell (98:14)

That's exciting.

Darren Kitchen (98:15)

The Carpenter's Son, a movie where a horror movie where Nick Cage plays Joseph to Teen Jesus, or he might be Teen Devil. We don't know.

Darren Kitchen (98:26)

It. I don't know. It was really good. I. I did end it this year by doing something that I planned on doing last year, which was watching the Unbearable Weight of Massive Talent. And honestly, Terry.

Leo Laporte (98:40)

I didn't.

Darren Kitchen (98:41)

I think it was.

Leo Laporte (98:43)

It makes fun of Nicolas Cage. Maybe that's.

Darren Kitchen (98:45)

It makes fun of Nicolas Cage and also just fundamentally misunderstands. What about Nick? Like it. It. So the plot of it is that Nick Cage plays Nicholas Cage, plays Nick Cage, who is Nick Cage, but he's down on his luck, running out of money, ends up accepting like a million dollar offer to go hang out on this island with a fan, and then like kind of an action movie spirals from there. But it's all a bit tongue in cheek, poking fun at him. Him. Basically, they name check a bunch of his movies and it's kind of an homage to his career. But I found it frustrating because despite the fact that it's ostensibly an homage to his career, it just doesn't understand things that make his movies good, which I think a lot of people have this warped understanding of Nick Cage that dates back to a, I think like 20 tens YouTube video that went super viral called Nick Cage Freakout Compilations, which is given everybody this understanding that Nick Cage is just a guy who screams a lot. And that's his whole shtick. And he does scream in quite a lot of movies, but that's not the shtick. Like, it's different than that. It's not just that he's shouting so.

Leo Laporte (99:53)

Much more than that.

Darren Kitchen (99:54)

It is a more Robust character.

Leo Laporte (99:56)

It's the blue in his eyes.

Darren Kitchen (99:58)

It's him eating a live cockroach after, you know, and that's. It's him eating a live cockroach like 35 minutes in to Vampire's Kiss. And then him and the director are being like, well, how do we escalate from here? It is. It's so much more.

Jason Howell (100:12)

Man, you are a connoisseur.

Darren Kitchen (100:14)

I am a cage.

Jason Howell (100:16)

A sewer.

Leo Laporte (100:17)

Is he Renfro in the. What's his name in the Vampire's Kiss? Is he.

Darren Kitchen (100:22)

Oh, he's in Ren Faire. I didn't think Ren Faire was a good movie, but I did like Nic Cage in it. Yeah, I think that he's the best part in that.

Leo Laporte (100:28)

He's the best part in any movie that he's in.

Jason Howell (100:30)

Okay, all right, maybe you've already answered this then. But if you're such a cage of sore. Favorite cage. Least favorite cage.

Leo Laporte (100:38)

Oh, don't ask her. That's impossible.

Darren Kitchen (100:41)

It's hard.

Jason Howell (100:41)

So I'm just authentically curious. I just plugged it in. Gemini, by the way. It says 120 to 130 films somewhere in that.

Darren Kitchen (100:51)

Yes, Ballpark. Because as a brief digression, he went through a period that I call the pyramid debt period, where he bought, made a lot of unwise purchases like multiple pyramids and exotic animals, and then went into debt and had to make a lot more money movies to get up to it. But to answer your question, my instinctive answer for favorite cage is Face off. Because it's a perfect movie. And what inspired me to do this.

Leo Laporte (101:12)

That is a great.

Darren Kitchen (101:13)

But I think technically speaking, I'm either adaptation or Vampire's Kiss is probably talking about movies that make my heart sing.

Leo Laporte (101:25)

Come on.

Darren Kitchen (101:26)

I mean, Moonstruck is also fantastic is the issue. There are just so many, many great Nick Cage films out there.

Jason Howell (101:34)

Like the whole set. I get it.

Darren Kitchen (101:35)

Yes, it's wonderful. Matchstick Men. Honestly. Also a banger. Even though it's more of kind of a straight to like, it's sort of movie you'd see on cable tv. But least favorite movie is Zandile by far. Which is a. I gotta look this one up. It's. I don't even. I don't even know how to describe this.

Leo Laporte (101:52)

Would you say that you're a cageouseur?

Darren Kitchen (101:55)

Yeah, I am a cajouseur. It is. Ostensibly. It's. I'll read the description. Bored with her marriage to burnt out poet turned corporate executive, the Harry Zandile falls prey to an old friend of her husband, Nick Cage, who's got a crazy goatee, the manipulative and egotistical Johnny and becomes enmeshed in a sensual passion, passionate and destructive affair. It could I guess, be Challenger esque if you were doing it right. But every aspect of the movie is just done wrong. It's shot terribly. The dialogue doesn't make sense, the plot doesn't make sense. They're. The actors aren't really doing a good job in any of it. It's just. It's the closest I've come to turning off a Nick Cage movie.

Jason Howell (102:41)

O man, I've only known about this whole aspect of your life for like five minutes, but I never would have imagined there would ever be a Nick Cage film that you'd think about trying to.

Leo Laporte (102:53)

It was so nagger on, so bad.

Darren Kitchen (102:57)

I can't emphasize.

Jason Howell (102:59)

Does Judge Reinhold doesn't save the film. I mean seriously, you would expect.

Darren Kitchen (103:04)

Listen, you'd expect the craziness of his. As someone who literally has a Nick Cage coaster right next to me right now, you'd expect that I would be enamored by some part of it. But it's bad. It's like not even a bad movie in the way that I'd recommend watching. I think my letterboxd review for it was this could have been Challengers, Instead it was boring. That's the issue is I like a movie. I love a bad movie if it's bad in an interesting way. But to be bad in a boring way, criminal.

Jason Howell (103:35)

Yeah, totally.

Darren Kitchen (103:37)

I forget how we got on this.

Leo Laporte (103:38)

I don't know, but I'm gonna take a break before we do rage bait. I have. Yeah, that was rage bait. That's how we got it.

Darren Kitchen (103:44)

Huge rage bait.

Leo Laporte (103:45)

I have a two parter for you. This is Paris's Thanksgiving, as you know every year.

Darren Kitchen (103:56)

I'd also like to add to my thing I forgot Wild at heart. Also probably Wild at Heart.

Leo Laporte (104:01)

That's the classic.

Jason Howell (104:02)

Yeah, it's one of my favorites.

Leo Laporte (104:04)

That's the classic.

Darren Kitchen (104:04)

Sorry, we do have to. I have to add a caveat. So last week I was home in Florida. I literally as I was. Because we talked about, oh, maybe my dad should pop in and be on the show. I as I was and I was picking me up from the airport, I was like, you know when I'm filming on Wednesday, like you should come by. I'm sure the boys would like love to chat with you. Haha. And he like looked at me, he's like, I don't know, maybe and seemed.

Leo Laporte (104:26)

Like really uninteresting he was playing it cool, wasn't he? He was just.

Darren Kitchen (104:29)

He was trying to play it so cool. And apparently he spent the entire night that I was recording just kind of waiting around hoping that I would invite him in. And then afterwards was like, I guess you didn't want me to come on after all. I was like, you seemed uninterested.

Leo Laporte (104:45)

Broke his heart.

Darren Kitchen (104:47)

I did. He was genuinely.

Leo Laporte (104:49)

There's a lot of backstory, Jason, on this. There was earlier, a year or two ago, they were at a wedding and where was it, Bruce?

Darren Kitchen (104:57)

They were at a wedding in Peru. And for some reason, because like the wedding had.

Leo Laporte (105:02)

Because he had bodyguards and was wearing a Tuxiri tuxedo, people in Peru thought.

Darren Kitchen (105:07)

He was some American celebrity. And there's this video posted by a random TikTok account of people mobbing my dad and my dad's a ham. So of course he responds with me like, I'll sign your autograph, I'll take.

Jason Howell (105:18)

A photo with you. Sure.

Darren Kitchen (105:20)

And my mom is in the background being like, scott, come on, we gotta go. But he's like, he.

Leo Laporte (105:25)

I think he. Next time you gotta get him on the show. Because he really.

Darren Kitchen (105:28)

He had questions for you guys prepared. It was really.

Leo Laporte (105:31)

I think he really. I feel terrible. You know, maybe we should call him up anyway. This might ease the pain a little bit. It's a two parter. We'll start with part one before the turkey goes in. Let me know.

Jason Howell (105:46)

All right, here we go.

Leo Laporte (105:47)

The insertion of the deep frying of turkey is always kind of an important component. I hope we do have the destined fire department ready to go at any moment. They'll be just around the corner.

Jason Howell (105:57)

That is right here.

Leo Laporte (106:00)

That's your grandma, right?

Darren Kitchen (106:01)

That is my grandma. Yeah. She hangs out by the deep fried turkey so that she can smoke cigarettes all night.

Darren Kitchen (106:08)

Going to be a little profound.

Leo Laporte (106:10)

So, you know, here we go. Get ready. It's going in the. Don't try this at home, kids. Oh, God.

Darren Kitchen (106:16)

We do it every year and every year I do think that it's going to explode.

Leo Laporte (106:20)

It seems so scary and dangerous.

Darren Kitchen (106:21)

You pointed out after I said you this that she has a fleece blanket like one foot away from.

Leo Laporte (106:26)

She's sitting so close. That is fair.

Darren Kitchen (106:28)

She would.

Leo Laporte (106:29)

She would go up. Oh, well, she didn't.

Darren Kitchen (106:32)

She didn't.

Leo Laporte (106:34)

Now the drinking can begin. We got about three.

Leo Laporte (106:38)

Oh, wow, they're ready. Go again.

Darren Kitchen (106:41)

Yep, the drinking can be good.

Leo Laporte (106:42)

Minutes per pound.

Darren Kitchen (106:43)

17 didn't understand the pan.

Leo Laporte (106:45)

We'll be going at this about 51 minutes. We'll give you updates along the way. He's very good at math, your dad.

Jason Howell (106:52)

No kidding. Well, how was the turkey?

Leo Laporte (106:55)

Oh, you're gonna see how the turkey was in a moment when we come back.

Darren Kitchen (106:58)

Crispier than ever and juicier than ever. He also had watched apparently a lot of TikTok videos about different turkey carving techniques.

Leo Laporte (107:07)

Oh, carved in a.

Darren Kitchen (107:09)

Very interesting.

Leo Laporte (107:09)

So that is an issue. That is an issue. Now the quality of the, of the, the second part is lower. So I don't know, maybe you have it a better quality version. It looked better on my phone. I don't know, maybe if Apple has downgraded it or something. Do you have a great Android?

Darren Kitchen (107:26)

Says of my grandmother, she's smoking with a fleece blanket right next to the vat of boiling oil heated by a propane tank. What could possibly go wrong? This is a person. Great question.

Leo Laporte (107:36)

Who has no fear. It's true. She is a brave, brave.

Benito (107:41)

She is part of the greatest generation.

Darren Kitchen (107:44)

Yeah, it's true.

Leo Laporte (107:45)

No, it's funny because I saw the video. I said, so who is that woman sitting there? Is that your grandma? And she said, yeah.

Leo Laporte (107:53)

Yeah, that's grandma. I love it. We're gonna have more in just a bit. Jason Howell is visiting. It's great to have you. Jeff will be back next week. And of course, Jeff joins Jason every week. In fact, he didn't miss the show this week, I might add, on a inside. What is he doing that show? I don't know. What is he. Do you know what he's doing, Jason?

Jason Howell (108:12)

He is speaking at a media conference, I believe, somewhere in a different country. That's about all that my memory can recall at this point. But yeah.

Leo Laporte (108:24)

We will see the turkey emerge from its hot bath in just a moment. But first, a word from Monarch. Between travel, gifts, hosting, entertaining, you name it, it's easy to lose sight of your money and sad to say, maybe even of financial responsibility this time of year. If you want to keep your finances under control this holiday season, you need to be using Monarch, rated Wall Street Journal's best budgeting app of 2025. Monarch is the all in one personal finance tool that brings your entire financial life together in one clean interface on your laptop or on your phone. And right now, just for our listeners, monarch is offering 50% off. 50% off your first year. Now I can tell you I use Monarch. I live in Monarch. It is so useful for me, especially now I'm on a fixed income. I have to pay attention my investments or am I going to survive? Am I Being Alpo next week. Monarch lets me keep track of it. I never in my whole life budgeted before. I always found it too fiddly. Too much data entry. I don't have to even do that anymore. With Monarch, it automatically is hooked up to all my accounts. By the way, that was an easy process. Stays hooked up, which is great. And then automatically imports all the transactions and automatically filters them, files them away in the right place. So I know exactly how much I spent on food, how much I spent on groceries, gifts. Not much charity, even less. But I know it's in my head and I know where I am. Monarch is built for people who don't have the time to spend doing a lot of data entry. Putting in your checkbook. Remember we used to do that in the early days of computers. You'd get your bank statement, you type in all your. You don't have to do that anymore. If you put off organizing your finances. Monarch is for you. Or if you, like me, in the old days, used to do it by hand and got tired of it. Monarch is for you. Monarch does all the hard part, the heavy lifting. You link all your accounts. It takes just minutes. You're going to. And by the way, the data visuals, the graphs and everything, they're beautiful. You get smart categorization of your spending. You get real control over your money. You know, I told. I was talking to Henry about this. My son, when you're making good money, you know you're in your prime of your earning. It's very easy to say, yeah, I got plenty of money. I think about it. Don't do that. Don't do that. I told him. Not I said, I did this. This is not good. You're going to have kids to put through school. You're going to want to buy a house. You're going to want to retire someday. When you're young, you don't think about that. Now's the time. Now's the time. Don't be complacent. If you're not on top of your money, you could be missing chances to save more, to invest smarter, to hit your financial goals faster. Like buying a home, saving for retirement. Whatever it is. I just thought of something. I know what I'm going to give Henry for Christmas. Monarch helps you stay informed so nothing slips through the cracks. Monarch, it's not just another finance app. It's a tool. And real professionals and experts use it. It's Forbes best app for couples because you can share your finances with one another. That's one of the Things couples get in fights over more than anything else. If it's on the paper there, you know what's going on. It's much, much easier. CNBC named it one of the top fintech companies in the world. It's got a very active and passionate Reddit community. Paris, you might be interested in that of over 34,000 users. What's really cool, Monarch pays attention to that community. Those users are shaping how the product is developed. Monarch cares. Don't let financial opportunities slip through the cracks. Use the code iamanarch.com in your browser for half off your first year, 50% off your first year. Monarch.com the code is. I am so glad. I'm trying to think what to get Henry, he called me last week. He said, I don't know, should I, I'm in the stock. Should I get. What should I do? How should I, should I save? And I, I tried to give my best advice. I'm going to give Monarch 50% off your first year. Monarch.com use the code. I am knowledgeispower. Monarch.com thank them so much for their support. Yeah, there's a website. I use the website, but there's Also iPad app, iOS app, there's an Android app. So, you know, you got it covered everywhere. Oh, I don't know. Did you put it in the discord? Because this is kind of low quality. I wanted to show everybody how good that turkey looks.

Darren Kitchen (112:53)

Yeah, hold on one second.

Leo Laporte (112:54)

I'll put in the discord because I'd.

Jason Howell (112:57)

Love to get more hungry than I am already.

Leo Laporte (113:00)

Could you. Is it possible, hungry for turkey this short, this recently? That Thanksgiving was this recent.

Darren Kitchen (113:05)

I mean, what did you do for Thanksgiving?

Leo Laporte (113:07)

I made my own turkey. I used Alton Brown slash Kenji Lopez Alt's method, which involves a baking steel on the bottom of the oven. So that. And then you turn up the oven all the way to 500 degrees, get it really, really hot, put the turkey and then turn it down to 300. But the heat from the steel cooks the dark meat faster. That's the problem. Right? The light meat cooks. The dark meat takes more time. So. So the heat cooks the dark meat faster. It actually came out perfect. It was. I was stunned. The dark meat was fully cooked, the white, the breast was not dry. It was really, really good. So. But I think Paris will vouch for me, vouch for this. But I think deep frying produces a pretty good result.

Darren Kitchen (113:49)

It is delicious. The skin is crispy and brown, but it is juicy and flavorful in the inside.

Leo Laporte (113:56)

And it's in 50 minutes. And that was a big turkey. That was a giant. How many people?

Darren Kitchen (114:02)

17 point something.

Leo Laporte (114:04)

That's huge. We had a 10 pound turkey. And I still have leftovers.

Jason Howell (114:08)

All you risk is a bonfire. I thought you were gonna say salmonella.

Darren Kitchen (114:13)

That's why you do it in your.

Darren Kitchen (114:16)

Backyard.

Leo Laporte (114:17)

Yeah, by grandma. Out by grandma. The fire department.

Darren Kitchen (114:20)

Oh, right next to grandma.

Leo Laporte (114:21)

That's why.

Leo Laporte (114:23)

Alcohol. That's there as well.

Darren Kitchen (114:25)

Yeah. There goes so much alcohol. All right. I posted it in the chat.

Leo Laporte (114:29)

Okay.

Darren Kitchen (114:29)

Yeah, I was going to say it is. It was close. It was close to the alcohol and frankly, it wasn't far enough away from the garage.

Leo Laporte (114:35)

Yeah.

Jason Howell (114:36)

As I'm looking at it, I'm like.

Leo Laporte (114:38)

All right, Here we go. 7:59 minutes later. It's now time for one hour.

Jason Howell (114:44)

How much bourbon has been drank in that time?

Darren Kitchen (114:47)

Infinite.

Jason Howell (114:47)

Coming out.

Leo Laporte (114:48)

Look at that.

Darren Kitchen (114:49)

Oh my God.

Leo Laporte (114:50)

Look at that temperature gauge. Yes, please don't touch it.

Anthony Vinci (114:54)

What is he crazy?

Darren Kitchen (114:56)

He's crazy. Yes.

Jason Howell (114:58)

What? What did he just do?

Jason Howell (115:03)

I guess I should have asked. How did he do that? That's.

Leo Laporte (115:05)

Yeah, no kidding.

Darren Kitchen (115:06)

No.

Leo Laporte (115:06)

This is now headed to the dining room table. First we'll put it. This is. What a trooper your dad is. He doesn't. Doesn't let anyone anything.

Darren Kitchen (115:14)

Oh, look at that crispiness.

Jason Howell (115:16)

That does look delicious. I want to eat that right now.

Leo Laporte (115:18)

Should be ready to go. We're going to let her rest for 20 minutes. Then it's turkey carving Now. How do you get all that metal out of there? Does it just come right out?

Darren Kitchen (115:26)

It's kind of just like a. A little frame that goes in there. You kind of pull it from the other side.

Leo Laporte (115:32)

There's your dad.

Jason Howell (115:33)

Destroy the turkey.

Leo Laporte (115:34)

There's your dad.

Darren Kitchen (115:37)

Wow, it's good.

Leo Laporte (115:39)

By the way, next time have dad have a k extinguisher. HD editor has some experience with this. He says an ABC fire extinguisher is not good enough. You need the foam.

Darren Kitchen (115:49)

I don't even know if they own a fire extinguisher, to be totally honest with you. I should get that as well. I think this is actually wise because I started making a list of things I need to buy to bring to them in when I come home for. Got a little nervous the holidays. My parents, their whisks are broken. They've got a messed up cheese grater. They've never properly cleaned their cast iron.

Jason Howell (116:14)

Oh, boy.

Leo Laporte (116:15)

What kind this is.

Jason Howell (116:16)

It's properly seasoned.

Leo Laporte (116:17)

Then I am ready in case.

Jason Howell (116:19)

Dang. Look at that.

Leo Laporte (116:20)

Should break out. Well, I got nervous at one point. I think There's a lot of electronics up here in the attic.

Darren Kitchen (116:24)

Will that work for electronic fires?

Leo Laporte (116:26)

This is an abc. This is an abc.

Jason Howell (116:29)

What is the difference?

Leo Laporte (116:29)

I get a K. I need a K, don't I?

Darren Kitchen (116:32)

No, I. I don't know.

Leo Laporte (116:33)

He's the one with the foam.

Darren Kitchen (116:35)

Yeah, but do. I don't think fire extinguishers work for lithium battery fires. I know this because I thought about getting a $300 box for my E Bike batteries.

Leo Laporte (116:44)

Yeah, they don't. You're right, because I have all batteries make their own. I also have an emergency fire blanket. I do have that. The prepared Hero emergent. I saw it on TikTok.

Jason Howell (116:55)

I have a massive amount of old smartphones up there. Many of them actually came from the Twitch studio when. When you were shutting down the studio. Yeah.

Leo Laporte (117:03)

You got them. Thank God. I don't want them.

Jason Howell (117:04)

These are the, you know, ancient batteries that are just daunting me right now.

Darren Kitchen (117:08)

I'm constantly trying to pawn my old smartphones off on friends because I'm like, they're gonna explode one day.

Jason Howell (117:13)

Can it be your problem and not mine? Thank you.

Leo Laporte (117:16)

So we're very excited. You may have seen us talk about Marumi, the furry companion robot. It is now available on Kickstarter. We are very, very excited.

Darren Kitchen (117:33)

Oh, my God. And this is different than the one that's just a little flu.

Leo Laporte (117:37)

This is better than the one that sits in a little bowl and just goes, yeah, yeah. This is a. What they call a charm robot. See, it's attached to her purse. And apparently all this is. Look around.

Jason Howell (117:46)

This is a spider vice.

Darren Kitchen (117:48)

Oh, yeah.

Leo Laporte (117:49)

Yeah.

Jason Howell (117:49)

I don't know this.

Leo Laporte (117:51)

I think you're right. I think Anthony Vinci would say, do not. Where is this made?

Jason Howell (117:56)

I wonder what he would have said about Labubus, because I've heard some really crazy, weird things they don't have.

Leo Laporte (118:03)

Do they have.

Jason Howell (118:05)

I think there's just some. Some. Some wacky conspiracy.

Benito (118:08)

There was some viral stuff. People cutting laboos open and finding cameras, but I don't know if, like, there's that are planted or, like, you know, you can't believe anything on.

Jason Howell (118:17)

Totally.

Leo Laporte (118:17)

You're not.

Jason Howell (118:18)

I know, and I don't.

Darren Kitchen (118:19)

But still, VO should get on that. We could make a good view.

Leo Laporte (118:24)

I am glad that your father did not fall for the AI Slop recipes that are taking over Thanksgiving dinner food bloggers. And my son is a food blogger, sees traffic dip as home cooks turn to AI inspired by impossible pictures.

Jason Howell (118:42)

Oh, okay.

Darren Kitchen (118:44)

Oh, boy.

Jason Howell (118:45)

The recipes are not AI generated. The pictures are right and there's, well.

Leo Laporte (118:49)

No, I think it's all AI now.

Jason Howell (118:52)

I wondered about that. I mean I, I pretty much assume that if an AI is generating a recipe that didn't exist before, it's, it's probably your chances are pretty low that it's going to be amazing.

Leo Laporte (119:01)

So Bloomberg quotes E.B. gargano, who has been writing recipes online for some time. The easy peasy foodie. You might know him as.

Leo Laporte (119:14)

She not he can predict when us readers begin searching for her stress free turkey instructions or when her Christmas cake will start its annual climb up search results. This year though, the easy peasy foodie says familiar patterns are breaking. Instead of sending home cooks to her decade old, well tested recipes, Google increasingly inserts AI generated summaries stitched together from bits of her work and others work that often get the basics wrong. For instance, an AI assembled version of Gargano's Christmas cake would have people cooking a six inch cake this big for three to four hours.

Darren Kitchen (119:52)

Oh no, you're going to need that fire extinguisher.

Jason Howell (119:56)

Yes.

Leo Laporte (119:57)

She says you'd end up with charcoal. Meanwhile, traffic to her turkey recipe is down 40% year over year. And it's because. And this is what Google's doing to everybody, right? This is zer. This is the end of the Internet. Google is replacing traffic to people's websites with their AI summaries. The thing that's bad is these AI summaries aren't good. They quote a Mexican food blogger, Muy bueno, Yvette Marquez Sharpnack.

Leo Laporte (120:31)

She says earlier this month she posted two AI generated tamale photos. One with sauce poured over. Now if you know tamales, and I don't know if you do, they're delicious. It's corn with a filling. And then they're wrapped in a corn husk right where. And then steamed. But you, I know Jason knows this because he's, he's around, you know, he's in California. You take the husk off, right? And then you eat the insides. You don't eat the husk.

Jason Howell (120:59)

Well, no, no, no, no.

Leo Laporte (121:00)

That'd be crazy.

Darren Kitchen (121:01)

You can't.

Leo Laporte (121:02)

Yeah, yeah. It would not be good. She posted two photos and one AI generated with sauce poured over the husks, another showing tamales lying flat in a steamer. Both were obvious mistakes. She said. One, the husks aren't meant to be eaten and tamales are supposed to steam upright. She said little details like this are big red flags. These are the AI generated images. And she says this is a bad sign. People are Going to get the wrong idea.

Leo Laporte (121:32)

Hmm.

Jason Howell (121:34)

I didn't know that tamales were supposed to be steamed standard.

Leo Laporte (121:36)

I didn't either. And if you fell for AI generated photos, you might not know.

Jason Howell (121:42)

Not that I know.

Leo Laporte (121:42)

Wrong. She said her husband trusted the post. Her husband wanted to try a recipe on Facebook for maraschino cherry chocolate chip cookies.

Leo Laporte (121:53)

She said she was suspicious of the photos. The cookies were too perfectly pink. But her husband trusted the post because it was, quote on Facebook.

Darren Kitchen (122:03)

Hey, that's all you need.

Leo Laporte (122:04)

The result was a melted sheet of dough with a cloyingly sweet flavor. A disaster, she said. So, just a little word of warning.

Leo Laporte (122:15)

You know, there's lots of. Lots of AI garbage out there.

Jason Howell (122:20)

I mean, in general, when we're talking about, like, you know, an AI, I don't know when it's an AI summary of facts from, you know, like a news story or something like that, generally, you know, I take it with a grain of salt, but often it gets things mostly sort of right.

Leo Laporte (122:37)

I feel like talking about myself and say, I can tell the difference. Right.

Jason Howell (122:41)

Well, there's probably a little bit of that, too. Yeah, that's true. That's true.

Leo Laporte (122:45)

Paris is always saying, I don't know.

Darren Kitchen (122:47)

You don't because you.

Leo Laporte (122:49)

Because I'm Grandpa.

Darren Kitchen (122:51)

No, just because anyone reading something that they do not intimately know.

Darren Kitchen (122:58)

The mistakes that these. These tools make are going to be inherently not very obvious.

Leo Laporte (123:05)

And they're so confident, they're so convincing stated. So, like, I know this answer, but yet.

Jason Howell (123:12)

And I guess what my point was is, is that I would trust a lot less an AI interpretation of a recipe when we're talking about amounts and tablespoons and teaspoons and quarter and.

Leo Laporte (123:24)

But Alex likes all the time that he'll give the AI a couple of, you know, ingredients he has and have it give him a recipe. And he says, I cook them all the time. They're great.

Darren Kitchen (123:35)

Okay. We have been having. I did this week because I have been busy and not buying groceries as often as I did. I just asked because I've been trying out Claude a lot more. I asked Claude to do a full research of recipes that I could make quickly after work for dinner that would also give me leftovers to have for lunch or breakfast the next day. And I have, during the show, been eating a cold sesame noodle thing that I had it specifically prepare for this evening because I knew I wouldn't have time to eat or cook before the show. And it's been great.

Jason Howell (124:08)

And that was a Claude recipe, a cloud recipe.

Darren Kitchen (124:10)

I mean, it's pretty.

Leo Laporte (124:11)

It probably stole it from somebody.

Jason Howell (124:13)

I was gonna say. Yeah. I wonder if it was like stolen from somewhere originally.

Darren Kitchen (124:17)

I believe they'd had.

Darren Kitchen (124:19)

Where were the links on this?

Darren Kitchen (124:23)

It was originally from Gimme some Oven and Table for two blog.

Leo Laporte (124:30)

So earlier today we were talking about AI Generator recipes and you know we have a fun group in our Discord Chat Chocolate milk mini sip. Paul tried putting this recipe in. I have broken Lego overripe bananas and tuna. What can I make for supper? But I was so impressed that ChatGPT did not fall for it. Short answer Please don't cook the Lego longer and tastier answer here's what you can do with what you've got with overripe bananas and tuna. There is no traditional dish that combines them and for good reason. But you can make two separate components that actually taste good. Banana pancakes, which is overripe bananas, egg and a little bit of oats. Tuna salad Tuna melt gave it the correct recipe for that. Although I would put a little chopped celery in there. But that's just me. Serve them as separate dishes. Your taste buds will thank you. And then, and this is hysterical like a little laughing emoji says. If you meant this as a joke, here are some quote creative combinations not recommended. Lego crusted tuna crunchy, colorful and 100% a dental emergency. Banana tuna smoothie. The last thing you'll ever drink voluntarily and a Lego banana casserole. Great structural integrity, zero flavor integrity. I think that was the right answer. I thought that was pretty good.

Jason Howell (126:02)

You know, not immediately groan worthy humor coming from LGBT there.

Leo Laporte (126:08)

It warned you not to to do it.

Jason Howell (126:09)

Yeah, not bad.

Leo Laporte (126:13)

Finally, since we, you know, we only had one nun story today, I think we need another Catholic Church story. So I'm going to put one more in the race to AGI pill the Pope. Now this one I will run by Father Robert. I'm sure he'll have something to say because Father Robert, as some of you know, is our good friend who is in the Vatican and is a consultant to the Pope on technology and AI. This is a story from the Verge today. Robert Hart, writing a team of believers want the Vatican to take AI doomsday scenarios seriously.

Leo Laporte (126:48)

There is a movement afoot to get the Pope to speak out about the dangers of AI. These are AI doomers that want Pope Leo to pick a side in the AGI debate.

Jason Howell (127:05)

Hmm.

Darren Kitchen (127:06)

What do you think he's gonna pick?

Leo Laporte (127:08)

I don't. I don't know. The Verge says he's convening experts. He's Paying attention to relevant conferences and gatherings on AI. I know. He asks Robert. According to several Vatican observers the Verge spoke to, it's something of an open secret that Pope Leo is preparing an AI focused encyclical.

Leo Laporte (127:26)

Which would be, in a way, you know, the doctrine of the Church.

Leo Laporte (127:32)

So.

Leo Laporte (127:34)

Maybe they should be lobbying the Vatican.

Darren Kitchen (127:37)

We gotta get, we gotta get a representative from the Church on here that knows a little bit about AI, whoever could be calling. If only we knew someone that also had great expertise when it comes to cats.

Leo Laporte (127:53)

And drones. He's also, he's also drones. So there are other people, they actually flew to Rome and they're trying to get to talk to the Pope and, and get him to say something about AI Doom.

Leo Laporte (128:10)

I don't know.

Darren Kitchen (128:12)

I also, for anyone interested, I have posted the text of the meal plan recipes recommended by Claude in the Discord live chat. It's been.

Leo Laporte (128:22)

Oh, that's a benefit for members of Club Twit.

Darren Kitchen (128:26)

Yeah.

Jason Howell (128:26)

Lucky you.

Leo Laporte (128:27)

Lucky, lucky you. If you're not a member, you see what you're missing. Twit tv Club Twit. It is a very important part of our monetization. Just as you have your Patreon, Jason Howell, we actually, we use Memberful, which is a Patreon company. We have our own little club. The club, though, makes up a big portion of our overall expenses. 25% of our operating operating costs come from your Club Twit membership. That's a huge deal, as Jason knows. Well, if, if, if we, we. There's no deep pockets here. There's no investors bailing us out. If we don't make enough money to keep the shows on to support the staff we have to cut and we have in the past. We're the reason I'm in an attic with my little ABC fire extinguisher is we didn't want to spend the money on a studio. So we've done everything we can to cut. It's up to you really, at this point if you want to keep Twit doing what we're doing, if you want to keep these shows going, if you want to get the extra shows that we do in the club and we do a lot of stuff. Our AI user group is coming up tomorrow at 2pm Pacific. We're actually. This is going to be fun. Larry lrau in our club Twit is going to. He's been working on a demo of Google's new vibe coding IDE Anti Gravity. So we're going to get a demo of Anti Gravity. Hi, Larry. So that'll be fun. I'm very interested in, in this, but that's the kind of thing we do in the club. It's great. It's so much fun. We have conversations about interesting things going on in our life. Anything geeks might be interested in. If you're not a member, please consider it ad free versions of all the shows. Of course, of course. Because we don't need to show you ads. You're a member. Twit. TV Club Twit. This is a good time now through Christmas Day. We have a 10% coupon for annual memberships. There's a two week free trial. There's also family and corporate plans. The base plan, $10 a month, $120 a year. But again, 10% off when you sign up for a year. Be a good gift too. Twit. TV Club Twit. Enough said. Not gonna belabor the point. None.

Leo Laporte (130:35)

I think this might be a good time for our picks of the week so we can wrap things up and Paris can get back to our new.

Benito (130:41)

Actually, before we get to that, I wanted to ask Jason something.

Benito (130:46)

About all the AI music news that's happened in the last couple of weeks, because that's something I'm personally interested. I know Jason's a musician as well, so I want to know what his opinion is on Warner making deals or Universal making deals with the AI companies.

Leo Laporte (131:00)

Yeah, yeah. And by the way, Spotify has announced that they are making so much music that Suno is creating, or rather Suno not Spotify, that Suno is creating an entire Spotify catalog. What is that, seven, eight million songs every two weeks?

Jason Howell (131:18)

Two weeks.

Darren Kitchen (131:19)

And this is the context also is that Warner Music Group struck a deal this week with Suno settling of its copyright lawsuit against so.

Leo Laporte (131:29)

And I've used Suno and I love Suno, by the way. I am so impressed with what Suno can do now. Jason, he has albums. He's yellow. Is it Yellow Peas?

Jason Howell (131:38)

Yellow Gold. Yellow Gold.

Leo Laporte (131:41)

Yellow Gold.

Jason Howell (131:42)

Yeah, yeah. One word.

Leo Laporte (131:45)

I'm sorry, I was teasing you.

Leo Laporte (131:48)

And. But now they're going to remember it, right? And so you are a musician. As you can see, there's, you know, guitars behind him and stuff somewhere back there. I think that's. I'm glad he's also a musician. Bonillo's said in the past, and I think this is true, the, the joy of being a musician is making music.

Jason Howell (132:04)

So 100%. Yeah. I mean, as a musician I can say that, right? As, as someone pretending to be someone who does not make Music, though, I think these services fill a gap for someone who has an idea or wants to get into the fun of making music and doesn't have the time or the energy or the money or the whatever it takes, the commitment to learn a musical instrument, it gets them at least an entry point into what it's like to have an idea and to come out with something on the other end. As far as the deals are concerned, I think this was just inevitable. I knew at some point, knowing what we've seen, these services were capable, regardless of how they were fed their information and the legality of all that collection of the data that built up the services and all that. Putting that to the side, that's a whole other question. But the quality of what they've been able to do. For me, as a person who loves technology and the magic of really good music technology, I just look at it and I'm in awe. Benito, you would appreciate this.

Jason Howell (133:11)

After 25 or 30 years of making music, I have like tons and tons, hundreds, 500, let's say, old recordings of songs that I've like demoed, or just me and a guitar, you know, from 30 years ago into a. Into a tape recorder or whatever. And what I realized with Suno is that I could take that old recording, put it in there and tell it I want this song, but I want this song modernized. And what it allows me to do as a musician is get re excited, like re excite myself about these old ideas. Because then I go, oh, that's what I could work towards is something like that. And so these tools, I think they get a lot of flack and a lot of pushback because it's so easy to just type out a paragraph and end up with something and then people claim ownership or they claim, like, I created that and, you know, a lot of people will push back on that, but I think there's real value there.

Jason Howell (134:05)

There's a whole lot of complicated issues though, connected to that, and I recognize that from. But from a pure technological standpoint of, you know, just being a fan of music technology, I think it's really interesting and I think. I think I'm not surprised at all that these companies are being swallowed up and they're going to turn it into something legit.

Leo Laporte (134:23)

Does it scare you that Suno? I mean, I don't know if you've used the new version 5, but it's so good.

Jason Howell (134:29)

Yeah, it's really good. That's what I've been using.

Leo Laporte (134:31)

Where you use it too?

Jason Howell (134:33)

Yeah, well, I use it I use it to prototype. I use it to get a sense of like, what could this song be if I was to, you know, spend, spend some time with it, you know, take it, take an old crappy demo and see, like, is this, is this a path worth pursuing for me? And I find myself getting excited about what I hear there. I'm like, oh, that's what I could work towards. And then I'm invigorated and then I go and I make it like as a creator. I then put in my effort as a musician and I create it so, you know, for my. But I'm also not like a career musician. I'm always going to write music because I love it. It's dear to my heart, it's part of my expression, it's part of who I am. So the existence of a suno doesn't offend me outright, you know, like, like it's going to take my job or take. But, but I can totally understand why people are freaked out about it.

Darren Kitchen (135:23)

Does it at all worry you to be uploading, I don't know, old demos of you into something like, you know.

Jason Howell (135:30)

It doesn't worry me. Again, I can, I can understand why it worries other people, but for me, like, these things are 30 years old. Like, who care? I mean, I don't care. It's a 30 year old, you know, song that when I was a teenager I recorded into a microphone. It's going, you know, when I pass away, those recordings are probably going to get thrown in the trash, you know, like no one's going to ever care about them as much as I do and I'm not making any money off it. So I don't care. I just want to do cool stuff and that allows me to look at my music through a different lens. I'm very interested.

Darren Kitchen (136:04)

Cool use of it.

Leo Laporte (136:05)

Yeah.

Jason Howell (136:06)

Yeah.

Leo Laporte (136:07)

Well, it's interesting because iHeartRadio just announced that they're not gonna play any AI music or use AI generated personalities in their radio stations or create any AI generated podcasts. They are one of the biggest are.

Darren Kitchen (136:23)

Now going to say that they are guarant, quote, guaranteed human.

Leo Laporte (136:27)

Guaranteed human.

Leo Laporte (136:30)

For a certain group of people that's like yay, right?

Darren Kitchen (136:33)

Yeah, I mean, I think that's fantastic.

Benito (136:36)

I do think we're about to split here though, where with there's going to be a section of people who just straight up don't want AI music and then there's a section of people who want AI music. Music.

Leo Laporte (136:46)

It's not just music or just don't.

Jason Howell (136:48)

Care, like, because I think there's a certain segment of music, like. Like, I'm kind of of the camp that, like, I know a song I like when I listen to it, and I realize that I like it. Like, do I need to know the provenance of that song? Do I need to absolutely know that a human made that? If I like it, is that the point? And I think that's the case for some people. It's like whether AI is the tool that created it or created a piece of it. You know, Hollywood's going through this too. They're trying to figure out, like, are we okay with AI I'm sorry, but AI is part of your tools, whether you like it or not. It's going to be the same way that Photoshop is used for, you know, editing pictures and everything. That old trope.

Leo Laporte (137:25)

But can you tell if it's AI Generated?

Jason Howell (137:28)

When I hear an AI generated song, yeah, I think I can. It's got a watery quality to it that I feel like is pretty identifiable.

Leo Laporte (137:35)

That's a nice watery wet. Yeah, but for how much time?

Jason Howell (137:39)

Yeah, exactly. Like, the way everything is improving in all other facets. Video is pretty remarkable.

Leo Laporte (137:45)

Right now, they're talking in the discord about Walk My Walk, which is a. Yep, that's right. Artist called Breaking rust. It's a 100 AI generated song, number one on Billboard's country chart.

Darren Kitchen (137:58)

That was, I believe, a couple of weeks ago. An overstatement, I think it was. It wasn't exactly number one. There were some.

Benito (138:04)

It was number one on the chart, briefly. The chart was for the people who purchased digital music. So, like, that's a very small, small list. You can absolutely game that. It only takes, like, three or four thousand buys to get on that list, right?

Jason Howell (138:17)

It's very small. But, yeah, no one's talking anymore compared.

Benito (138:22)

To, like, the actual billboard top 100. Like, this is like the country charts.

Leo Laporte (138:26)

The Billboard Top 100, based on album.

Benito (138:28)

Sales, a bunch of aggregate stuff, a lot more stuff.

Leo Laporte (138:35)

Look, there's only two ways you can listen to music now. You stream it or you buy it. And this is the. That was also streamed 3 million times in under a month. I think this is legit.

Jason Howell (138:46)

I mean, at a certain point. Would we all agree that at a certain point, these things are going to get so good that it's going to become really difficult, if not impossible, to tell the difference between an AI generated song and a real, like. Of course. So then at a certain point, you know, these. I heart. And I'm not saying this because I feel strongly that I, you know about the I heart story.

Leo Laporte (139:07)

No, but there are people who will, right?

Jason Howell (139:09)

But there definitely are people who will. And I just think like I get it right now. Like, you know, right now the music, the technology is not there. But if, if you want to live in a world where you're like. And it's never going to happen because I'm not going to let it happen. I'm sorry it's like that. That's not how technology works. That's not how it's ever worked. And this is just a weird spot that we're in right now.

Jason Howell (139:32)

That's how I feel.

Leo Laporte (139:33)

What do you think Billboard's real charts are based? I mean really, seriously, they're record stores anymore? I mean I used to report to Billboard when I. When I worked in radio and when I worked at a record store I used to report to Billboard. But I don't know is. Does radio airplay count?

Darren Kitchen (139:48)

The charts can be ranked according to sales streams or airplay.

Darren Kitchen (139:54)

And for I'm.

Jason Howell (139:56)

Sure.

Darren Kitchen (139:58)

How much song charts, charts such as the Hot 100 or Global 200, all three data are used to compile the charts. So sales streams and airplane.

Leo Laporte (140:08)

So how much of what you end up listening to on Spotify is AI generated? I bet a lot more than you'd like to think. Unless it. Unless it's the roll.

Darren Kitchen (140:16)

I don't know. What was your Spotify wrapped? Do you have any AI generated songs in that today?

Leo Laporte (140:20)

I have not. I don't use Spotify. Wow. What was your spot so different? Are you a spotifier? Of course. I am a young person.

Darren Kitchen (140:28)

Is that a young person thing? I think Spotify is kind of the big.

Leo Laporte (140:32)

Everybody uses Jason Spotify.

Jason Howell (140:34)

I have Spotify and YouTube Music.

Leo Laporte (140:36)

YouTube Music is a Apple music. The problem, the reason I had Spotify, I got rid of it because I already have YouTube Music, which I was going to get rid of. Then I realized but that also gives me YouTube Premium so I don't have to see ads. I have Apple music because I have an Apple one account. I have Amazon music because I have an Alexa. I have too many. At least three or four.

Leo Laporte (140:58)

I have other. There's others. So I think I just heard of Spotify because both my kids have Spotify.

Darren Kitchen (141:05)

So we should have a. I'm just scanning through. I was on the Wikipedia page for the Billboard charts and the first chart published by Billboard was last week's 10 bestsellers among the popular songs. A list of best selling sheet music in July 1913.

Leo Laporte (141:24)

Winchester Cathedral.

Leo Laporte (141:28)

You ain't Hear nothing yet, mama. All right, now we're gonna take. It's. Well, there's no break to take. We've done our breaks. Now we can do our picks of the week if you would like to. We like to start with Paris Martineau.

Darren Kitchen (141:43)

I'm trying to decide what I want to do. We didn't get to talk. Should I just read some quotes from.

Leo Laporte (141:48)

Sure. This is the ones you can read.

Darren Kitchen (141:52)

The ones I can read. Hers actually isn't very crass. So this is the book published by Olivia Nuzzy, who famously got fired from New York magazine a year ago for having an emotional affair, an affair of some sort with RFK Jr. Who she was covering.

Leo Laporte (142:09)

He says it was emotional.

Leo Laporte (142:12)

Lizza, her ex.

Darren Kitchen (142:14)

Other people say it's not. I don't know. Anyway, this is her memoir that she wrote that's being ripped apart now kind of in the press for its interesting writing choices. I'll read some sentences picked out by Scotchy, the Great blocker or the Great Gawker blogger who says she had to review Olivia is Nezzi's new book for Slate, and she did a blog called Sentences from American Canto that sent me to the hospital. The first is see if you can track. If you can follow this. The flag winked beside lanes that bent to borders that faded to barriers that felt the lines I crossed. I am talking, of course, about how it happened between me and the politician.

Leo Laporte (142:57)

It might be helpful to understand she claims she writes all her works on her phone.

Darren Kitchen (143:03)

Yeah, and I feel like.

Darren Kitchen (143:06)

Here'S something that is a useful thing that I guess could go with. I think I mentioned this in the chat. I was once gifted a drone. I named him Pistachio since he was so tiny. Page 44.

Darren Kitchen (143:21)

The movie star. Not a fraud at all. In fact, a shock of honesty in this weird little town approached and grabbed my face. Olivia, the secret to life is to be rapeable, she told me. You are rapeable.

Leo Laporte (143:34)

Oh my God.

Darren Kitchen (143:35)

Weird. What are. Oh, one of them is. This is a line pointed out in the Washington Post Review as a description, as an example of her metaphors growing in feverishly mixed. She writes, it was as though the media was holding up a doll of me and gesturing at random to different parts in search of kindling to feed the fire of the story.

Leo Laporte (144:00)

Oh my God.

Darren Kitchen (144:01)

It's just then that we came from fire and returned to fire, that we move forward only because we have learned to tame some fires. That here, at least untamed fires is the greatest present threat to life.

Leo Laporte (144:15)

Honestly, sounds like she might have A mental illness issue.

Darren Kitchen (144:20)

Seriously, I think that she. She.

Darren Kitchen (144:24)

Went through something. I. I really actually enjoyed the Washington Post's review of this, which is Olivia Nessie tries and fails to save her reputation in American Canto. And where was the.

Darren Kitchen (144:39)

Basically, she says that you can't write a memoir unless you are willing to kind of bury your soul in it. She said, in the end, Olivia did go native in her time in Washington, which is one of the things she says. But her account of her infatuation and the ensuing ordeal is a string of overwrought emotions. You shouldn't write a memoir unless you are willing to make yourself look foolish and pathetic. Nuzzy breaks this cardinal rule, flattering herself by admitting only to the chicest types of disintegration. And I think that that is. It's part of it. She goes on. This author goes into good literature and good gossip have been common that they are both savagely and mortifyingly honest. They plumb the depths and reveal the sort of details. They don't save face. They rip it off to expose the raw veneation beneath. And I think that that's kind of good. We're all having a lot of fun making fun of the back and forth between Olivia Nazi and Ryan Lizza and all the media drama of it. But the thing about this that I do find interesting and a bit disappointing is if you go through, most of the reviews of Olivia Nezzi's book begin with some great author who's writing this reviewing like, I was genuinely a fan of Olivia Nuzzi's reporting.

Leo Laporte (145:50)

Right. Almost all of them say that everybody.

Darren Kitchen (145:52)

Was that I was too, and still am. I think the stuff she had written was great, even though now I feel complicated about because I don't know who she has and hasn't slept with.

Leo Laporte (146:00)

But why would that matter? Explain to people not in the business you.

Darren Kitchen (146:05)

It would be.

Darren Kitchen (146:08)

She has been alleged to have had romantic, if not physical relationships with two of the politicians she covered, which is not acceptable.

Leo Laporte (146:18)

And to have written a profile of a politician that her lover at the time was trying to influence.

Anthony Vinci (146:27)

Yeah.

Leo Laporte (146:28)

And there's also perfect juxtaposition between you and her. You won't even. You won't even vote. You won't talk about politics and won't even vote because you need to not be biased.

Darren Kitchen (146:41)

Yeah, I won't like participate in political demonstrations and things like that because I think it's important to continuously maintain neutrality.

Leo Laporte (146:52)

And this is what everybody says. Well, maybe she's more of a Tom Wolf or a Hunter s Thompson, where she's gonzo. She's one of the things embedded in.

Darren Kitchen (146:59)

The story Liza, her ex, published and claims is a document he found. I think on her computer is a memo that Nuzzi allegedly sent to RFK Jr. Basically spelling out how he should approach being excluded from the presidential debate. And it goes so far as to be like, you need to wear a suit, a blue suit, because you're a summer. You're a summer unlike some other politician, you know, and you need to pitch your stream to these specific news networks, and if they don't agree, then you need to go to this host and basically acting as a political advisor, which is the completely antithetical to journalism. But I think the thing that is so sad about all of this is it would have been a. I was originally really excited to read this book, American Candidate, despite all the things I just said, because I was like, well, Olivia Nessie is a famously good writer. She clearly made a terrible mistake. Grave lapse in judgment. It would be. It'll be a fantastic book to read. How she fell into such depravity, how she lost herself so much, what covering Trump World from the front lines took from her. And it is upsetting that there doesn't seem to be as much self reflection in that as anyone would have hoped.

Leo Laporte (148:16)

She's 32 now. She was very young when all of this began.

Darren Kitchen (148:21)

I mean, yeah, that's also, I think, a huge part of it.

Leo Laporte (148:23)

Yeah, it's. It's just pretty sordid and tawdry and I prefer to think of better things. But.

Darren Kitchen (148:32)

But I will keep texting you every.

Jason Howell (148:34)

Time.

Darren Kitchen (148:36)

Leo's saying this, but yesterday he was like. To me and Jeff were going back and forth, everything. He was like, why does this always happen when I'm on the air?

Leo Laporte (148:43)

I'm in the middle of a show and I'm getting text after, after text after text, and it's like I can't even read the things you're talking about. It is. Look, it's amusing. Mostly I'm amused by you and Jeff and.

Leo Laporte (148:56)

All the energy around it is very amusing.

Darren Kitchen (148:58)

I was gonna say, this is perfect. Me and Jeff bait.

Jason Howell (149:00)

Yeah, it is.

Leo Laporte (149:02)

Thank you, Paris. Did you want to share a pic with us?

Darren Kitchen (149:06)

That was my pick.

Leo Laporte (149:07)

Oh, I share this pic with you, actually. This is a good pick, Jason. Go ahead.

Jason Howell (149:11)

Yeah.

Jason Howell (149:13)

I forgot that there was a whole pick thing. And I was like, well, I guess the Suno thing could have been a pick, but I gotta have a pick. And then I was looking around, I was like, oh, yeah, Loop earplugs. So.

Jason Howell (149:27)

I just started in a new band. I haven't been in a band for, I don't know, 25 years. I've just done music by myself. And then friends of mine were talking about their ween. Have you ever heard of the band Wean? Wean?

Leo Laporte (149:38)

You ever heard Wean? That sounds familiar. Are you in Wean?

Jason Howell (149:42)

Well, I'm in a Ween cover band, and I play keyboards, which is very interesting because that's not a key. That's not a. Like an instrument that I. I consider at the top of my. My list of. Of capabilities. But I'm learning anyways. It can be a loud band. And I forgot that I had these loop earplugs, which are kind of like. They don't look like your standard typical earplugs. The cool thing about these is that they don't just block everything. You know, sometimes you put in earplugs, and if you're a musician and you're in a live music environment or you go to a show, you put in earplugs and you can hardly hear the music because they just dampen with, you know, reckless abandon. And the cool thing about the loop earplugs is, A, they're not that expensive. B, you know, I wish I had. Oh, actually, I do. I have an overhead shot. Do I? Yeah, I do.

Benito (150:32)

Wow.

Jason Howell (150:33)

They kind of have a little bit of a. Kind of, you know, fancy quality to them, I guess, the little metallic piece and then the nub that goes in your ear. But they come with a lot of different pieces that you can attach to the end that give you different noise reduction, also give you, of course, different fit. And I really enjoyed wearing these when I'm doing the band practice because I don't feel like I'm isolating myself from. Into, like a muffled room off in the distance. Yeah, so that's. I was on the website, scrambling, trying to.

Leo Laporte (151:04)

You didn't get the Swarovski crystal one. I can see that.

Jason Howell (151:07)

Not the Swarovski crystal.

Leo Laporte (151:08)

I have two pairs. I have one for sleeping, which is smaller and is out of the way.

Darren Kitchen (151:14)

Do you have a hard time sleeping without.

Leo Laporte (151:16)

No. And I never wear them, but I.

Darren Kitchen (151:17)

Got them just in case.

Leo Laporte (151:19)

It was actually when we were going to go on that trip up the Mississippi, and I'd heard it was very noisy, so I got them for that. And. And I. And I'd done some research. I think I saw it on Reddit, and then I got a pair because I always wear earplugs. When I go to shows. You're so smart, Jason. I have bad tinnitus anyway, but I don't want to make it any worse. So I got some of the kind of more standard ones. But what's cool about these loops is you can put a little rubber ring in them that attenuates the music even more. So you have control about how much attenuation there is on the music. And, yes, they do. Unlike just stuffing something in your ear, they really do give you a pretty good quality. They don't sound too.

Darren Kitchen (151:59)

Have they been tested? Do they actually dampen the sound enough that it protects your hearing, though?

Leo Laporte (152:04)

Oh, yeah. There's. It's. I think it's 10. 10 decibels. I can't remember. They say they have the. There you go. Ready to rumble.

Jason Howell (152:11)

Yeah. So these are the Loop Experience plus. That's what they are. And then when you get. The box comes with all these different kind of fittings for the tips.

Leo Laporte (152:20)

But there's also a little rubber ring you can put in the. In the loop, I believe. Oh, so they have. So the experience mode is 23 decibels down. Quiet mode is 26, which is quite a bit. And so, you know, if you're at a concert and it's 100 to 120 decibels, that's deafening. That's very, very loud and will damage your hearing. Anything over 90 for an extended period of time will damage your hearing. So if it's 120 decibels and you're able to get 26 decibels down, that's going to protect your hearing. That's going to really help it. You should still get out of there.

Darren Kitchen (152:56)

I need to read this review that's just casually at the top of Loop's thing. It was just left by someone. 5 stars. Awesome. They look amazing. They also really help block out sound. My bird screams a lot and putting these in blocks out most of his screaming. I can still hear people talking to me.

Leo Laporte (153:14)

Very useful.

Darren Kitchen (153:15)

It keeps going on. And I was like, your bird screams a lot.

Leo Laporte (153:19)

Yeah, Bird people are different. They're different.

Jason Howell (153:21)

They put up with a lot. Yeah.

Leo Laporte (153:23)

Different type, different breed. Yeah. So, yeah, the loop experiences are for concerts. Let me see what they do. I think there may be even more or reduction. But I really like them. I think they're really good. And they. You can get a little magnetic, which I did. Little magnetic cable so I don't lose them.

Jason Howell (153:41)

There's like a whole loop loop road that I haven't driven down.

Leo Laporte (153:45)

There's a whole Loop road.

Darren Kitchen (153:46)

It's a loop.

Jason Howell (153:48)

It's. It's a whole loop loop. But part of the reason I bring it up is because, of course, because it's close to Christmas, they're having a sale, and they're not that expensive. You know, it's like sometimes you. You know, it's like dirt cheap or really expensive when it comes to ear protection and stuff. And these are really affordable, but not dirt cheap. You know, they're probably like 30, 40 bucks, but they got.

Leo Laporte (154:07)

So I have to hear that.

Jason Howell (154:08)

You.

Leo Laporte (154:08)

You're protecting your hearing. Paris, when you go to concerts.

Darren Kitchen (154:12)

I do, yes. I used to hate concerts until I realized that you can wear earplugs and sit down.

Leo Laporte (154:20)

The AirPods also, you don't have to dance. Huh?

Benito (154:24)

I will.

Anthony Vinci (154:24)

Yeah.

Darren Kitchen (154:25)

If I end up. I. Like when I was at a bachelorette party, we ended up at a club, and I hadn't brought earplugs, and I was, like, so physically uncomfortable because I'm hearing getting damaged. But then I put the air plug.

Leo Laporte (154:37)

AirPods work really well, actually. Yeah, they're. I've. I've gone to shows with AirPods, too.

Darren Kitchen (154:41)

Is it. Does it actually help, though? Because it's active noise cancellation.

Leo Laporte (154:45)

Is that not damaging if you have a good seal?

Jason Howell (154:48)

Yeah, if you have a good seal, you're sealing them. Yeah.

Leo Laporte (154:51)

So it's not that sound is being protected. Absolutely.

Darren Kitchen (154:54)

But the active noise cancellation isn't producing other noise that cancels it out. No, but it is also damaging. Okay.

Leo Laporte (155:01)

No, it's.

Darren Kitchen (155:02)

I don't know how noise works.

Leo Laporte (155:03)

It's bringing it down. Well, you got a wave like that, and then the active noise cancellation gets a wave in the other direction.

Darren Kitchen (155:10)

But the fact that there's another wave, that's not.

Leo Laporte (155:13)

No, you're not hearing the other wave. They're canceling each other out.

Jason Howell (155:15)

No, they cancel each other out.

Leo Laporte (155:16)

So there's a. There's a trough and there's a hump and the hump and the trough, and they just zero out. It's amazing. It's a miracle. Yeah. Yeah, that's. That's a good. You know what? I bet you people don't know that I legitimately.

Darren Kitchen (155:29)

Noise cancellation is not making more noise multiple times before. I guess it was many years ago, but I never found a straight answer. Yeah.

Leo Laporte (155:38)

So they do two things. They seal your ear because of the tips. If you've got good tips. So that's cutting it. And then the noise cancellation is bringing it down even more. Yeah. It's not putting more noise in your ears. Good news.

Darren Kitchen (155:50)

Good news.

Leo Laporte (155:51)

That's because they have microphones on them and they can. And the circuitry in there takes the music.

Darren Kitchen (155:56)

They figure out what this is and.

Leo Laporte (155:58)

They make the music, shifts it over 90 degrees and then the trough fills the wave and the wave fills the trough and you get nothing.

Jason Howell (156:09)

Or takes the air noise of the airplane and does same thing.

Leo Laporte (156:13)

That's why it works with continuous noise. But it doesn't work very well with, with transit noise, like because it doesn't have enough time to generate a canceling wave. Yeah, that's a good. I'm glad you brought that up. I thought that was common knowledge. The more you know, the more you know. Paris Martino. Everybody should read her. Are you working on like the next big exposure Consumer Reports soon? You gotta subscribe.

Darren Kitchen (156:40)

You don't need to subscribe, but you should because you should support good journalists journalism. Yes, but you don't need to subscribe to read my stories. But you should because it's really affordable.

Leo Laporte (156:49)

Are all of your stories outside the paywall?

Darren Kitchen (156:52)

Yeah.

Leo Laporte (156:52)

Oh, I didn't know that. Is that part of the deal you made?

Darren Kitchen (156:56)

I mean, no, it's just our special projects team does. Because you're doing good work for people, consumer advocacy work. So that's inherently an in front of the paywall thing because we want it to reach as many people as possible.

Jason Howell (157:09)

Nice.

Leo Laporte (157:10)

Very good. And again, Jason Howell.

Jason Howell (157:13)

Yeah.

Leo Laporte (157:14)

All inside. All inside. AI inside. I want to say all inside. A inside. AI inside show. AI inside show.

Jason Howell (157:23)

Yeah. And the lower Yellow Gold.

Leo Laporte (157:24)

What's the yellow gold website?

Jason Howell (157:26)

Together. You know, I don't really have a yellow gold website. Just go on to Spotify or YouTube Music and do a search for yellow gold. One word. You will also find an artist called Gustavo. Gustavo. Yellow Gold. I think that's not me. It's just yellow gold.

Leo Laporte (157:40)

Just plain old yellow gold. And where does the weaned tribute band perform?

Jason Howell (157:44)

Well, soon in the Bay Area.

Leo Laporte (157:46)

Oh, exciting.

Jason Howell (157:47)

Legit. When I went to my first practice with them, I was like, oh my God, you guys are like really good. And so I've got a, you know, I've got a level that I've got to work up towards and I'm working on it, but I think, you know, it's fun.

Leo Laporte (158:00)

Do they have a clever name? Because, you know, if you're a cover band, like we have the Illegals, we have Fleetwood macrame, we have a She D. She?

Jason Howell (158:11)

Yeah, ACD She.

Leo Laporte (158:13)

Yeah, it's all girl. ACDC tribute band. Do you have a like ween, like not ween or we. We ain't. Or. I don't know. You need a clever name. You have a clever name.

Jason Howell (158:22)

There needs to be. There's an old ween song that I really like called Voodoo lady that I think would be a good, good band.

Leo Laporte (158:30)

Voodoo Ladies. The Voodoo Ladies.

Jason Howell (158:32)

Voodoo Lady. But, you know, there's a million names. We actually have a group chat going on and I swear, half the. And it's always rolling and half the time it's. It's like, band name is.

Leo Laporte (158:41)

So the name was going to be. Yeah.

Jason Howell (158:43)

Before we go, can I mention something real quick?

Leo Laporte (158:45)

Oh, just this. I wanted to. I'm sorry, I meant to bring this up.

Jason Howell (158:48)

Yes, it's fine. No, I'm happy to just have like 30 seconds.

Leo Laporte (158:51)

Jason's doing something really cool.

Jason Howell (158:53)

I mean, I'm playing around with ideas. It's the end of the year. It's, you know, and I'm like, looking at December, I'm like, I want to. I had an opportunity to help someone, a friend of a friend who had a podcast, and they were looking to take it more seriously. And I hopped on a couple of calls with him and it was just. It just made me feel good to like, take my 20 some odd years of podcast experience and tell him things that I take for granted. But that really impacted his, you know, how he approaches his. His podcast. And I was like, I want to do a little more of that. So I decided that for the month of December, I'm opening up my calendar for like 10 people, 20 minutes. It's not a huge amount of time, but it's the amount of time I have. And if you have a podcast or if you're thinking about podcasting or whatever, like, I want to get a sense of, like, what kind of need is out there or what kind of problems people are looking to solve around podcasting. I might not have a solution to everything or an answer to everything, but I'd certainly love to talk with you about it and maybe I can give you, you know, some insight that you might appreciate. And I would learn a little something about, you know, the demand for something like this. Go to Bit Ly Podcast Tuneup. I just set that up before the show. That'll take you to a substack post that I did like a week and a half ago talking about this. And that gives you the. The way to. To say you're interested. So anyways, that's. Anyone who has a podcast or thinking about it, reach out.

Leo Laporte (160:17)

What a great thing to do. Yeah, Lisa told me about it and I thought, oh, that's really cool. Bit ly.

Jason Howell (160:25)

Yes. Podcast Tune up.

Leo Laporte (160:26)

All one word Podcast Tune up for some help from a guy who's been doing it practically as long as I have. When did you start with my professional.

Jason Howell (160:36)

Career pretty much is podcast. I started in 2005. I got an internship at ZDNet and I think at the time I was helping out with a show called Dan and David with David Berlin and Dan Farber. Remember?

Leo Laporte (160:48)

Yeah.

Jason Howell (160:49)

And then I got pulled over to Buzz out loud in 2006 and I've.

Leo Laporte (160:53)

Asked ChatGPT and it suggests the Bugnish Brothers. Pure Guava Revival.

Jason Howell (160:59)

I like that one.

Leo Laporte (160:59)

Chocolate and Cheese Whiz. The Mollusk Men. Dean and Gene adjacent. Voodoo Lady Review. Mutant Book Nish. I don't know what any of these mean. Pork Roll egg and tribute. Captain Fantasy's crew.

Jason Howell (161:14)

Captain Fantasy is another one that I'm. That I'm amping, amped for or Buenos tardes.

Leo Laporte (161:19)

Imposters.

Jason Howell (161:20)

Love it. Yeah. By the way, yeah, Ween is a very unique band, but they also rock. It's everything. The weird thing about Ween is they are everything. They. They have us. They might have a bossa nova song next to death metal song next to a. Like, they're everything and they just do it all so well.

Leo Laporte (161:43)

Wow. Yeah, that's really cool.

Jason Howell (161:45)

It's really unique. And when I found out that they were looking for someone, I was like, oh my God. When I was a teenager, the band that I emulated when I first started recording music was Ween.

Leo Laporte (161:55)

So, you know, I know a lot.

Jason Howell (161:57)

Of the songs and they were like, well, I guess you have to be in the band then. Like, okay, great.

Leo Laporte (162:02)

Hey, let us know when you play. I'd love to come see.

Jason Howell (162:04)

All right, we'll do.

Leo Laporte (162:05)

Thank you. Jason Howell. Great to see you again. Paris Martino. Thanks to all of you. A special thanks to our Club TWiT members. We do intelligent machines every Wednesday, 2pm Pacific, 5pm Eastern, 2200 UTC. You can catch us live in the club Twit Discord, of course, but also on YouTube, Twitch, X dot com, Facebook, LinkedIn and Kickstarter. You also can watch us after the fact, which is probably more convenient. You can download shows, audio or video from our website, Twit TV IM. There's a YouTube channel you can use to clip little bits and send it to friends and family or just watch the video there or subscribe to the audio or video versions of the show in your favorite podcast player and leave us a great review because Paris has got to do some dramatic work readings next time.

Darren Kitchen (162:52)

That's true.

Leo Laporte (162:53)

I do make it interesting, make it funny, make it exciting.

Leo Laporte (162:59)

Thank you, Paris. Thank you, Jason. Thank you, everybody. We'll see you next time on Intelligent Machines. Bye. Bye. I'm not a human being.

Darren Kitchen (163:11)

Not into this animal scene. I'm an intelligent machine.