Who Doesn't Like A Show About War Crimes and Gambling?

The problem in our current era is that we've turned spending money into passively clicking and tapping on stuff. Great for the tech companies. Real bad for your wallet. BJ is also joined by David Gerard from Pivot to AI to talk about Palmer Lucky's sketchy trip to the West Point Military Academy.

Share
Who Doesn't Like A Show About War Crimes and Gambling?
Photo by Mohammed Ibrahim / Unsplash


Hello everyone, welcome to another edition of Privacy Tips, the sister podcast for Stupid Sexy Privacy. I'm your host, BJ Mendelson.

Today's show is made possible by our friends at DuckDuckGo. So if you enjoy what we do, check out their DuckDuckGo subscription. It comes with a simple and easy-to-use VPN, identity theft protection, data removal service, and more.

Additional information can be found at DuckDuckGo.com/subscribe.

(And some features and offerings will vary based on your location.0

Now my friends, Stupid Sexy Privacy has wrapped, and you can find all episodes online at stupidsexyprivacy.com. All you have to do is click or tap on where it says episodes.

The purpose of Privacy Tips is to give me an opportunity to read from my version of How to Protect Yourself from Fascists and Weirdos. Think of this podcast like the DVD commentary track that you can listen to after you finish the book. At the time that I record this, which is Tuesday, October 13, 2026.

We appear good to go on getting the audiobook published. So I'm hoping within the next two weeks I can actually tell you the book is now available. And then I can hopefully tell you how I'm going about promoting it. I have to confess though that I am dreading book promotion. It's been a while since I've done any sort of media outreach, and it was always that fear that maybe time has passed me by and my marketing abilities have atrophied.

I used to be a world-renowned expert in word-of-mouth marketing. But that was also before the pandemic. So we're living in a whole new world six years later. If I'm as good as I used to be, then things will go well promoting the book. If not, well, we'll see what happens.

In other words, if I try to promote this thing and just utterly fall flat on my face, first feel free to laugh, but second know that I'll be laughing along with you, because I don't have high expectations for my abilities six years later, so be prepared.

I do have a plan B though. You may see our Stupid Sexy Privacy host, Rosie Tran, do some publicity for the book and the podcast more than you see and hear from me. Meaning I may work my ass off to get some spots and promo lined up.

But it's actually going to be Rosie doing all the talking. That's my hope anyway. We'll see how that plays out.

If I can have anything I want, I'd love to have Rosie do all the promotion. And I can just keep my head down and work on this show and the next book. And that's, you know, something I actually wanted to do years ago.

And this is the whole point of this podcast. It's behind the scenes, right? So I don't think I've ever told anyone else, but years ago when I first got attention and first got an agent, I was writing fiction and I didn't want to do publicity at all. I just wanted to write. And so years ago, I had a friend whose name is EB, and I said, "EB, when I publish this novel, as far as anyone knows, you're the author. The only thing I'm going to ask you is that if you do any interviews, you have to do it like completely in character as this author."

And we had like this whole plan where she was going to do the press and promotions in character and it was going to be funny.

And I just didn't finish the book because I wound up writing Social Media Is Bullshit instead. And I guess the rest is history. So we won't be doing that here with How to Protect Yourself from Fascists and Weirdos. Rosie is very much Rosie. If you see Rosie doing the interviews, that is Rosie being Rosie. She's not doing a character.

For me, you know, I take care of three disabled adults and one of whom right now is sick, and my schedule is going to be kind of erratic, just as this book is coming out. So it's hard for me to schedule something with a press without the possibility that I may need to reschedule it at the drop of the hat. And that's not very good when it comes to marketing and publicity. So I'm kind of hoping that Rosie might be able to help alleviate that. So if things go well and you see and hear a lot from Rosie and not a lot from myself or Amanda, that's good. It's totally fine. That's part of the plan. That is what we're doing.

And just an FYI, Amanda is over in Australia. So my co-author is not in the correct time zone to do media or to be effective at press outreach over here in the United States. But she may do some press outreach over in Australia. So we'll see what happens with that. Anyway, that's all behind-the-scenes stuff. It's something I just want to flag for you because the book is coming out and the publicity is about to begin. It just might not be me doing all the talking. We'll see.

Now let's get to this week's book excerpt. And make sure you stick around for a bonus interview we did with David Girard from Pivot to AI. This is an interview that ran over on our sister website, The Monroe Gazette, where we talked with David about Palmer Lucky's bizarre West Point presentation that he made to cadets about his adorable little weapons systems company and the role industries play.

We covered it for the Monroe Gazette because it happened at West Point, which is in Orange County, and that's what the Monroe Gazette covers. But for people who are a more national and international-facing audience, which is all of you listening to Stupid Sexy Privacy, I wanted you to also check out this interview. It's really good.

This Week's Book Excerpt

Okay. Now we're going to be doing our book excerpt. This is a really short excerpt. It's the end of chapter two. And like I said last week, I kind of wanted to flag a point of separation between the first two chapters and the rest of the book. So I'm going to read from you the book excerpt, and then I'll explain why chapter one and chapter two are kind of different from everything else you're about to hear and read. So let's do the excerpt first.


Lastly, transactions on your phone are great because they're frictionless. But the flip side is that you're more likely to overspend because of how easy it is to buy something or get addicted, which is how predatory gambling apps and those pretending not to be gambling apps by saying they are prediction markets make their money. Whenever possible, try to make all of your transactions in cash. Buy now, pay later apps, like credit cards and crypto, are a scam. Cash, while still traceable, is far less so than a digital transaction. Resist the efforts to get rid of cash. Cash is the most surveillance-resistant choice we have when it comes to money and financial transactions. I don't want to blow by that point there. We kind of blow by it in the book, but ProPublica just recently—I think just in the last week that I'm recording this episode—put out a short documentary about someone, a reporter who essentially decided to behave like a problem gambler on DraftKings and FanDuel and document all the ways that these companies are constantly pushing you to make bets. And as you might have guessed, you always lose money on these things. Even when you think you're making money, you will, on the whole, lose. And one of the things that the reporter says in the video is that, you know, it's so easy to get hooked on this because it's not even like you're spending money. You're just tapping buttons and the money is spent as part of that. And that's true of a lot of these apps, which is why it's so important that, you know, when we say to you transactions on your phone are great because they're frictionless, that's true. But the flip side is you may not have a full accounting of your money when you use all these apps and when you get addicted to the gambling apps or when you do anything else online. Like, when money becomes a simple matter of clicks and taps, that's very easy to lose track of. So when we tell people to use cash instead of their phones to do banking, we always get puzzled looks.

And they usually ask this question: "Well, how are you supposed to keep track of your money if you're using cash in envelopes instead of a phone?" So we're going to close this section in chapter two with an old-school recommendation: the envelope method. It works exactly how you think it does. You figure out what your budget is; you label each envelope according to each line item in your budget, which means you have to make a budget; and then you allocate your cash based on that budget into each envelope. I'm going to read that again: "The envelope method works exactly how you think it does. You figure out what your budget is; you label each envelope according to each line item in your budget, which means you have to make a budget; and then you allocate your cash based on that budget into each envelope." It's deceptively simple. And I know, you know, look, the idea is supposed to be when you're out of money in the envelope, you're out of money for the month. We get it. This is imperfect. Depending on your circumstances, people may have a very good reason for not wanting cash in their house. We get that too.

But if you can—if you can—we want you to consider this: You won't believe how much people lose each year to recurring subscriptions, for example, that they forgot to cancel and to overspending because of how easy it is to tap and purchase something on your phone. To avoid losing money to recurring subscriptions, print out the last 90 days of your bank statements and make sure there's nothing you forgot to unsubscribe from. And yes, we do recommend manually printing this out and holding it in your hands as opposed to looking on your phone. Don't let an app do this for you, by the way. You don't want anyone getting your financial data. And then if you worry about the subscriptions coming back or creeping back up on you, which they do, all you have to do is repeat this process every 90 days to look for shady transactions. Which, by the way, is a great way to look for identity theft. So it sounds simple. It is. The problem is that a lot of people don't do the simple things, which is how we get tripped up and trapped by fascists and weirdos.


Now just real quick, I want to give you a behind-the-scenes note about chapter one and chapter two. They are the original pitch for this book. So when we originally pitched or at least discussed how to protect yourself from fascist weirdos starting back in 2022—it might even start in 2021, to be honest with you; I've lost track because I smoke way too much weed to sometimes keep track of these things—but we've been talking about this book for a while. And chapter one and chapter two was essentially what the book as proposed would have looked like.

The thing is, we ran into a bit of a hiccup. You know, in 2022, I suffered from pretty serious bipolar depression that lasted for a couple of years, and I didn't come out of it until fall of 2023. So there was a long time between this idea of this book and what it would have looked like and the book that you're actually about to listen to when it comes out on sale. After rebuilding my confidence as a writer who can meet consistent deadlines, which I did over at the Monroe Gazette, I went back and pitched DuckDuckGo about doing a revised version of How to Protect Yourself from Fascist Weirdos. And that was in late 2024, going into 2025. The original agreement was again for 20,000 words, and that again is basically chapter one and chapter two. So chapter one and chapter two is what I pitched DuckDuckGo on after coming out of that depression, saying, "Hey, we really should publish this book." Chapter one and chapter two are what we pitched them on. But as you might have guessed, something happened. And the 20,000 words that follow—which you're going to hear starting next week on Privacy Tips; we'll start the full book next week—is all new. Everything that you're about to hear in the next section and throughout the rest of the book was written in February of 2026, following the murder of Renee Good and Alex Prady.

Essentially, I just kind of looked at chapter one and chapter two and I said, you know, this is good, but I really struggled because I didn't know if it was good enough. You know, I originally thought about cutting chapter one and chapter two off and releasing them separately as originally planned with DuckDuckGo. But cutting them off kind of felt like cutting off an important part of the umbrella we're trying to give you. You know, the whole point of this book is to shield you and help you give you the skills you need to shield yourself from the torrential rain made of bigotry, ignorance, and evil that we're all trying to fight. So I wound up keeping it. But I just wanted to let you know that because chapter one and chapter two are definitely different from the rest of the book.

They're a little more technical. That's not to say there aren't some technical things that don't pop up throughout the rest of the book. There are a couple of things, but chapter one and chapter two are kind of front-loaded with all the basics, the specifics, the things that you can do. The rest of the book is much more tactical and much more focused on the long-term vision and strategy. So chapter one and chapter two are no less important and no less valuable, we think, to share with you, which is why we went through and which is why we included it. I just wanted to let you know because I realized there's no print version of the book. There might not ever be. There might be, depending on how it does. I can't at this point in time tell you because I don't know. I really don't know if there will be a print version or not. But if you were to read the print version, if you look in the manuscript, it's chapter one, chapter two, and then there's a break. And then at the end of chapter two in the manuscript, it says, "And now we'd like to present How to Protect Yourself from Fascist Weirdos."

So if you're listening to the audio version, you're not going to see that line of demarcation in the produced version coming out. It's just going to go chapter one, chapter two, right into chapter three, without that line of demarcation. So I wanted to just stop here and tell you that there was a line separating one and two from the rest of the book.

Okay.

All that being said, that's probably more information than you need or want to know, but now we're going to go into our interview with David Girard. I recorded this a few weeks ago with him over at the Monroe Gazette. Like I mentioned, that newsletter is very focused on Orange County, New York. So if you're listening to this, just based on the statistics, you are not based in Orange County, New York. You are based all over the world and throughout the United States. So you probably missed it. It's a very good interview.

You should know what Palmer Lucky is up to. David does a really good job of explaining the weapons systems that is essentially what Mr. Lucky is selling. And also, he was doing this in front of a whole bunch of cadets at West Point, which I found to be really uncomfortable.

Our Interview With David Gerard

BJ Mendelson: David, I had you on our other show, Stupid Sexy Privacy, but there’s a lot happening over in Orange County, New York. And so I was hoping for a chance to have you back, and here you are. But for people who did not hear that first interview, would you like to take a moment just to introduce yourself?

David Gerard: My name’s David Gerard. I’m a journalist. I do Pivot to AI, which is a blog, a YouTube channel, and a podcast about the AI bubble and why all of these people need to go to jail but probably won’t.

BJ Mendelson: Mm-hmm.

David Gerard: And how we’re heading for a Great Depression to depress harder. Just—it’s a very cheerful blog, as you can tell.

BJ Mendelson: That’s pretty much the vibe of The Monroe Gazette. It is: all these people should be in jail, but they’re not going to be. We’re just going to write about it anyway, because someone has to. You had a great video last night just to give people an idea of what Pivot to AI puts out, because I watch it regularly. Like, whenever there’s a new video, I will watch it.

David Gerard: Comes out daily.

BJ Mendelson: Yeah, so tell us about the video last night where there was almost a little whoopsie between the United States and China caused by AI.

David Gerard: That was very good. Where the existential risk of AI is not that the AI will take over and turn us all into paperclips—because that’s science fiction fantasy promoted by a literal cult. The hazard is that stupid people will think the magic truth machine is a magic truth machine and do stupid things because, well, “computer says kill.” And it’s worse because the current Department of Defense—or to use its proper name, the Department of Defense—is full of very, very stupid people who believe the chatbot is the magic truth machine. Pete Hegseth is a person of limited capacities, but he absolutely believes AI is the best thing ever. And yeah, I can’t say I’m convinced myself.

BJ Mendelson: I can’t either. And you know, Hegseth has a lot of influence in our community over what happens at West Point. You know, there was just the lawsuit that we covered with Professor Tim Bacon, where Hegseth and Trump essentially wanted to censor the civilian professors at West Point so that they can only talk about whatever the Department of Defense approved. So my ears always perk up whenever you cover that stuff, because there’s a trickle-down effect. But before we get to that, just before we came on here, I wanted to get your take on commercials that just came out. Like, as we’re recording this, they’ve just come out from Kathy Hochul—all of a sudden she’s reversed course on these data centers. And I was just hoping to get your impression on this new commercial.

David Gerard: Yeah, I had a look at the commercial just before we popped on. It’s just a 30-second campaign ad. The entire thing is: I’m against data centers. My opponent loves data centers. That’s the whole thing.

BJ Mendelson: Right.

David Gerard: Right? Now that’s very good political marketing in the lead-up to the November elections, because the populace hates data centers with all their hearts. There’s a bunch of Democratic strategists who absolutely cannot handle the concept of being handed a slam-dunk win. And so they’re saying we shouldn’t talk about this stuff. We might get some of their money. And that’s a very dumb, losing strategy. And if you’re a candidate listening to this and you want to think, well, maybe I should moderate on the data center question—no, don’t. Go for it. Your electorate will love you.

With Hochul, it’s interesting because I read her statement where she talks about risks of AI, and a lot of it’s the doomsday cult framing. Now, to be clear, people like Jacob Cox and the kid who got marketed into the news a couple of weeks ago is an idiot. These people are weird, stupid cultists. And I’m not saying that lightly. At their extremes they become murderous lunatics. Not all of them are crazy people, just many of them. But it’s sort of an engine for making weird and disturbing people. And their basic idea is they believe superintelligence will destroy all of humanity. And we can’t prove that the chatbot is not the first stage of that. So this might sound really absolutely inane, but these people honestly believe that. They also love using the chatbot, by the way. So they’ve actually had a great hit with a lot of politicians. They got Bernie Sanders and Steve Bannon up on the same stage, and boy, is that an unfair forced error from Bernie. I’m sorry, Bernie—this is not a guy you should be on a stage with, you know.

BJ Mendelson: Yeah.

David Gerard: This is not a guy you should share a stage with, you know.

BJ Mendelson: I gotta say, that’s probably been my biggest disappointment with him—he seems to have fallen hook, line, and sinker for a lot of this AI doomsday stuff.

David Gerard: Yeah, it’s unfortunate. Hochul’s press release that she put out—when was that? It was a few days ago, I think.

BJ Mendelson: Yeah, it was just as we were chatting to set up this interview last week when it came out.

David Gerard: Yeah, here it is. September 21. So this was like Monday. She was requiring starting in November, New York will direct large AI developers to register with the state. So this is about safety incidents with AI—meaning we programmed our automated hacking tool to hack people, and then we enabled it to hack people, and the people we hacked are idiots. So we’re reporting that. That’s what that is.

BJ Mendelson: Yes.

David Gerard: So I found that actually really interesting because it’s like: obviously everyone hates AI. I’m gonna get on the anti-AI side because I want to get reelected. Which is the job. So I was really interested by that. One thing that really caught my eye about that press release was the regulator is part of the Department of Financial Services. Now, remember that New York is the state through which every U.S. dollar goes. It’s the New York Fed. It does all the banking arrangements for international banking. Every U.S. dollar that’s electronic goes through New York, more or less. So it’s like New York is the city the money lives in. It’s the city with the U.S.—it is the home city of the U.S. dollar, functionally.

BJ Mendelson: Right. Can you explain that a little for people?

David Gerard: The DFS is big and scary. They often put in financial regulation before the U.S. federal government does. So a lot of the SEC setup and laws were modeled on New York Blue Sky laws from the 1920s—they put them in place in the 1930s. Now Trump’s gutted them, but of course. But the DFS—it’s interesting she’s gotten the big regulator with teeth on this one: the Office of Digital Innovation, Governance, Integrity, and Trust. Digit! Isn’t that clever? So yeah, and I don’t know—they might do something. Who knows? It happens.

BJ Mendelson: You know what’s interesting, though? Originally the data center ban that came out of the New York State legislature this session was for three years. She weakened it down to one. And up until the 21st, when that press release came out, she looked like Gavin Newsom in terms of being completely in the bag with these tech companies. Do you think that she just changed directions because that’s where the wind was blowing?

David Gerard: Hochul is not the sort of person who has beliefs. She is not a politician who has beliefs. I presume she is doing it from electoral calculus. And in this thing, the electoral calculus says two plus two equals ban the data centers. So yeah, it’s pretty amazing. Even Greg Abbott in Texas—arch-Republican who never saw business in Texas he didn’t like and was very much in the pockets of the data center developers—had to walk it back with various restrictions on the data center industry. And it was interesting because everything he did is stuff that he could backtrack on in five seconds. It’s all governor decrees. His Democratic opponent says that’s weak. We need actual legislation, because that way you can’t just switch it off the day after you get reelected. And it’s quite something that a Democratic candidate in Texas actually has a chance and is hammering on the AI thing as hard as possible. Because the Texas grid doesn’t work because ERCOT is more or less rubbish. So they’re really feeling the pain. So yeah, if they can get someone like Greg Abbott to make anti-data center noises, imagine what Hochul in New York is having to do. So yeah, the AI companies are throwing as much money as they can at U.S. politics, which is what happened because it works so well for the crypto companies. But now you can just say: my opponent is funded by the AI companies, like Hochul just did, and it’s a strong, powerful message that will resonate. So, with gritted teeth, I say: well done to Governor Hochul.

BJ Mendelson: I think that is also pretty much the position of The Monroe Gazette. Whenever she does something well, we always give credit where credit is due—through gritted teeth. Let me ask you because this is something that I mean, you and I know that we’re not getting Ultron. Ultron is not coming from this version of what we keep calling artificial intelligence. And you know, they want you to think it’s AGI and it can do all these amazing things. But for people that are listening to this—we do have readers who skew a bit older. We have a lot of retirees who maybe get the news from CNN and Fox, right? And that’s not to slight them, but they’re not necessarily getting the truth about what this stuff can and can’t do. So just for the record, could you help me tell those people that Ultron is not coming?

AI Is Not Going To Kill You

David Gerard: So Ultron is not coming. The AI is not going to destroy us—not by building a superintelligence that takes over the world. That’s not a thing that happens. We don’t know how to do that. We don’t know how to define that. No one is on a path to that. The chatbot is literally a word completion machine. That’s what it does. It comes out with in-context sentences because we fed it literally the entire internet and then whatever else we can pile on. So it’s got a lot to complete from. So it’ll give you a nice, absolutely statistical average of all the writing in the world. And some of that will be stuff you haven’t seen before. So people try the chatbot and they go: “Wow, it was so useful. I love it. I’m convinced.” And they’re one-shotted by it. Other people use it and they go: “This is garbage.” And they’re not so convinced. It’s a super-duper autocomplete. It is not going to eat you and kill your dog. That is not how it works. What will happen is that the bozos running it, paying it, and trying to make all the money in the world from it do not care about you and will happily destroy your life, your town, your environment, and your industries in the process of making another buck.

But you know these guys: whenever someone says the magical robot will do all your work for you, that is the same pitch as “there’s a way to get rich for free.” In both cases, they are going to pick your pocket 100 percent of the time. And that’s what we see. The chatbot has limited use cases. People can find limited use for it. I’m deeply unconvinced by any of them. But yeah, it’s sort of a machine that uses up ghastly amounts of power and water at huge expense, which is subsidized by venture capitalists to do a mediocre job of some things. So yeah, if it destroys the world, it’ll be destroying the environment first.

BJ Mendelson: Yeah, and they don’t care. I think that’s the important point—that the tech bro oligarchs do not care about any side effects or any consequences. And I think that is so important for people to get because, you know, I get into this thing with my dad. He’s almost 80. And when we talk about AI for his generation—they don’t have some of the more contextual information that people our age and younger might have. And so the question I constantly get from my dad is: “Don’t they realize that what they’re doing is taking all your water right from your town?” And I keep saying to him: they don’t care. They only care about the bottom line financially.

David Gerard: Yep. I mean, my mum is in her 90s, and she got it immediately: “They’re just trying to take all our jobs.” Yes, that’s literally what this is for. That’s literally why they created this. That’s literally what they promote it to do. They’re trying to take everyone’s job. That’s how they sell this stuff: “You can fire all your employees. The robot will do your work for free.” It’s another instance of the robot pitch, you know. And again, it’s just picking their pockets because actually it turns out chatbots are really bad at every actual job. The only use for AI is for doing work that should not be done. You know, like transcribing meetings that nobody will ever read or summarizing meetings that nobody will ever read the summary of. If it’s ever important, you must not use the chatbot. It will get stuff so wrong and cause disasters and try to start a war with China and so on, purely by accident and stupidity.

BJ Mendelson: See, this is why I wanted to have you on, because I don’t think there’s anyone that puts it as clearly and as succinctly as that.

David Gerard: Well, Ed Zitron does, but he also swears a lot more.

BJ Mendelson: Ed does. I know Ed. We have a history. So like, I’m a little more reluctant to go to Ed. And honestly, I like your stuff better because it takes me—it takes me five minutes to watch. I watch everything on 2x speed on YouTube.

David Gerard: I must say I’m a huge fan of Ed Zitron. He does super valuable and important work. He reads all the financial documents.

BJ Mendelson: Yes.

David Gerard: I used to do that stuff with crypto, so I deeply appreciate precisely what he’s doing, you know. I’m glad he’s doing it so I don’t have to.

BJ Mendelson: Absolutely. Yeah, and let me just be clear to people listening. Like, you should read Ed Zitron’s work. My own stuff with him is my own business between me and him. But it should not affect whether you read his stuff. You should absolutely be reading his stuff. But I like your stuff a bit better because it’s more digestible, you know? It’s five minutes.

David Gerard: It helps that I do a five-minute blog, a podcast, or a few hundred words a blog. Turning that out every day is a lot of work. Did you know that a video takes about an hour per minute?

BJ Mendelson: Yes.

David Gerard: Just to edit it. That’s the old rule I learned from radio: for every hour on the air, it’s two hours worth of prep work that people don’t see. And yeah, producing a YouTube video every day—I can imagine just writing, editing, and coming up with stuff.

David Gerard: There’s one of me.

BJ Mendelson: Yes, yeah. I know the struggle well. The reason why we stopped doing video is I realized I could not write, break, and report news every day at 7 a.m. and still do all the other stuff on top of it.

David Gerard: Goodness me.

BJ Mendelson: You put out great work. And so I just want to recommend people: the Ed Zitron newsletter is very long, your stuff is digestible. And I think that’s the key difference.

David Gerard: Yeah. His stuff is very long, but you should dive in if you can, because

BJ Mendelson: Yes.

David Gerard: It’s absolutely quality stuff with a fair bit of strong language, because it warrants it. But—

BJ Mendelson: Yes. And that’s something too, like with Monroe Gazette. Like, we swear, and we get some of the older readers who don’t like it. And I’m constantly telling them: the planet is literally on fire. We have a dictator in the White House. At what point can I be allowed to swear? When is it okay to acknowledge that we’re living in this?

David Gerard: Yeah. The rule is: on YouTube, you have to indicate when you’re using strong language. On Apple Podcasts, you just gotta bleep it. In writing, AP Style says that if it’s in a quote, you can say fuck.

BJ Mendelson: Yes, that’s right.

David Gerard: Anyway, you have to bleep that for Apple if you want your clean rating, but that’s your problem.

Palmer Lucky Is Selling War Crimes In A Box

BJ Mendelson: Right, I thought it’s all good. But let me ask you: I originally wanted to talk to you about one of the things I mean, there’s always a ton of stuff—but I’ve been trying to work on this story involving Palmer Luckey and an appearance he had at West Point.

David Gerard: Yeah.

BJ Mendelson: So, for people listening to the Monroe Gazette: at this point, we’ve help break the story that Pat Ryan used to receive a lot of money from Peter Thiel’s Palantir. Recently, Ryan declared that he was going to give all that money back. But what he didn’t do—and what his office refuses to clarify—is whether or not he also gave back the $28,550 from Palmer Luckey’s Anduril Industries, who was recently, this summer, at West Point giving a really weird presentation to cadets.

And I was just hoping you might be able to, because I know you’ve come across Palmer Luckey: tell us a little bit about who he is, his connection to Peter Thiel, and what Anduril Industries is.

David Gerard: Right. So Palmer Luckey—he’s a bro. Like, he is the most broish bro who ever broed. And in fact, that’s his presentation, how he markets himself. He wears this sort of beach-bum outfit as his business uniform because if he wore a suit, you’d think he was a normal guy in a suit. But if he wears the beach-bum outfit, you think, “Wow, this guy must be special. He’s a nerd.” Same way Sam Bankman-Fried marketed himself at FTX: if he dressed in a suit, SBF wouldn’t have got nearly the money he did by dressing like a weird-looking nerd. So it’s business clothes—his beach-bum business clothes. He’s from a comfortable middle-class upbringing. His dad ran a car dealership. He got money, started his Oculus virtual-reality thing, and it was bought by Facebook. Then he got fired by Facebook for being Palmer Luckey. And he started a company called Anduril, which was going to be his defense contractor. And he got the attention of Peter Thiel. And they’re pretty obviously of like mind: Founders Fund, Thiel’s venture firm, led Anduril’s first funding round.

Basically, Luckey is a huge reactionary of a slightly disturbing kind. I saw the video of him at West Point. Well, I watched a bit of the video and I went, “I can’t stand this guy,” so I just read the transcript for the rest.

BJ Mendelson: Yes.

David Gerard: The transcript’s bad enough. He tells the cadets, “I think that Anduril is way more valuable as a company that is ideologically motivated. You know it. It’s the flame of the West.” Now, “West” is a direction, but also there’s a certain kind of guy—when he says “West,” you think that’s reactionary. You see a guy like that on Twitter—well, first of all, he’s on Twitter, but also he’s clearly got a statue icon, or maybe an anime icon, or maybe an anime-statue. He’s got a reactionary political stance. Luckey is telling you straight up who he is and what he is. It’s not obscure. He’s of like mind with Thiel. And putting money into politics—Thiel throws around money like water to causes he likes. It’s not even slightly secret. He funds a ton of podcasts and media operations. He funds all sorts of people he thinks are great, because he thinks the use of his money is to enjoy himself, and he certainly does.

But yeah, so Anduril is like—I mean, that presentation was like a commercial for Anduril. It totally was. That’s what he was there for. They were like, “Welcome, Mr. Luckey. We’re so delighted to have you. Gosh, you’re cool. We love you so much.” And it’s absolutely a commercial. It’s 100 percent an Anduril ad—always be closing. And manifestly, Luckey is a good salesman, right? He knows how to sell this stuff. His job is selling to defense. He talks at length in that video about selling to defense, you know. And the other thing is that Anduril’s stuff, their drones and so on—they don’t work. They don’t work. They keep failing in testing with the Defense Department. They’re not very good. They fail so much. But that doesn’t matter. You don’t need stuff that works for the Defense Department to buy it. That’s the origin of artificial intelligence. They spent about 30 years on the Defense Department dime on a field that was not going to give the Defense Department robot slaves. And they finally stopped giving these guys money, and that was the winter.

BJ Mendelson: Right.

David Gerard: Like, literally, that’s the history. John McCarthy invented the term “artificial intelligence” to market a summer program to the Department of Defense. It was a buzzword—a very cool buzzword with fabulous marketing potential, right? That’s a real achievement. He came up with a great marketing term. And that’s why, if you ever talk about artificial intelligence, someone says, “This technology...”—stop them right there and ask them to be specific about the technology. Otherwise—

BJ Mendelson: Yes.

David Gerard: —you get bozos who say you should absolutely adopt the lying chatbot in your business, because machine-learning systems scan X-rays. If you don’t say the name of the technology, suddenly it’s transparent that people are equivocating and selling you nonsense.

BJ Mendelson: Yeah, and it’s rare how often they say “machine learning.”

David Gerard: Image classifiers for radiology and so on—they can be good. Basically, if the radiologist is in control of them, they’re using them in their job, and they have a say in how these things are used, they help the work go along nicely, you know. They’re not super reliable. Machine learning is not reliable, but it’s good enough that it can assist people. A good system can assist people. Chatbots aren’t the same class of object. So anyone who uses X-ray analysis as an excuse for the chatbot—they might be lying, or they might just be a bit dumb.

BJ Mendelson: Good.

David Gerard: Yeah. With Anduril, it’s like—I don’t think Luckey is a military strategist. He doesn’t know it.

BJ Mendelson: Exactly. That’s why that was my whole—that’s what set me off in the first place: why is this man—

David Gerard: He doesn’t know it. He knows about as much about war as I’m sure he’s well-read on the subject and finds it interesting, but so is any war theorist on Reddit. He’s a Reddit guy about war, and now he’s saying this stuff in front of officer candidates at West Point. And they’re actually educated in this thing. They are there to learn it at West Point. They’re there to get a proper education in it. He’s got a Reddit-bro understanding. I’m sure he’s read Wikipedia articles, you know. And they’re just lapping it up because he’s a good talker. He’s a good popularizer. I mean, I’ll give him credit. He’s good at explaining simple stuff he’s read up on. Because, you know, a CEO’s job—half the job is public relations, you know?

BJ Mendelson: Yes. Absolutely.

David Gerard: Yeah, I’m sure Luckey is competent at the things he’s competent at. He’s an ass, but he’s not an incompetent one—unless you want your drones to work. In that case, he might not be your dude.

BJ Mendelson: So that’s—I mean, that ties this all together, right? Because I’ve noticed with the AI bros in particular, they seem to have an issue with China. And I don’t know about you, but as I was watching the video, he seems—and it’s not just him, it’s Peter Thiel and Musk and all of them—they seem hell-bent on having some kind of conflict with China. Did you get that impression? And do you feel that there seems to be some marketing going on there in beefing up some kind of imaginary feud with China?

David Gerard: I mean, yeah, that’s the whole deal. The great funding era for defense contracting ended in 1992 with the collapse of the Soviet Union. What do you do then? They tried starting a few more forever wars, which are always good for money, but not as good. But if China is the new super-power enemy, well, we have to arm up against them. And he’s got this sort of fantasy scenario of how China invades Taiwan. And maybe they will. Maybe they will. But if they do, it’s because they’re stupid. You know, it’s like—no one would have predicted Russia was going to invade Ukraine because it was obviously stupid, but they did it anyway. So we can’t rule out stupidity. But it would be stupid.

And then he painted this entire—I believe the phrase is “scenario planning”—where you make up a story. And he painted this whole scenario where China, one, successfully invades Taiwan with an amphibious assault, and two, proceeds to invade the rest of the South China Sea. Sure? Yeah. I’m not sure the People’s Liberation Army is quite up to that. It’s not in a great state, but if they really wanted to do that, you’d know about it, because they’d be getting the PLA really up to high quality. Which they could do. But I think China’s too busy making money.

BJ Mendelson: There was an interview on Stupid Sexy Privacy where we spent a lot of time talking about China. And forgive me for forgetting the interview name [Editor’s Note: It was the author Patrick McGee]—I will find it—but the guest had pointed out that the United States seems hell-bent on causing a war, and meanwhile, China’s kind of over there going, “We’re just making money, man. We’re just in this for the money.” Now, that doesn’t excuse or—and I’m not meant to make light of some of the more serious things going on there. But generally speaking, it certainly seems like we’re spoiling for a fight, and it’s the tech bros that are agitating for it, while China is on the other side going—

David Gerard: They’re marketing for defense dollars.

BJ Mendelson: Yes. And I think that’s key. I think that’s such a big point. And so, you know, it struck me as weird that this defense contractor is sitting there giving an ad for his company to cadets who may be in position to make requisitions with those defense dollars. That brings me to my last question: Is all of this a bubble? Like, if we’ve got these crazy scenarios, right—where we’ve got a defense contractor eagerly and enthusiastically selling his stuff to cadets—something doesn’t seem right here. Like, it seems like this business model for the AI companies, and including the weapons that use the AI, is a lot shakier than it looks. What would you have to say about that?

David Gerard: Yes, it’s absolutely a bubble.

BJ Mendelson: Great.

David Gerard: It i it was the bubblyest bubble that ever bubbled. my theory, my basic thesis of the AI bubble is that the whole thing is venture capitalists running a bubble party. And it’s literally the same actual guys, not just people very like them, but literally the some a lot of the same actual guys

BJ Mendelson: Yes.

David Gerard: Who tried to do a bubble in crypto and a bubble in the metaverse and web three and those all fell flat.

And really the crypto the AI bubble makes a lot more sense when you remember that Sam Altman is still a crypto bro.

BJ Mendelson: Yes.

David Gerard: If you think, that guy’s a crypto bro, suddenly he snaps into focus. And Palmer Lucky, I think, doesn’t give a hoot about the AI really. He’d like he he talks a lot in the video about has a lot of how we have certain kinetic weapons which we have

Have troubles with the law. What he means is that he wants to make autonomous drones to kill people and that’s what is actually called a war crime. He didn’t say war crime, but he’s talking about he really, really wants cell machinery to do war crimes. He wants to sell well Pamelahi wants to sell you war crimes in a box. And he’s

BJ Mendelson: Yes.

David Gerard: …very upset he can’t. You know. So, yeah, it’s it’s just necessary to America’s defence, you understand, to give

Palma Lucky, all your money for drones that don’t work and war crimes in a box. But yeah, I mean i he’s very frustrated by war crimes laws.

BJ Mendelson: It seemed it would seem so. And disappointingly no one no one called him out. None of the cadets they were just cheering and applauding, which was l a little disturbing.

David Gerard: It’s somewhat disconcerting, yeah. I was they they were just far too enthusiastic for his defence of the West thing. Yeah. I don’t want to see…

BJ Mendelson: Yeah, that was really disconcerting.

David Gerard: … a bunch of weird reactionaries being set up in the army, but I guess it’s a bit late really.

BJ Mendelson: We’re out of time, so is there anything I did not ask that that you wanted to share?

David Gerard: I don’t know. I mean, he’s I I’ll say one thing for Palmer Lucky, he’s not an AI doomsday believer. He thinks that’s garbage. He wants AI so that he can make his drones that commit war crimes. he wants to sell the Defence Department as many drones as possible. That’s his patriotic duty to bring about a proper reactionary, world of the sort which he believes in. The flame of the West, you know.

This guy is deeply unwholesome and I don’t like him.

Episode Outro

Today’s episode of Privacy Tips was recorded in Hollywood, California and is sponsored by DuckDuckGo.

It was written, produced, and hosted by me. B.J. Mendelson.

Before we go, I want to give a shout out to the production team at Stupid Sexy Privacy which includes Andrew VanVoorhis, Rosie Tran, and Amanda King.

And, of course, our friends at DuckDuckGo.

If you enjoyed today’s show, we hope you’ll take a moment to leave a review on Spotify, Apple Podcasts, or wherever you may be listening.

This won’t take more than two minutes of your time.

You see, we have this crazy goal, and we need your help to achieve it.

We want 5% of Americans to be 1% better at protecting themselves from fascists and weirdos.

Leaving us a review can help make that happen.

That’s because your reviews will help other people find this show.

So, please take a moment to leave us a review, and we'll see you next Thursday on Privacy Tips.