Hacked - How Much of Facebook’s Revenue Is Scam Ads—And Other Chat Worthy Questions

Episode Date: November 16, 2025

It’s a chatty chat. I repeat, this one's a chatty chat. Today we’re digging into the big weird questions on our desk: what percentage of Meta’s revenue allegedly comes from knowingly running scam ads, what exactly recently pardoned Binance founder Changpeng Zhao did and why he walked free, and what happens when AWS—the concrete pad foundation of the modern internet—goes down. If long-form nerd talk isn’t your vibe, totally fine—we’ll catch you next time. For everyone else, here’s some good good chattin’ for your dishes, your commute, or whatever you’re up to right this second. Learn more about your ad choices. Visit podcastchoices.com/adchoices

Rapid fire questions. What percentage of Meda's revenue is knowingly running ads for scams? Damn. I'm going to say knowingly. Knowingly. Got to be 30%. Whoa. Oh.

You guessed higher than I did. What exactly did the recently pardoned founder of finance Changpeng Xiao do? And why was he pardoned? Money laundering and pardoned. because he money laundered for the right people who have power that gets you pardons? What happens when AWS, the concrete pad foundation on which the house of the internet is built, goes down? Well, you kind of set it in your question. It goes down. It goes down. We have questions. So, Scott, we have to chatty chat.

We have to chat. We've got some in-depth interviews in the hopper for upcoming episodes. But today, we just have like a big old pile of strange text stories on the stuff. side of our desks that we got to get through. You ready? We're just catching up. We're just catching up with the times. Catching up with the times. Let's get into it on this episode of Hacked. How you doing, Jordan? I'm doing good, man. How are you? I am good. Busy, good, happy. Things are good. Filled. Fulfilled. Fulfilled. Spiritually enlightened. Active. Active. All of the checkboxes being checked. We're truly living our best lives, Scott. And sharing it with you, the audience.

So like I said in the intro, we have some fun stuff coming up in upcoming episodes. These things take time together. We have a lot of stories, though, that we haven't been able to get to over the last little while. So that's what we're here to do today. It's a good old-fashioned chatty chat. We're just going to catch up. We're going to gab about what's going on in the world's technology and cybercrime and weird tech stuff. If a couple of nerds talking long form isn't your vibe, we'll catch you in those later ones.

but for everyone else, we're just going to make sure you got some good, good stuff to, I don't know, wash dishes to or commute or whatever it is you're doing this very second you're listening to this. One thing I want to touch on is a couple episodes ago we were talking about like open AIs, AMD investment and kind of like the round. Oh, we're right into it. The circle. The soft edges of like, I don't want to call it a Ponzi scheme, but like, you know, some kind of, market manipulating investment stuff. It's the snake eating its own tail, the Owaroboroboros of investments.

I hear you. Yeah, I remember talking about that. So we talked about that. That was what, like a month ago? Yeah, about a month ago. Anything happened since that? Yeah, I feel like a week ago, the entire internet caught on to the fact that this was going on

and like so many big YouTubers and other articles started coming out about like, is Open AI a Ponzi scheme? Question mark. And like there's a bunch of things to talk about in the open AI space. I know that they've made a bunch of, we'll call them market gaffs as of late. You know, they claim that they needed the federal government in the United States to backstop data center development, which then kind of indicated that maybe they didn't have the $1.4 trillion that they've committed to invest and spend, that maybe they didn't actually have that much money and kind of stock market got rocked by it.

But yeah, I've got. It was just really interesting because I felt like we talked about it like a month ago. And then it's just like all of a sudden my whole timeline was like people talking about it like, you know, three weeks later. And I was like, ah, we were ahead. We scooped it. We scooped it. Yeah, it definitely felt like there was a lot of video essayists. God bless them.

Just sort of digging into that shape of the investment. I don't want to call it a funnel. The funnel. What is happening with companies investing into other companies, investing into other companies that are investing into the, first one in that loop, so to speak. Cyclical. Cyclical investment cycles. Not cyclical.

This is not a financial. Remember that. Subsequently, if you had been listening about it being a financial, AMD had a massive announcement that they think that the market for data centers is about to just explode, which I feel like everybody kind of knew, but nobody knew that AMD was going to be a part of it. So AMD's shares have had quite the lift. I've been buying, I was at Costco.

As you do. Sure, everyone's familiar with Costco. I don't need to, it's always funny. Like, what do you explain in a show like this? I was at Costco, and I noticed they sell gold bars at Costco. And I was like, do I just, do I just bail on the entire thing and become a weird guy with a duffel bag full of gold bars in the backyard I don't have? Still thinking about it. Still thinking about it.

So funny enough, so they've been selling those gold bars at Costco for, what, three to four years? years? I don't know. No, no. So they have. So a few years ago was one of my friends 40th. And we were trying to plan this like kind of elaborate boys golf trip weekend away, like go somewhere nice for his 40th. But he couldn't make it work. He had work conflicts. So we ended up having to scrap the whole plan. So we were just like, well, why don't we just take all the money and buy him like a gold bar as like a joke for his 40th birthday? And they were $2,200 at the time. And now they're, like this is Canadian like 2200 Canadian at the time and now I think they're like over 6,000 so we should have done that because then that would have paid for

this year's gold going up the things I don't know and hence the things I don't know okay this first thing that I want to talk about this episode we we teased it in the intro is meta and your big swing guess on what percentage of their revenue they knew

was coming from ads for fraud yes Do you want me to justify my big swing? You went big. It's like when someone asks, you were like, how much did you think? And they go beyond what it actually was and it deflates the whole thing. That did happen. I rooted it for you.

And the reason I took a big swing is because my baseline of like that they know is fraudulent means that they get moderated afterwards. So they take the initial cash. So my 30% is an end. estimate of if all ads put onto the meta platform, how many of them get removed for being fraudulent? Because meta, not denying fraudulent ads from going up is to me essentially allowing them because they're still monetizing from them for the first little while. Yeah. So this is really interesting. There's some really cool reporting on this, and it's somewhere in the middle. We're talking about this pile of internal meta documents that Reuters dug into

that showed that the company knowingly earned as much as $16 billion a year from scam ads, showing users 15 billion scam ads a day, and sometimes charging the scammers more money to run the ad than simply banning them. While ignoring, I think the number is 96% of scam reports and planning only very slow-tempered reductions into the future to protect that revenue stream. So I have a knock on to this that drives me nuts. Where in free-to-play video games, we're going back to hacking and gaming. But they don't really want a good anti-cheat because they want to talk about daily active users.

And most cheaters have hundreds of accounts that they cycle through. So it inflates their daily active user goal. So when they go to share meetings, they're like, yeah, look at this game we make. It makes billions of dollars. And like the daily active users are X. And they actually don't want to remove cheaters. I have a theory on this that they don't actually want to remove cheaters because it will hurt their share price. It will ruin their game.

But they'll take the long-term pain for the short-term gain of having a really high DAU number. We don't want to kick the person selling drugs in the club out of the club because then the club will be less business. and people won't have easier access to drugs. It's like... See, and Meta probably doesn't want to turn off the $16 billion in revenue they're making. No, and according to this reporting, the links they've gone to to avoid turning it off are quite fascinating. No surprise. Small upstart journalism enterprise called Reuters,

uh, obtained a tranche of internal meta documents from 2021 to 2025. And these documents showed that meta has been knowing, learning, you know, billions of dollars from ads that are four things that are scams. illegal or banned under its own rules. The internal projection as of late 2024 was that about 10% of all meta revenue, not the ad but all meta revenue was coming from these scam ads, $16 billion, $15 billion scam ads every single day. That is importantly only paid ads.

We're not talking about organic scam activity, things like marketplace or DMs or groups. Those aren't being monetized and those I think we should think of them as a sort of a different phenomenon. But 10% of all revenue, according to this documents reported by Reuters for a fang company, like one of the biggest companies in the United States. 10% of gross rev is a huge amount. Like a chunk of the economy. Like that's a really baffling amount of money. But like if you think about that as the impact of the share price, like 10% of gross rev is going to hit the net income line or the EBITA line, the earnings line, that share price,

are valued based off of substantially because the overheads are still there. So you're essentially taking 16 billion clean out of it. Just gone. Yeah, massive impact to share price, max of impact. So it's like people, I'm sure, flirt with this all the time, just like they do in video games. It's like, how can we keep the community happening while still keeping the investment community happy? It's tough. According to the files, meta system, like they use a system that just has a likelihood of

fraud slider somewhere inside of it. And it's only when they hit this like 95% threshold that they will ban an ad. If meta, there's like a window up until that point where if they're less certain, but there's still something a little iffy going on here, meta will raise the price of ads. The internal name for this is called a penalty bid, which means that the effect, according to this reporting, is that meta charge scammers, suspected scammers more, not less, increasing that profit, which would, as you have kind of brought up, make them more money and would be good for the shareholders, but would not be protecting the users of their platforms.

And I will know, too, that this isn't just a meta problem. No. Anybody that watches YouTube. 100%. The amount of deep fake ads that I get served now is, and like not deep fakes of like a certain celebrity, it's like national leaders, like the president, the prime minister of Canada, like I get deepfakes of all of those. Somehow those get through the threshold filter. Like you would suspect that AI would capture that and be like, no, this is clearly trying to be

Donald Trump. Like, get out of here. Yeah. But they don't. No, no. No. To their credit, trying to figure out whether or not it is Donald Trump hawking a cryptocurrency or a deep fake of Donald Trump hocking a cryptocurrency is a non-trivial task.

We're talking about fake e-commerce stuff, counterfeit brands, fake investments, crypto schemes, illegal casinos, banned medical products. At the heart of this is like meta's like the internally used risk calculus. And the documents kind of show a bit of a deliberate strategy here. The line was do not reduce scam revenue unless regulation forces us. the plan to do that as there has been more legal push to regulate scam ads on these platform is to be very slowly reducing that revenue over a period of years. They have a, there's a chart, allegedly, according to the reporting, I just keep saying

that 10.1% in 2024, which is kind of like the current alleged number, down to 7.3 at the end of this year, down to 6, down to 5.8. They're like, there's sort of a whittling process. so it's not to just turn the $16 billion spig it off all at once. And I would imagine hope that some, I don't know, meta-ray bands offset that loss in revenue. What's meta going to do here? Like, seeing as this is a straight chatty-chat.

Let's talk about it. Sure. Because like they bought Instagram, great buy by them. Exceptional. They paid a billion dollars for what is probably now a trillion dollar company. Worth every penny. Yeah. With every penny.

Yeah. WhatsApp. Yeah. They bought. Good, good spend. Yep. I feel like they haven't bought. really anything since.

I think they just bought some 28-year-old kid who's going to lead their AI. Yeah, yeah, yeah, yeah, yeah, look shocked. I read somewhere that his comp package is something like $14 billion. What? And he's being brought in to run meta-a-I. I don't know how to Google 14 billion-year-old teenager for sale. It's something about they bought the kid. I'm like, I don't know how to Facebook.

Yeah, I think it has to do with. them taking a massive stake in scale AI, but really what they were mostly interested in is Alexander Wang, the founder. So you remember we talked about Johnny Ives company getting bought by Open AI and like a, what was that, $6 billion? Yep. This is one of those. And then we got one of those going on in the meta side.

Just pay $14 billion for one person. Wow. Yeah. Tell me, tell me it's not a bubble by telling me it's a bubble. Yeah, $14 billion. That's wild. Um, and maybe they can use all that AI to, uh, the last staff that was in here that I was intrigued by was that meta platforms are tied to allegedly one third of all successful

scams in the United States. Wow. UK regulators pegged it at 54% of payments related scam lit losses related in some way to a meta platform. Anything about like WhatsApp, Instagram DMs. Totally. All this stuff makes sense. That's not included inside of this knowingly ran the ads, but it is just, it's

an interesting insight into like, what? What are the surfaces on which people lie on the internet? It's a fascinating thing's reporting. Someone owns them. And his name is Mark. Yeah. Well, for every, every, like, text message scam I get, you know, hey, can I send you the job offer?

Hey, you're still awake or whatever. You know, you get so many of them. For every one that I get on that platform, I get at least 10 more across every other platform, whether it's an Instagram, DM, whether it's a random, what's app message, whether it's, one of the things that I have been finding is like when you travel internationally, everybody just uses WhatsApp. Like you check into a hotel and the concierge is like, hey, connect on WhatsApp. You need something. Hit me up. And I don't trust it because I know so many people that when they get back from vacations get scammed over WhatsApp. Yeah. The amount of

spam that comes in over WhatsApp is crazy. But it is like it's the, it is. Especially if you have a group of people that are like cross platform between say Android and iOS. It is like the most reliable like one of the most reliable messaging apps, reliable in the sense that people reliably already have it on their phone. There's other options. Text messaging is getting better, but everyone has WhatsApp. It's where the conversations already are. I've never, I've never figured this out because they paid what like, can't remember what they paid for WhatsApp, 600 million, $460 million, something like that. I'd have to look it up,

but it's not consequential to this conversation. The thing for me is, what is the monetization? Like, I get that they have the audience. But as somebody who's been a WhatsApp user for 10 years or whatever it's been, I have no idea how they make money from it. I don't get served ads in it.

I don't get... It used to be encrypted end-end, so they didn't even get the data of what's going on in the conversations. So to me, it's always been an interesting one. Like, it's good to own more of the social space, I guess. Sure. And that could be an argument, but actually from a monetization perspective,

you know, what are they doing to monetize it? Yeah, it's interesting. If WhatsApp wasn't encrypted end-to-end, I would say it's like a massive trunch of data for, yeah, I was going to say it's a ton of information to be able to potentially train on or sell off to marketers and stuff. Yeah, it's an interesting question. I don't really know. I don't know why you would buy WhatsApp.

but we all use it, but there's no ads in it. It's weird. The thing that I liked about this story was that I'm, so I'm reading this Reuters reporting on the meta scam ad ratios and these baffling numbers. And then I, I go over and I go back to my good old Canadian news and I immediately find a piece of Toronto Star reporting about how they were able to purchase hard drugs off of Facebook ads.

Like it was just this one two punch of like 10% of the ads on Facebook are scams and Facebook knows about it allegedly. And then Toronto Star Report is being like, let me explain to you how I bought cocaine off of Facebook. And they did. Oxacodone cocaine. They bought them from websites that they found through meta advertising, did an intrac e-transfer and got mailed hard drugs through Canada Post. My favorite part of this story was that the drugs arrived. You can tell there was some like, it almost had drop shipping energy to it because the, the, the, the, the, the, drugs arrived in a bag plainly labeled regular Coke. Not the good Coke or the bad Coke.

Regular. Yeah, no, totally. Executive Coke. No, this is regular. This is Junior Sweet Coke. Um, the start found like 20 different websites that just this one reporter was able to find advertising on Facebook since March of this past year. Kept kind of going for months.

Star was flagging the sites after they confirmed that they were real. and meta was removing some of the ads. Others remained live at the kind of at the time that they reported this story. Yeah, sponsor posts for cocaine. It was an interesting kind of follow-up story to the scam revenue one. I'd be intrigued to have somebody on from like a National Police Service for an interview about this, because I just, given how overrun the Justice Department is with so many things going on in society, I've noticed that like getting, like there are like where Jordan lives in Vancouver,

you can walk into a store that is blatantly advertising that is the selling illegal narcotics and buy them and just walk out. And they might get like a bylaw fine for every day that they're open, but they easily make that up in the sales that they make. And the online websites where they're just bypassing like essentially the tax system because you're paying with interact transfers or you just, essentially venmoing people money. They're not tracked.

It's, I imagine there's, and who's policing it, and are they policing it? Like the amount of illegal stuff you can buy from nicotine to magic mushrooms, psilocybin, marijuana, which isn't illegal in Canada anymore. Neither's nicotine. Yeah. Well, well, it is, it's regulated. Regulated, sure. And so many of the things that you can buy online are unregulated versions. Same thing goes with marijuana.

It's also regular, or cannabis, it's also regulated. And the product that you're buying online is often bypasses the regulatory system and is therefore illegal. But like, doesn't seem like there's a big crackdown on it. Okay, are you ready for a real pivot, but like a solid segue? Yeah. I'm pretty proud of this segue. Okay. Speaking of cocaine, actor Michael Cain has signed a deal licensing his voice to AI company 11 last.

And Matthew McConaughey. And Matthew McConaughey. This is a great story. To be clear, I'm not implying Michael Cain has done cocaine. I'm merely referencing the meme that the easiest way to jumpstart a Michael Cain impersonation is to say my cocaine. Which is great because this is ultimately a story about impersonating Michael Cain. I give you yours. Thank you.

I give you. I need my roses. This is exceptional. Thank you. You brought me this story, so why don't you explain it? So I sent this story to Jordan literally last night as I was laying in bed, reading the news before I passed out, same as I do every night. And it had broken, I think, yesterday, late afternoon.

So here's the big twist, is that Matthew McConaughey is an infestor in 11 Labs. So 11 Labs is one of the first big AI audio voice companies. So you can upload a book, pick a voice that you want, and it will generate you like the audio outputs of somebody reading them in a prompted voice. So you can be like, I want a Dutch woman with a, but speaking English, etc., etc. And like Jordan and I both use this tool and it is very good. Yeah. I mean, a couple things.

People have for sure heard the outputs of this tool on like TikTok. A billion percent. A lot of the times when you hear that kind of, because there's two different ways you can use it. We'll talk about it. But when you hear that kind of like video that starts with guy talking like this and you're like the rhythm of him talking isn't quite right. But the like the sound of the voices is accurate. A lot of the times that is tools like this, if not explicitly 11 labs. Like I will recognize voices from 11 labs out in the world sometimes.

And that is as you said, people feeding text into the system and having a generate audio. The other tool that other use case you'll hear sometimes is people feeding spoken. spoken word like audio of a person talking in and you can re-skin it. So it has different features that it can use, but it is sort of the most popular AI voice generation platform. It does also have the ability to upload. I think you need a minimum of five minutes of clean signal of someone's voice and it will actually make a fingerprint of that voice that you can then apply to text copy or previously recorded audio copy. And that must be what they've done here. So Matthew McConaughey and Michael Cain have both signed a deal licensing their voices to 11 labs.

I haven't logged into the system to see if we can generate something with Matthew McConaughey's voice yet, whether it's currently in the system. But yeah, big, big shift. So the knock on to this is I sent this to you at like 11 o'clock last night. And in my morning version of my evening thing where I sit and check up on the news stories that happened. Uh, yeah. There's already massive fan outrage that people are, are very, very angry about this. Shocker. No shock.

No. But anyway, it was almost instantaneous. It was just like big news, Michael Kane and Matthew McConaughey. And then literally the next article was just like the people are speaking and they do not like this. Before digging into all of the what does it mean of it all, it is just sort of funny to me. that these two people, because it's like all the headlines just have Michael Cain's face and Matthew McConaughey's faces. Michael, Matthew McConaughey is, as you said, an investor in this company.

That seems very relevant to all of this. Um, but it is the fact that these two actors are maybe some of the most impersonated people on earth. Like to have their voices be the sort of like tip of the spear in this whole thing of celebrity voices can now be deep faked sort of with consent of the actor. it is fascinating that they're the two people that most people can do an impersonation of. Yes. All right, all right, all right.

And my cocaine are like memes. Like it's like they're, it's a joke. So it's really funny that it's the two of them. It would be like, I guess them and Morgan Freeman maybe would be the third one that you might expect to see. There are other people in the, um, they have a name for it. It's their like iconic voices. They have a like a short list of famous people essentially. Um, and there's.

some, I would say, yeah, like more cursed stuff in here, just given that the people weren't alive. Babe Ruth. When this was created. Yeah, Robert Oppenheimer. So that hits a little differently, I would say, than like, if Michael Kane wants to get that bag, that's, you can think what you want of that, but it is different than Babe Ruth's ancestors deciding that they will get that bag.

It's like that that is in some material way different. So when I originally read the original launch story, it didn't really say anything. It just said that their voices were being added to the system. Yeah. To which I thought like, okay, all TikTok brain rot is just now going to be Matthew McConaughey. Oh, God. It sure is. But it turns out that they have a request system.

So you actually have to request to use it and provide what you're going to do with it. And then it will be approved or disapproved. I'm not sure if the artist themselves has the ability to approve it, I doubt that they want that much headache because so many people are going to request it. Yeah, I have to think it's not Michael Kane getting a WhatsApp message being like, hey, do you want to be in a TikTok explaining how trains work or whatever it is? I have to think it's there's some intermediary that Matthew McConaughey isn't just getting lit up with like supplement ad requests.

Totally. Left, right and center. And I also imagine that there will be a surcharge. like this will not be a free service. So to use one of these prestigious digital legend voices or something from the marketplace will come with a supplementary cost, which I imagine will be pretty high. Yeah, and I would imagine that would prevent them. It would be unreasonable of me to expect Michael Kane to be deeply tapped into the use of

AI in meme culture in short form vertical video. But I would imagine he as a representative, some kind of manager, someone who is responsible, responsible, an agent who is responsible for that kind of thing. And that creating some sort of a filtration system, some sort of like, even minute cost to prevent like people, because a lot of the time, some of these things are being produced by like an automated agent system of like grab topic, create script, output audio, generate visuals for it, upload directly. And that even you would need some kind of expense, some kind of little box that has to get checked that costs a little bit of money to penalize everyone just making Michael Cain slop.

Like, I would hope that there is some sort of filtration system against a bottomless deluge of Alfred Slop. He played Alfred and Batman. He did. Yeah, I would think. Or the bag was just so significant that they just simply don't care. Could be. Could also be that.

That AI money. That just crazy amount of AI money. A lot of the coverage on this cited that there are other celebrities, Leonardo DiCaprio, Ashton, Cush, like, there's other celebrities that have invested in AI companies, and I think we're starting to see a path towards an answer to what do you do with all of this money that you have gathered when the technology reaches. It's like 11 labs is, it could get a little less gravelly. The quality could go up a little bit. There's room for improvement, but it, it functions. What do you do with that money once your product has reached maturity? And it's like,

Why don't you just, like, fireblast it at people with big kind of cultural relevance to get their buy-in to manufacture that sort of, like, societal consent seems to be the answer. I don't know if we've ever mentioned this on the show, but it would have been over a year ago. Oh, yeah. Jordan and I received a very strange email from a, from a company. We never have talked about this. Now, from a company that I won't name, not even that I remember the name of them. But they were essentially doing the same thing as what 11 Labs is doing. They were building personal assistance and like building a compendium of voices so that people like Apple could get them for Siri.

And they were, they were like launching a podcast thing. But really about two minutes into the meeting, it became very evident that really what they wanted is licenses to podcasters' voices. And so being from the marketing advertising licensing space, we started asking very serious pointed questions about licensing rights, you know, approvals, finances, like what kind of fee they were planning on paying, whether it was revocable, et cetera, et cetera. And they got very uncomfortable, I would say, and said that they would send us an email with all the follow-up.

Really at this point, Jordan and I had been recording the call because we were going to make it into an episode because we thought it was hilarious. And it got really serious. And they never got back to us. They ghosted us after that. Because I think we asked the questions that they were hoping we wouldn't ask.

And I think that they were just more excited to have Jordan's beautiful, silky voice imprinted into AI for the rest of MET, like life. Yeah. It struck me that they were finding people who had, and I buy volume large amounts of recorded voice that were available and we're hoping that you know, content creators don't know what data is worth to AI companies.

And we're hoping they could throw a really small amount of money, if I can guess. If any money at all. To be generous. If any money at all at people to try and, you know, see how many hundreds of hours of content we can get from you. And then we'll take your voice. And then license it to other AI companies that need voices. Forever and ever amen in perpetuity throughout the universe.

Correct. Anyway, so we had that odd incident, like probably a year, year and a half ago. We never got a follow up from them. We still have the original meeting recorded. We never made an episode out of it, but thought it would be interesting. And I guess this is our roundabout way of talking about it. Yeah, it finally could love. I live in Labs is now doing this. They're now licensing known voices for people to use.

So what we could do would be take that recording and for our own protection, anonymize them by filtering their voice. So it sounds like Michael Kane. Like we could voice mask that. Like, you know how it used to be you'd pitch the person's voice down in the interview to make them be anonymous? We could have it be a conversation between us and Matthew McConaughey. and us being like, so you're not going to pay us anything for that? And you could listen to the whole thing, but without us exposing ourselves to legal liability. But I feel like we also just described it all. So there's nothing too exciting for us.

There's not much left. There's not much there. Yeah. I love it. I love it. Thank you. The internet going down. So this was like a month ago.

Yeah. It's not a news story. But we just had to talk about it because it's funny. Not funny. sad, scary, but also funny. Telling a forewarning an omen. So Amazon went down. AWS. One set of their servers crashed

was taken down. Funny enough, it was caused by two threads in a system trying to write to a DNS entry at the same time, which just left the DNS entry blank, apparently. And boom, AWS, a big chunk of it goes down. A big chunk of the internet goes down. It was kind of like a blackout day on the internet

where I can't remember every service that was impacted, but it was substantial. Yeah, all because they didn't put like a mutex in the code to protect multi-threaded applications from trying to write the same piece of data over and over at the same time. And pretty basic thing to do if you write multi-threaded code, and they didn't do it. And then it took down a big chunk of the Internet.

So I imagine they fixed that since then. For anyone that doesn't know, AWS Amazon Web Services is like, Amazon's cloud computing platform, it is in addition to e-commerce, like the, I don't know if it's the golden egg or the golden goose, it is a money printer. It basically lets people in companies rent computing power, storage, databases, all the back end of their websites, they can rent it from Amazon and then it's distributed around there.

There's a bunch of benefits to it. It's like sort of infrastructure on demand. It's the cloud. It's the cloud. Netflix, Twitch, Disney Plus, Etsy, Shopify, Slack, Reddit. Like, just name a website. Our service. Or service.

It is why I described it in the intro as the sort of concrete foundation on which the house of the internet is built. It is essential. Help me understand this error. Because you described it as like a, it's not simple, but there's the sense of like, oh, that's what took it down. Like, help me understand that. Sure. So you've got two systems looking and reading and writing records in the domain system at the same time. two of these systems at the exact same moment

try to write to the same record. Something in programming called a mutex, a mutual exclusion, which is essentially a lock. If somebody gets to something first and says, I'm going to edit this, they lock it so that nobody else can edit it. And then they make the edit, they take the lock off and they go about their business.

The next system... You don't get conflicts. Yes, the next system's waiting for the lock to go away. The lock gets popped. They go in, they make their edit. They lock it, they lock it, they make their edit, they pop the lock and they go away. Very, very basic.

Yeah. Like if you read any kind of multi-threaded programming, muttexts are essential. The, that didn't occur here. So they didn't. They didn't. So two systems were trying to write to the same DNS record and essentially overran each other leading to it just being left blank to the best of my knowledge, which then caused the massive outage.

So when I read this, I was like, okay, like Amazon AWS was the original foundation of the cloud, like all the internet services. AWS was first to market largely, very dependent on, very integral and integrated into lots of massive Forbes Fortune 500 companies. Like they've just been, they were the gold standard forever. Microsoft Azure. There's a bunch of other platforms now that have come out in the tail end of them. So when they had their big outage, I was like, we're going back to the hacked stock market show. I was like, okay, now that they've shown that they're vulnerable to outages and what the cost of those outages are, every major company currently has a bunch of IT people writing up risk assessments and are looking to be like, how do we hedge it?

So that if one of our cloud providers goes down, we don't go down. Right. So I assumed a lot of big companies are about to start a mitigation strategy of hedging across multiple cloud providers so that if one does go down, like in AWS, they fall back to Microsoft Azure or something like that. And then within a few days, Microsoft had a massive Azure outage, almost the exact same. It wasn't the same cause, but essentially Microsoft services across Azure started to go down. and they went down for about two hours. And I was like, oh my God, is anybody safe from this? It's funny.

It's like my sense of AWS and probably Azure to an extent is like that they are, they're trying to be the one stop shop for cloud computing. It's like if you can think of a tacked on thing that you might have to go to a third party for, they've brought that into their feature set in this idea that the one feature they would never bring in is redundancy that kicks you over to one of of their competitors. It's like the one thing they would never want to bring in house is the ability to go like, well, if we mess up, it'll go over here.

It's like you, I just, I can't imagine they would ever intentionally funnel traffic over to Microsoft and vice versa. In spite of the fact that that seems like it would be extremely useful in a situation like this is some kind of turnkey solution for like, and if you're on AWS and you subscribe to like a little bit of a redundancy over here with Adjury, in the event of something, it gets boop like flipped over almost like a train track switching system. Yeah, it goes, we'll fall back onto Google. Exactly.

Or like Microsoft. So the, and that's especially relevant if you're something like an airline. Jesus. Yeah. So like a lot of these cloud services, they're, they're not just like websites and online stuff. They power everything now. So booking systems, you know, I'm sure like the FAA uses them for like flight

scheduling systems, et cetera. Like they're everywhere. So essential services throughout the country. So Alaskan Airlines actually did go down when the Microsoft to zero outage happened. So I think the first thing that jumped into my mind, like if I was the CIA of these companies, Monday morning, I'm like, okay, what is our mitigation strategy? Interesting. And then Microsoft goes down.

So you're like, okay, like, is our mitigation strategy going to be planned for at least two of these things going down? So it's like, okay, maybe we need some AWS, some Azure, some Google Cloud, some big, big news when it happened, obviously. Ignoring like Amazon obviously was effective. Amazon is built on AWS. It was affected by the Shopify, Instacart, Slack, Discord, Zoom, Twitch, Reddit, Pinterest, LinkedIn, Netflix, Disney. I'm not just listing people that are part of AWS. I'm listing people that have, according to reporting, been affected by this outage. UPS, FedEx, there was an estimate by a analytics firm that pegged it. It's like the sort of losses during this window as being like kind of creeping up towards a half a billion dollars, just like a power outage at this scale essentially of like internet infrastructure potentially being that high. It's a fascinating story.

And it's interesting that these kinds of hundreds of million dollar implying outages just are sort of happening all the time. It was like, like you said, AWS goes down, and then Azure does shortly after. It's just sort of a feature at this point. It's so interesting. Well, the Microsoft one was due to a global configuration change, apparently, that applied to all of the containers, like all of the client clouds, which then reminded me of the crowd strike incident, because that was kind of what happened with CrowdStrike, which again took down massive amounts. of, you know, dependent systems, airports, flight things, etc., banks. So, so, yeah, it's, you know, we are so reliant on technology now that any of these

small things just caused such a massive impact that I think that, I think that there's going to be no way for big companies and big services and government departments to get away from looking at a hedged, leveraged, like, usage of clouds. there's no longer going to be a single provider. There's probably a massive opportunity right now for somebody to come out with a product that essentially is a cloud that sits above the clouds and has redundancies.

It moves your stuff over to the next one, sure. It's like if you're part of hacked cloud, we use all the clouds underneath. And if one of them goes down, the service will never be disrupted because we'll just fail over to one of the other clouds. It is the kind of thing, though, that it's like, and then what happens when hacked cloud goes down?

Like, it's you've, you've. it's all over. That it's truly all over. We've truly lost. I looked this up. It looks like there was an AWS outage in 2021. That was about eight hours. This one was about 15, so like almost double.

So it looks like every two to three years, we get sort of just a, a rough doubling. Because the previous one to that was, again, I think four years prior and it was four hours. So we literally every three to four years, it just doubles in duration. It seems like if we go back. So the next one, we should expect it to be in 2020. 29 and it'll be 30 hours long, which honestly speaks to the redundancy of how quickly they were able to get this thing back up and running. I don't have a transition for this, so I'm just going to say it.

Another big piece of do is that broke in the last couple of months that we really haven't talked about on the show or off the show was all of the sports gambling crackdowns that the FBI's done. And the reason I bring this up, but it really has nothing to do with hacked, except for there was one piece of it that is kind of interesting. So some of the people that were indicted in this crackdown on sports betting, which largely was related to like NBA players, prop bets, like situational bets inside games. So people were kind of rigging it and their friends were betting, et cetera, et cetera. Even rolled all the way up to Portland Trailblazers head coach has been allegedly charged with throwing games.

So essentially putting out fielding like a B squad instead of an A squad so that he can like decide. the game outcomes better and having that bet on. But the same person, Chauncey Billups, Hall of Famer, and a former NBA player, Damon Jones, were also arrested,

allegedly involved in mafia poker games. So they were bringing in a lot of these high-earning, you know, stars, people with lots of money, and bringing them into poker games, high-stakes poker games, that they were then using

hacked card counters, Like, if you've ever been to a casino, there's, like, shuffling machines. Yeah. Yeah. They had figured out how to tap into the cameras inside of them, and we're literally getting, like, full insight into what the card ordering was. So they knew what cards everybody had in the back of the room. Wait. Yes.

But, but, but, uh. The organized criminals had passed into the, the mafia, the mob. had hacked into the card count, the shuffling machines at like real casinos. These aren't underground card games. No, no. These are underground card games. Okay. So there's underground card games.

Should I get the shape of this in my head? Yeah, yeah. And the shufflers and organized criminals hacked into the shuffling machines. Correct. And at what point do athletes get involved? Why are athletes involved? I think, I think. So there was a big crackdown from the FBI.

on athletes around sports gambling. Okay. But then a few of the people that they were cracking down on for like throwing games, like sports, sports betting has long since compromised the purity of sports, right?

Okay. Like the second you can gamble on something, rigging games, you know, there's been referees and umpires that have been convicted of this. There's been players that get convicted of it, coaches. and usually it's at lower leagues. Like the worst thing you can do is open a betting app

and start betting on like div 3 football in the Europe because it's a good chance that like somebody who's, you know, making $45,000 a year to be the coach of a soccer team in Venezuela or something has a higher likelihood to be persuaded to rig the game. Got it. Then somebody who's, you know, making $250 million a year like Cristiano Ronaldo. So this one was a big deal because it actually impacted real leagues. Like the NBA was a large makeup of the people that got allegedly charged or charged with alleged crimes.

But the big, the hacked kick on this for me was the fact that Chauncey Billups and Damon Jones, both won an active coach in the NBA and won a former player, were essentially bringing people into mafia run high stakes poker games. And those games were rigged because they had hacked the deckmate two automatic shufflers by jacking into the USB service ports on them. Okay, I got it. Accessing the cameras. And so they would essentially know the order of the cards would be texted or sent to, like, their phones. So they would just know what everybody had. Oh, so it wasn't.

Okay, so that was the last piece of this that was missing. The games are run by the mob. Correct. And the mob has hacked their own shuffling machines. Correct. And they're not just using that to presumably bias towards the house, which is them, but also towards this small selection of certain players, allegedly these athletes. Yes.

So in poker, there really is into house. No, no, so kind of. You kind of got it wrong. So the house gets a rake usually in high stakes poker. So poker is like everybody at the table versus everybody at the table. The house doesn't really play. The house just deals. But the house takes a rake on the pot size usually if it's like a real casino.

They were bringing in high net worth individuals, pro sports players, people that have a lot of disposable income into these high stakes poker games. and then they were essentially fleecing them because they already knew what the cards were. So they knew what you had. So when you're betting and raising and calling and playing your poker game, they're seeing right through you. They already know what you have down.

So I did see some other coverage about this that they might have been using like glasses to show them literally what everybody had. So these deckmates would send a list of the cards that came out. That would be transposed across the, deal and they would actually know what everybody had at the same time. But yeah, essentially the mob was paying these guys a kicker to bring people into these rigged poker games so that they could fleece them and take millions of dollars off of them. Hacked shuffling machines, huh? I was sure,

I couldn't, yeah, hacked. I couldn't tell. I was like, I was trying to figure out if it was that they had hacked someone else's shuffling machine, but this makes much more sense. It was their own game. and the only way you can extract money out of other players is if you have a player is to beat them and that you have a player that is essentially on your side. You need that alignment between at least one of the players and the people who are compromising the shuffling in order for this thing to kick off any money.

You need a winner that's going to be in on it, essentially. But yeah, big, big deal. FBI crack down on it, a bunch of arrests. Big, big catastrophic shake through the entire gambling in sports industry because I think that they feel that there's going to be some changes regulatory side coming.

The only thing that's really associated with cyber crime, the rest of it's just human crime. The only thing that was associated was this piece, this mafia-run poker games with essentially hacked card dealing. And there weren't even smart glasses involved in it. What do you think of this is a consumer tech tangent? The Rayman meta,

the new gen that have the wave table display right in them that you look at. What do you think of those? Because I know you've been intrigued by this product before. Yeah, the meta ray bands, I'm less intrigued by the Google XR glasses. I'm going to be much more inclined to be interested in. I think having the display in that makes sense. I'm actually, truthfully, the first time I went into a store and tried on the mida raybans the first time, I was expecting to see a display. Yeah, sure. And the fact that there was not a display, I was like, oh, this is not a product for me. Like, I don't, for somebody that shoots, like, I'll go on like a three week vacation and take like four photos. And usually it's of like a parking

stall number, a menu that I need to show somebody and something else. Like I'm not somebody who's constantly looking to capture and create content for the internet. Right. And I think that's who this product's for us. It's not for me. Interesting. Yeah, because my take on it was that the only thing, I think that they've seen the popularity of the basic version of Meta Raybans, which is just a pair of glasses with headphones in the arms and cameras pointed forward. And they've gone, this will replace the phone. The more phone functionality we can put into it. And there's a little part of me that's like, people might just like having a camera in their glasses. Like there's a signal noise thing going on where I'm like,

oh, I'm worried you might think this is what computing will be. It's like, maybe. But I think the only signal you really truly have is that people love to be able to post to Instagram really, really effectively and being able to click on their glasses to take a photo is pretty cool. So I'm curious to see, I'm curious to see where it goes. I think for me, this makes way more sense. You know, with the AI kind of revolution, air quoting. Sure. But as AI is getting better, interpretive, like image-based interpretations,

So, like, if I can give an image to an AI, I can tell me about it. That process is going to be cooked into these things at some point. So whatever I'm looking at through that camera, the little screen is going to be feeding me details, or I'm going to hear a little voice asking me if I want to know more about it. Hey, you know, you'll walk by a cafe in Tokyo and you'll look at it. It'll just be like, oh, did you know that this is one of the highest rated coffee shops and all of Japan for foreigners on travel? It's like I think we're going that direction.

And truthfully for me, that is more of the direction that I'd be interested in. Like just having a camera that's accessible in my eyes, very low output for me is I'm not one of those people. I know there are those people. I know a lot of people that I fly fish with wear metaray bands because when you catch a fish, they can just say record and then they get a whole video of catching the fish and stuff without having to really disrupt anything else. but that's just not part of my processes. I'm more information focused and glasses with the display in them and the ability to talk to me, I'd be much more inclined to get.

A few episodes ago we talked about a product called Cluley, which is like live AI assistant for meetings. It says off to the side. It's listening to what's being said and it's feeding back kind of like a large language model produced. prompt for what you could maybe say next. When it was initially announced, it was sort of presented as this idea that like cheating, like what previously had been considered cheating would no longer be considered cheating. And you could just sort of parrot what a chatbot says and that's the future. I have noticed that their marketing has pivoted hard away from lying is the future and is sort

of reframed around being much more of like a sort of LinkedIn-esque productivity tool. but let's put that to the side, that vindication to the side. I could see this type of small screen having applications for a certain kind of person who, I guess, feels the need to have chat, CheapT telling them what to say all of the time. You could theoretically in the same way that people were able to basically jailbreak Gen 1 Meta Raybans to do facial analysis

on people out in the world and feed information about people to a phone. Now you've put that screen right in front of the, the eye in a way that the person doesn't even need to look down. So the, I think that this technology is a platform for feeding people information surreptitiously. Totally. It's going to be extremely powerful. Like, it's icky, but the like social engineering implications of being able to like read information about a person on a screen that you don't have to look down at and they maybe don't know you're looking at is like, oh, people are going to get up to some, some stuff with that. That's, uh, I'm sure we'll be talking about it. Yeah, that was clue.

one point oh, I feel like that was like the vision that they foresaw. And I think that vision's still very much alive. I think that vision's coming. This is the technical infrastructure for that vision. For sure. Yeah. The, I, a friend of mine who has a public job speaks at press conferences to journalists and gets asked questions, has a set of glasses that have a screen in them and can receive text messages from his assistance and policy people. So if he gets asked or if they get asked a question, well, in front of a press conference and he doesn't know the answer. I should quit saying he and say they, but it's a he obviously. Yeah, I was going to say you ripped that band-out off. Dude, I don't know if I can edit that one out. If he doesn't know the answer,

somebody can quickly text it to him and it will flash up on his display in his glasses so they can continue the conversation, continue the response to the journalist with the accurate information. Wow. I think we're, you know, we're already doing that. And that's with a pretty old product, I would say. Like, it has an app that runs on their phone, connects Bluetooth to the glasses. We're something like a Google XR, a meta-rayban, AI, like whatever the next, next gen of this stuff is. Like, I think Google's XR glasses are meant to be like Q1 of next year.

Interesting. And the thing about the Google AR glasses, it make me a bit uncomfortable, is that I think they're constantly monitoring your environment. They know everything about what's around you. Because in the presentation that they gave when they launched them in Vancouver, I think it was Vancouver, she had put like a key card on a bookcase and then went about her presentation and then was like, hey, Google, where's my key card? And she's like, oh, it's on the third shelf of this bookcase near the book called this.

And it's like, okay. It's just sort of filming all of the time. She's kind of always there. Yeah. If you feel good about that and you want that, it will definitely be a product that you can get in the very near future. If not now. I don't know how often metarabans. My sense is that meta-ray bands provide like a light-based UI.

A light goes off if you start filming. But I am curious and I would like to learn more about whether or not that light goes on, whether or not the camera is active in any way. I think they're going to have to do it for the first couple of years. You know what I'm saying? I think that people will be very uncomfortable with it. But then as we get more comfortable with it, they'll just take the light away. That would be my futurist expectation.

That would be a bummer to be a person that just needs glasses as a growing suspicion that everyone wearing glasses is recording you. It's like, that seems almost unfair to me to just be like, oh, then we'll just sort of assume that if someone has glasses on there and potentially surreptitiously filming other people. It's like. And listening to everything being said. And listening to everything being said and feeding it off to a server somewhere. And it's like, oh, but that form factor is just a medical device for a lot of people. That doesn't seem great. You're talking to somebody that's born in that medical device since I was very small.

I was like, I'm aware of the fact I'm talking to you and you're wearing glasses right now. It's like, that's just, that's going to be a shift if these things do genuinely take off. And we reach a point where the camera is firing and there isn't any kind of like user feedback. Speaking of shifts. Speaking of. Mike, okay. Remember how we always talk about how Apple's just so behind in the air. AI game.

Yeah, yeah. Pretty famously right now. Pretty famously. Yeah. Except they just contracted Google for a billion dollars a year to provide them with their own custom models and to redevelop Siri to actually be useful. Wow. And I can't help but think that it's absolutely brilliant.

It feels like it's absolutely brilliant in a couple of different directions. It seems like it's very smart because you, you're going to want to get some generative, you're going to want a better intelligence inside of your smash hit smartphone. And they seem to be struggling with doing that on their own. So getting Googles in there. I see the value in that. In a pure business case, giving a billion dollars to a company to develop you AI that is giving you about $20 billion a year to be the default search is extremely trude. It's kind of, it's just sort of changing. It's like the cashier giving you change back, but you get an artificial intelligence out of it.

It's called a credit memo in business. Okay. There you go. I know we owe you $20 billion, but here's a billion off of it. Yeah. And I'm sure it will be more than that at the end of the day. But the thing that makes it brilliant for me is that Apple knew that they were way behind. So they essentially got to pick a winner and just pay for somebody that knows what they're doing to do it well.

And the craziest part of this is that Google Android is selling the Gemini AI integration so hard. It is like a value proposition differentiator for them. And now they're just going to give it to Apple. All of the product differentiation between Android and iOS, or I guess OS 26 or whatever it's called now, they're going to just give away because now they're going to integrate Google Gemini essentially fully, I would assume, fully into the iOS suite. It, they also seem like a good option. And again, this is getting technical in a way that is above my head.

But I know that when Apple rolled out Apple intelligence, they talked a lot about, I think they called secure enclave, which is just essentially an encryption process that allows processes to run on a server the same way a lot of other companies do. But in a way that is more meaningfully encrypted in its connection between the server and the end device. And then I think about what Google has been really good at with AI, which isn't just sort of the largest models, but it's those meaningful. models, those leaner, faster running models that still get most of the utility of the big ones.

I think of a secure encrypted connection, running something on a server, and then you have the option to run a model that is the lightest, leanest, most effective kind of one out there. It's like that, that starts to paint a picture of how this might work. Well, totally. And the fact that they're going with Google, who have already struggled and gotten through most of the roadblocks about deploying AI onto cell phones, ways of intervie. integrating it, hurdles that they had to cross. You know, they've got, they've probably got a billion dollars in knowledge that Apple's just

going to get for free. I know that there was also talk of them working with, I can't remember who was perplexity or, I think it was anthropic. Yeah. Yeah. As being a potential other, you know, group that could be running this backend for a better Siri. And it's like, you know what, Anthropic just doesn't have a mobile operating system that they've

already got AI running on. It's like it's to your point. It's like they, they simply must have solved a bunch of those we in the weeds type problems that come with weaving this into a smartphone. Yeah. It's, it's sort of acquiescing some ground to Google a little bit, but they're already in bed together. They already have such a relationship like we talked about tens of billions of dollars a year

for that search. They're already working together. I get how you get here. I also read on Bloomberg that Anthropic wanted $1.5 billion per year, where Google just wanted a billion. Oh. So they undersold them. Yeah, exactly. So you're getting all the knowledge and paying 36% the price that you would have went if you went with Anthropic.

So I don't know anything about scale AI, which to go back to the Mark Zuckerberg aqua hiring companies type thing. but I kind of got to give it to Apple on this one that like Facebook spent $14 billion on scale AI to sort of just do AI good and better and like Apple spent a billion dollars to make Siri be good. I'm like what I have to I don't know feels like Apple got a deal there yeah scale AI is a data company okay so I think their first big product I'm not sure what they've expanded into but their first big product was that they were like a data repository of pre-cleans index data that a lot of people used to train the world's models. So like a lot of all the big model generative companies, the company's training models

are licensing data from scale AI. Okay, interesting. So the founder of it must be very impressive. I'm sure that the company itself is very impressive as well, but a $14 billion. million dollar aquire is next level. Yeah. And it's part of it said they have a 49% stake in the data labeling company, which isn't even like that company can still license its data to other people because that's sort of

where I was going is do you do this in order to prevent them from letting other new entrance train on that information? And it's like, you haven't, they could still technically do that. Well, but they're probably less inclined to because you gave them $14 billion. If that doesn't buy goodwill, I don't know what does. Well, the other thing is, too, is like, how old is scale AI? So this company is probably, I would say, it's nine years old. But I think it started as a different company than what it became.

Right. So anyway, nine years, it took a startup to grow into that. If it was Google or Apple looking to build a competitive data set to trade models on or Anthropic. I'm sure they would do it in months. Wow. That would be my thought anyways. Like if that became a mission critical thing where like all of a sudden meta was like you can't use any of scales data. Like you have to find your own data to train. I'm sure they would get to it. As that is their like core business offering is these models trained on good information. It is fascinating that you can make a company with 14 billion. They didn't make the data.

No. They collected it. Which is the nice, and labeled it. Think about the last time you heard a breach story on this show. It always starts the same way. Someone somewhere saw something too late. An alert buried, a signal missed, an SOC that just couldn't keep up. Arctic Wolf set out to solve that problem by rebuilding security operations from the ground up for a world where attackers are already using AI. They created the Aurora superintelligence platform with fully agenic system powered by the swarm of experts.

Instead of single-purpose bots or lucky-guess LLMs, this swarm is full of deterministic agents that handle whole entire workflows. Humans stay in the loop and on the loop to validate the critical decisions and keep everything trustworthy, and all of this is just off running on their secure operations graph. A constantly updating intelligence engine fueled by more than 9 trillion telemetry events every week and over a decade of real-world incident response. The system reasons on real signals and real context not synthetic training data. and the result is the new Aurora Agent SOC. It's the first SOC that is agent led by design.

You get agents that coordinate, agents that investigate, agents that respond at machine speed, and hundreds more that automate the repetitive work that normally buries human analysts. Arctic Wolf didn't try and bolt AI onto an old model. They rebuilt the model entirely. What makes it even more effective is how it works with Arctic Wolf's concierge experience.

The team brings customer-specific context directly into the platform so every AI-driven decision reflects your environment instead of generic assumptions. The automation frees your concierge security team to focus on higher value strategy and proactive risk reductions while the agents handle the grind. If you want to see what trustworthy, production-ready AI and security operations actually looks like, go to arcticwolf.com slash hacked. Never feel like cyber threats are evolving faster than anyone can keep up? Last year, 2025 was nothing short of a record-breaking year. year for major breaches, from sophisticated ransomware operators to AI-enabled attacks that

turned defenses on their head. Organizations around the world saw headlines they never expected and cybersecurity teams were tested like never before. But here's the thing. These incidents aren't just news headlines. They're learning opportunities. And that's why Arctic Wolf is hosting a live webinar on February 5th diving the most impactful breaches of 2025. Their field CTO and security leaders are going to unpack not just what happened, but why these attacks succeeded. And most importantly, what businesses can do to fortify their defenses for it's too late. You're going to walk away with real insights into how threat actors are evolving, how defenders are responding, and what strategies can help you stay ahead of the next big breach. It's not fear mongering.

It's practical, actionable, intelligence from experts in the trenches. Register now at arcticwolf.com slash hacked. Is there anything that we need to catch up on? House robots? Did you follow that story? No, but I'll talk. talk about the knock-on to it, the Elon Musk compensation package. Yeah, sure. I feel like that's the real story about house robots. So I read an article the other day that had the potential market size for humanoid robots. It's something like $144 trillion.

So Elon, their humanoid robot, of which the name is escaping me right now. It is very popular. Optimus. Optimus, thank you. Um, they're all in. So now that Elon Musk's trillion dollar comp pack has been signed off, uh, they are apparently starting and breaking ground on building an optimist factory in Texas. So they're about to start cranking out humanoid robots.

Cool. Uh, the reason I brought it up was about a company called Neo. We'll skim past this because I do want to talk about that, which has started selling a $20,000 USD humanoid robot. The story about this was that the robot, you know, you would think would be sort of purely artificially intelligence driven. It seems though, as though this robot, at least in its initial stages, is going to be human piloted. So when you purchase this product, it's going to come into your house. It's full of sensors. It's a robot. But when you ask it to, like, do your laundry, there will,

it would seem be a person that sort of remote tethers in and with a VR, you know, headset and controllers controls the robot and, like, has it do stuff. Laundry stuff. for you. And that's just a product you can buy. Godspeed. Then there's the Tesla robot of it all. Let's hang on your story for a second. I made the quick jump to the Tesla robot. Sure, sure. Sure, sure. So if you were to have domestic staff, call that like a cleaner cook. Okay. This weird human-controlled robot in your house is kind of the same, except for the fact that you don't get to see the person on the other side.

Yeah. I don't know anybody that I think that would be comfortable with that arrangement. No, because of how uncomfortable it is. Like, you chose the word right, my guy. It's deeply uncomfortable. To be like, I want someone to clean up after me, but I'm so uncomfortable with the idea of having to see their human form that I will hire a robot visage for them to embody. and then very like just watching the video of it like trying with its like little

janky robot hands to like pick up a sock and you can tell there's another person on the far side of it like have you ever do you know games like baby steps like those awkward physics games where you're trying to control a person it's like that but it's a real robot in your house and they're just like trying to focus so hard to pick up the cup and put it in the dishwasher and it's like if you want a house cleaner and you can't stand the idea of it being a human and you are okay with it taking eight times longer and having cameras in your house, here you go. Have that or... Oh, man. Okay, let's jump to the real robot stories. The real robot story, the optimist, the trillion dollar compensation package.

140 trillion dollar market. Let's talk about something not controversial. Elon Musk's many enterprises. Well, he's down, he's down. So here's the thing is now that he's, now that he's been essentially, I don't know what you'd call it, he signed the world's largest compensation deal ever, I would assume. It's got to be up there. It's an unfathomable amount of money. Yeah. Granted, the comp pack is contingent on him reaching a number of organizational milestones. So it's not just like he became a trillionaire, but if he can turn, Tesla into like an $8 trillion company, then he becomes a trillionaire. And you know what?

I actually don't have a big B for this compact because he's worth like, what, $300 billion? Yeah, I'm so far beyond finding all of these numbers, like offensive to me as a human being, that it's like, well, this is just sort of a worse. It's like another drop in the bucket. Yeah. And the reality for me is if they said no to his compact, he probably just would have gracefully exited Tesla and then turned to XAI. into the humanoid robot company,

and then taken all of the trillions. So essentially in this way, he's providing this market does manifest and it does become a real thing, an optimist does become a tangible product. He's at least sharing that with the existing investors in Tesla. It's like I'm not super mad about that because he is super rich and could have just gone off,

raise the money. The man has a direct line to like the Saudi investment fund. if he needed cash, he could have got it. I'm sure his departure would have been graceful. I'm sure. And not the sloppiest thing ever. Yeah, it's, so I've never done fundamental analysis on Tesla, nor would anyone want me to. I can't tell you if these, I can't tell you if these like specific pay tiers are

reasonable, if they're overly broad, if they're offensive. I know there's been really considerable push back. Like the Norwegian sovereign wealth fund was like, this is insane. Like there, there was a lot of far better informed resistance to this. It also went through. I am not the person to provide commentary on that.

Yeah. I'm not a Tesla stockholder. I'm not on the board of Tesla. I don't beyond finding the idea that his labor is worth this just personally, like just not true. I don't have a horse in this race. I'm reminded of the quote that he said, which was like, the thing that this hinged on, and you alluded to this of like, if you don't give me a trillion dollars, I'm going to take my toys and go over to XAI. And the quote was, I just don't feel comfortable, like, building a robot army here and then being ousted because of, you know, internal politics.

I don't remember the rest of the quote goes. Yeah. And that, to me, feels like you're hinging this whole thing on like, well, how do we feel about this man overseeing a robot army? because that's what this is, is pay me the money, so I'll build the robot army here for you and I'll oversee it here. Otherwise, I'll go do it. And it's like the question that oversees both those options is how do we feel about one man controlling a robot army? Sure. And this is the premise for a movie.

Many films, a lot of science fiction. A lot of science fiction. And what happens when you give a person this much, an individual person this much power? I don't tend to think that a ton of power is great for human beings. The person that gets it, let alone the people over which they have the power, I'm reminded of Grockapedia. Are you familiar with Grockapedias? I have never looked at it, but I know it exists.

You know it exists. I feel like you have something you want to say about it, and I'm excited to hear it because I've literally read nothing about Grogapedia. So Grogapedia, this is just like a silly tangent way of understanding. all of this and how do we think about a person overseeing a company that has, you know, you know, robots and the word army, which I didn't use, he did. Grogapedia is an AI generated like online encyclopedia. It was developed by XAI, his other company.

It came out last month. It's intended as like a rival to Wikipedia. This was in response to Elon Musk's perceived biases of Wikipedia. I think that it is. I'm just going to jump in. I think the biases on Wikipedia are more than just Elon Musk. I think there's a substantial amount of the Internet that thinks that the Wikipedia is biased. Well, we can definitely talk about that.

It has hundreds of thousands of authors. Totally. So, Grockopedia is created by GROC, the large language model owned by XI. This is according to a bunch of reporting, it would seem that a lot of the articles are derived from Wikipedia, redrafted by GROC, with several entries seemingly reflecting Musk-aligned viewpoints and ideological framing. Well, I think we previously had a conversation

about how GROC was trained to look at Elon Musk's political ideologies for answers. This is well documented. To your point, Wikipedia, a far from infallible enterprise, but by its very structure, a product of hundreds of thousands of people like trying to hash out

and you can just read the logs. Correct. Something close to the truth. Going back and forth, trying to work it out. Grogapedia, it would seem, is a chatbot redrafting, except with those politically sensitive topics redrafted to reflect its private owners' political beliefs. If that is of more value to a person than hundreds of thousands of people, handwriting an alternative, fighting out what is and isn't the truth, there you go.

It's there. To me, it's kind of just a weird novelty. It's like, oh, man, what if I had a chatbot rewrite the encyclopedia my way? But if you are willing to wield your technology like a cudgel in pursuit of the realization of your political goals on your little encyclopedia, and then you go over here and say, if you don't pay me a trillion dollars, I want to build you a robot army. I would say, what possibly about your behavior makes you think I want you in control of a robot army, sir. Well, I think for fairness, I try not to get too political on this, but for fairness, I think he said, my future goal is to build the robot army, to build the marketplace for humanoid robots, to further the conversion from, you know, human labor to robot labor.

Like that's, we all know that that process is well underway. China's developing their own humanoid robots. This is the AI arms race between China and the U.S. is well documented. And his note was essentially, and there's a follow-up to this, so I'll get to that. But his thing was essentially, well, Tesla isn't really that company. It was never supposed to be the robot, humanoid robot company. The only reason we've even been considering in this direction is because I'm interested in going in this direction. I see the market for it. And I think that his statement was essentially,

if I'm going to build the trillion-dollar robot company, I want to be a trillionaire. So either you guys agree to let me become a trillionaire and I'll do it at Tesla, or else I'm going to exit and go over to XAI, another company that I own, and I'm going to do it there. because right after they approved his comp pack, they essentially came back out and said, it's been started to note and swing around the internet, that Tesla might now buy XAI and roll it in because they actually need the AI sophistication

from XAI to Power Optimus. So his many companies are about to reconsolidate just as they did with his solar company got rolled into Tesla. Like, I don't, not sure what's going on with this boring company at this point. feel like it has to be a Tesla company at this point. Yeah, I haven't heard about the boring company in a real long time. So I think the AI arms race is a more interesting topic than who's going to build the robot army.

I think as far as well-known characters in the AI space, the person who that I would assume was going to be the one building the robot army would be Elon Musk as he is as interested in AI as he is in machine. as he is in, you know, you name it, all the different component pieces that will be required to build said robot army. If you say said, it's like Bloody Mary, you see robot army into the mirror three times and Elon Musk makes a buck. Yeah. I suppose my, yeah.

My question, which is maybe bigger than Elon Musk, but is in this particular conversation localized on him because of said trillion dollar compensation package to build a robot army is, is anyone with the desire to build a trillion-dollar robot army of the disposition that we would want them in control of a trillion-dollar robot company? That question is bigger than Elon Musk. He just happens to have inserted himself in the middle of it. I'll counter this and say the government of China. Oh, no, that's not. He can't be that anything he wants to do is okay because if he doesn't, China will do it first.

It's like. No, but China's already doing it. Okay. If you were of the belief that the super cycle in the market and in society is that we're going to more artificially intelligent automation, that humanoid robots will start facilitating manual labor practices, etc., etc., then you're kind of accepting that the humanoid armies, the humanoid army is coming. Nothing about the existence of large language models presumes the need for humanoid robots. It's like we talk about them in the same breadth, but they're not. the same thing. Totally. No. Definitely not the same thing.

I think we all grew up watching and reading the same sci-fi. And I think we're at that place now where mechanically we're capable of building robots that do finite tasks, very complicated actuators. Like I know that Tesla's spent so much time focusing on Optimus's hands to make sure that those hands can do so many of the tasks that humans can do. it that way. Robotics is improving in the same way that machine learning improved and we hit this sort of like thrust forward moment with generally output. Totally. Sure. And those two things are going to collide and not only like not by like design but by will

and so the Chinese governments and a lot of Chinese companies are now driving this agenda on their side of the ocean. somebody in the States either has to pick up that baton and run with it in a market economy. And to me, it's like, of course it's Elon Musk. And so the question would become, if we know that we're going to a future where there are humanoid robot armies, do you want the one that's doing your dishes to be the one that's licensed from the Chinese government? Or do you want the one that's developed by Elon Musk?

I guess I reject the idea that the binary is the Chinese government and specifically Elon Musk. That Elon Musk is the only embodiment of that. Like, I hear what you're saying. I really, really do. Yeah, yeah. But I brought up, I did that whole silly Grogapedia anecdote to be like the disposition of a man that would rewrite the encyclopedia, except how he sees it is like, that's relevant in a discussion like this. what you are willing to do with your power. I'm not, and I do reject the idea that if I don't go, like,

it's awesome that Elon Musk is doing this. I'm like in favor of the Chinese government doing it. I'm like, that's not real. No, no, no, no. I'm specifically. I've totally created a false dichotomy. I get that. To illustrate a point, I take it.

Yeah, yeah. I'm drilling in on like a man's, a specific man's disposition. Sure. I think that, like, obviously there's going to be more entrance, like this is a free market. You can buy a Neo. You can buy a Neo right now. But if you start talking about $140 trillion market share,

you're going to start seeing a ton more entrance coming into that market. So the false dichotomy is very false. It's just that the fact that the two parties that are the furthest along in this mission appear to be Elon Musk and government of China back companies. So where we get to in the next five to seven years, for sure. Who knows what it's going to be? I haven't looked at Grogapedia. I can't look at it.

I'm sure it's the same sentiment as why he bought X. He wants a free speech, more unbiased internet. Whether you agree with that or not is what it is. For me, it's, if there truly is like the AI arms race between North America and China, I don't know who else is going to do it besides Elon. And that's why I don't hate his compact. Like, he could have gone over to XAI and done it, raised the money, built the factories. Like, the man is worth enough money to do anything that he feels like doing.

So, I don't know. As a Tesla shareholder, I voted yes. Yeah, I mean, it can only make the line go up. Yeah. It just leaves a bad taste to my mouth. I think the real umbrella question that you want to talk about, and maybe this is not the place for it, is when income inequality gets that high in a society. usually that society crumbles throughout like history.

That is definitely like skirting the boundary. I alluded to this up front, which is the idea of like, it's funny how if we let the clock on this show go more than 30 minutes outside of the normal hour, we end here is like I'm trying to separate out the fact that I find the idea that anyone's labor is worth this amount of money. It's like I know there's just a political difference between me and other people in the world on that. And I'm trying to put that off to the side of the desk because otherwise all conversations about these amounts of money just sort of end with, well, that shouldn't be the case. Hmm.

Okay. Totally fair. Look how far we got. Look how far we got. Meta running scam ads. Did we ever talk about Binance? No, we didn't. We didn't.

Okay. Well, I cut it up in the beginning. This is going to be a long one. No, and I have to say it. but I am, and I've said it on the show before, a pretty staunch believer in the freedom of speech. Okay. And I think it is a fundamental freedom that needs to be protected at all cost. I would agree.

I don't like seeing anything impacted. I think that there should be consequences to behavior and the utilization of it, but I don't think it should ever be made illegal. Of course not. Yeah. Wholeheartedly agree. I can't think of any better embodiment. of freedom of speech than hundreds of thousands of people hashing out what the truth is.

Okay, let's cut to Binance. Let's get to Binance. Okay. For anyone that doesn't know, Changpeng Xiao, known very widely as CZ. He's like a very influential cryptocurrency figure. He was charged. He's Canadian citizen born in China, migrated to Vancouver, where I live. He founded Binance in 2017.

It was one of the largest cryptocurrency exchanges on Earth within like 18 months. Like just stratospheric kind of take off for this guy. At its peak, two of every three crypto trades was done on Binance. Regulators claim Binance platforms were used for money laundering, sanctions evasion. November 2023, the DOG announced that CZ had pleaded guilty to violating U.S. money laundering laws that they had knowingly failed to like put in place protections for anti-money laundering. They were fined $4.3 billion personally.

Zee personally was fined $50 million. He was sentenced to four months in U.S. federal prison and barred from running the company for three years after. We were talking about this because he was recently pardoned. President Trump went on 60 minutes and famously said he does not know finance founder Chang Peng Zhao 10 days after granting him the presidential pardon. At the heart of this is a conflict of interest question centered on the Trump family's crypto venture world liberty. Liberty Financial and USD won a stable coin project that was used in a $2 billion, Abu Dhabi back to investment into Binance, CZ's company this past spring. We haven't talked about crypto in a while.

We haven't talked about crypto. It's been a minute. That's been a minute. Yeah. I was just thinking it was like, okay, so we've got CZ. He got the pardon. Yeah. But FDX and Bankman Free didn't get the pardon.

But I guess that's, it's actually surprising. that Biden on the way out the door when he just blanket pardoned everybody didn't also blanket pardoned Sam Bankman-Fried because Sam Bankman-Fried had given what like $100 million to the Democratic Party? A significant donor, totally. So it's surprising that CZ catches the pardon before Sam Bankman did, and Sam Bankman literally prepaid for it. It really didn't get what he paid for on that one. It really didn't get it. Yeah, you could maybe argue that it's one is a deal in the past.

It's like you said prepaid for it. Exactly. One's a deal in the past and the other is a deal, you know, still, still happening. USD1 and the Trump family cryptocurrency is still unfolding right now. So. Does CZ have anything to do with it? Has that been announced? I don't know that the pardon revokes.

I think the pardon would revoke the rule saying that he's not allowed to be part of Binance. So theoretically, it would be bringing the old leadership into Binance who has the connection to the World Liberty Financial kind of project. But we're just going to have to have to see. Yeah. Interesting. That's a good one. I wonder what Sam Bankman's thinking.

Probably pretty pissed. I'd be mad. I'd be mad if I spent that much money. I was just like, oh, for sure, get a pardon when this all blows up. And then I didn't. You be so mad. Especially when Biden was giving like preemptive pardons. Like just anything my son has ever done is like totally okay, right?

everybody's cool with this? There's not even charges against this person, but they're pardoned for anything that they might be charged for. It's like, is that even constitutional? Yeah, it's starting to look like this whole, I'm not an American lawyer, but like the whole pardon thing is like a pretty wacky premise when you really think about it, or it's just like there's like a golden ticket to a law not applying to you. It's like, ass wild. What's that ticket cost? What's that ticket worst? It would seem. We have a price tag. Yeah, so he made over, San Bagamette Fried made over $40 million in donations to the Democratic Party in 2022 alone. That's a, that's a beefy contribution. That is, that is like, that's like a,

like, what's it cost a monopoly to get out of jail? Yeah, a lot less than that. But so does property. So I don't know if their prices are pegged to inflation anymore. That's true. That's true. Anyway, I say we had it there. You wrap it up. This is a chatty chat. I hope, I hope your dishes are done. Hope you made it through your commute. We'll be back next episode with an interview. I'm very, very excited to share with y'all. I think fun. But until then, I think we catch you in the next one. Take care.