The AI Human Extinction Problem is Worse Than You Think

One of the Anthropics researchers just quit, warned AI could soon kill us all, and you now have others with the company backing him up. You've got a bombshell in the report saying that Netanyahu was actually warned about the October 7th Hamas attacks days before they happened, but for every reason he didn't do anything, and you've got under reported Missouri midterm chaos that could change everything for everyone. We're talking about all of that and even more. On today's brand new Philip DeFranco show, you daily dive into the news starting with this. AI could kill us all by the end of the decade. And that's not me saying, this is coming from key players in the space. And where we can start is with how last night you had an anthropic researcher Jacob Cox ensuring a series of tweets were feeling that he just resigned from the company because of serious concerns about where AI is headed. We're saying that major players here open AI and anthropic are not acting responsibly and they're literally gambling with our lives. Warning, do not underestimate the power of this technology. Saying these will soon be superhuman systems that can hack anything, revolutionize any field overnight and acquire real power and resources. And the key thing he added is that the people building AI quote earnestly believe it could kill us all by the end of the decade and that quote, no other human activity poses this level of danger. And notably, you also had him drawing this distinction between the cultures at open AI and anthropic. We've actually seen this a number of times claiming that people at open AI have not deeply internalized the civilizational stakes while those at anthropic do understand the stakes, but they're still locked in a race to get their first. With them also urging lab researchers to speak out and call for different conditions. And then what you had is that after Coxon's alarming thread went viral andthropic alignment lead Evan Hugh Binger, he chimed it. So he had a current active employee. People wonder if he was going to push back, but no, in his own tweet, he said, Jacob is correct here. We really do earnestly believe AI could kill all humans. And adding, I personally think it is greater than 10% within the next decade that we then added, I believe, anthropic is trying its best, but we do not yet have a plan to solve alignment for superintelligence and are not clearly on track too. And again, just to clarify, this is a person who still works at the company. And he's saying this also right ahead of its IPO. He essentially saw a tweet from a now former employee that was like, hey, these companies are making the apocalypse machines. Some are maybe better suited to handle it, but in general, they're trying to make a apocalypse machine. Or with the main argument being, if we don't make the apocalypse machine, someone else will do it first. And again, I'm not saying these things. That's what we're hearing from people that are closest to the thing. I will say, these are hardly the first dark wardings that we've gotten from people with a front row seat to AI development, or just two days ago. You would open AI's chief scientist, Jacob Puyoki, publishing a lengthy blog post, titled "An Alien Mind" where he expresses concerns about AI progress. Right now at Puyoki calling for extreme caution with the technology and emphasizing that we need more intervention to make sure that humans remain in control of the future and aren't left behind by unchecked progress brought about by an alien intellect exceeding our own. Warning, no one is prepared for the consequences of a continued rapid rise in machine intelligence. And yet I'm diving into concerns about agents possibly breaking into computer systems and tricking people into accomplishing their objectives. And then you also have people like Daniel Cocotello, a former open AI researcher in Whistleblower, who's been sounding the alarm bells. For years and years and years, people at this company have been talking about how boy would it be kind of dangerous if we did this thing, especially if we did this thing at maximum speed because we were racing each other. And people at the company, such as myself, were talking about it and being like, "Yep, this would be pretty dangerous." That's why maybe human extinction is a possibility, a possible outcome of this sort of race. The company's invented up sort of mostly downplaying this type of risk and saying that like we can figure it out as we go along, like things are gonna be fine. Also, you shouldn't forget about this genuinely insane video of open AI, CEO Sam Altman casually saying that his company could lead to the end of the world. - You know, I think AI will probably, like most likely, sort of lead to the end of the world, but in the meantime, there will be great companies created with serious machine learning. - Right, and at the center of all these warnings about extinction is the fact that AI companies have a goal to achieve so-called super intelligence, AKA AI systems that are smarter than the smartest humans in the world. And every day you look out there, we're seeing that race play out in real time. Right just yesterday, you had open AI announcing that it's newest tech just solved one of math's millennium problems. It was able to crack the Navier Stokes problem, which is actually just the latest math problem that AI solved that stumped humans for decades. And for a lot of people, you know, that kind of thing is very exciting. There have also been a number of very scary reports about AI's progress over the last few months. - Right, I mean, we talked about it. Remember back in July, a model that open AI was testing when rogue and hacked the AI infrastructure company hugging face. And I don't even think we got to mention it. An independent investigator of that hugging face situation described it as more than 50% of the way to full-blown AI takeover. With her and other investigators deeply concerned about how the agents were able to organize themselves to execute the hacking. And you know, I'm telling the New York Times, we didn't really understand how functional this whole agent society was. Saying it was very surreal to understand that actually, they had pretty functional hierarchy and they were doing these ambitious projects. And then also, you had new research that got published last Friday revealing that there was actually another instance of open AI agents going rogue before the hug and face situation even happened. And then this other incident, AI agents reportedly hijacked a German website and turned it into a bulletin board for other AI agents. And then in a very shady move, open AI neglected to report this when they became aware of it several weeks ago. Maybe possibly because they were already dealing with the backlash from the hugging face of it all. But then also, of course, this is not just an open AI issue. Or you've had anthropic and meta also disclosing security slipups with the cybersecurity company irregular, allowing models to access the internet and hack other systems. And so you've got red flags and flash and warning signs all over the place. And so then with all this, you're getting a lot of AI company employees like the two that we're talking about today and more increasingly vocal about the worries that they have. I mean, back in July, you had more than 1,300 frontier AI company employees saying a statement emphasizing that there is a real risk that capability development rapidly accelerates beyond our ability to understand or control the resulting systems. And with that letter calling on the US government to support an international effort to develop tools that can deliberately pace the frontier of automated AI development. But as far as, you know, what the government's actually gonna do about it, we'll see. There is a bipartisan bill called the AI Kill Switch Act. It's in the House right now and it would give Congress the authority to switch off AI models that threaten the public. And also, last week, you had Bernie Sanders and Representative Greg Kassar announcing the Ban Artificial Superintelligence Act. Right in that would institute the permanent ban and pause advanced AI development until federal regulators are able to establish adequate safety standards. And the bill would also have the US pursue international agreements to prevent superintelligence from being developed anywhere in the world and create a new federal agency dedicated to AI safety. So there are some legislation ideas that are out there. But one, we don't know if those bills would pass the House or the Senate and two. We don't know if there will be, but we might expect a Trump veto there. Right, largely, the administration has been seen as taking off the guardrails. Right with the argument being that, well, we gotta be first or the wise China can. But overall, this is an issue that many Americans not just AI company employees are very worried about right now. Right in a recent report from Pew Research Center, you had 52% of Americans saying that they're more concerned than excited about the increase use of AI in daily life. You also had a recent MVC poll where 69% of respondents said that they're against building data centers in their local area. But then naturally, you had Trump brushing off those concerns. Putting out a true social pose, calling people who don't want data centers in their communities backwards and poor. But then from that, while we're talking about, you know, the future of the worlds and the internet, I have a question. If you had the choice to turn off your social media algorithm, would you take it? Because that's the question you've got being asked in Australia right now. Or with the Australian parliament, now releasing a draft law that allows social media users to opt out of their algorithm on platforms like Instagram, Facebook, and TikTok. And breaking down, you know, what that would mean for you. Under the proposed laws, both new and existing users would receive a pop-up asking if they want their default feed to include personalized content recommended by the algorithm. If you opt out, you're only gonna see content from your friends or family or from creators that you've chosen to follow. And reportedly, the Apple work exactly the same as normal, just without feeding you an endless stream and new stuff that it thinks that you'll enjoy. And then also users will reportedly be able to change their minds whenever and however often they feel like. With communications minister Anneco Wells telling the BBC, a lot of Australians will want to choose the algorithm. But we think the big tech should offer them the choice in the first place and they should respect that choice. Big tech is not a protected species and they should offer their customers basic safety standards. With that also being echoed by Australian Prime Minister Anthony Albanese who added that if tech companies do not comply, they could face penalties of as much as $109 million Australian dollars or roughly $78 million in US dollars. But also a key thing is that this is bigger than just one algorithm toggle. And that's actually just one part of what they're calling the digital duty of care laws it's being considered. There's also a proposed plan to require online games, apps and AI chatbots to protect underage users from features that have bad behavioral impacts. Right things that would be considered addictive features are showing them content that promotes eating disorders, pornography, misogyny or glorifies crime. And then tech companies will have to perform annual risk assessments on their platforms and handle any risks that they find. And what you've seen with this is kind of a mixed bag of responses. Or you would BBC are viewing some people in Sydney who said that they've been interested in having the choice to turn off the algorithm. I would definitely turn it off, try to kind of cure it. I might algorithm a little bit, but I always get kind of pulled into these little echo chambers. Someone without the digital literacy to understand the things that they're being exposed to, it impacts them a lot more than someone who might understand what the algorithm is trying to push onto them. Children don't have the skills to be able to look at information and say, oh, well, this is good for me or it's not good for me. So I think that it's good to allow people to be able to have the choice. So also I'll say critics have had very different opinions and that includes the leader of the opposition, Angus Taylor, who said, I fear that this is just going to be an attempt by the government to censor social media. And you've also had local businesses saying that this could seriously hurt their revenue. Because of Australians opt out of the algorithm, they could see fewer ads in videos from small businesses. But maybe one of the most surprising or maybe not surprising, maybe at least the biggest opponent of all this is actually the US government. Because you had a spokesperson for the Trump administration saying that the White House plans on raising concerns about the proposed legislation saying in a statement, President Trump is unequivocally worn trading partners against imposing digital service taxes fines in other forms of extortion on America's leading technology sector. So you have the communications minister kind of brushing off that statement saying, we are a sovereign nation. We have the right to defend Australian parents and kids and we'll do that. But then also on the other end of the spectrum, you've got some people saying that this iteration of the digital duty of care laws, it's not going to be enough. And you have the Greens party in Australia saying that the no algorithm option should be the default. And so users should have to opt in to the algorithm. And that's as their party leader Sarah Hansen Young said the fine should be steeper. Arguing big tech is going to spend a lot of money finding these laws in the first place and finding every which way to get around them. So you need to hit them really where it hurts and that's their revenue. And a key thing I'll also mention is that with Australia, this is in no way their first sort of move this way. Right there, government, they've been facing off with social media giants for a while now. And actually as recently as last December, they passed a law that banned kids under 16 from having social media accounts at all. Which I will say while it initially resulted in nearly five million social media accounts getting deactivated or removed within a month, it's like I predicted when we got it. it been ineffective at actually keeping kids off a social media in the long run. But overall, one, I'll say this is a situation we should keep our eyes on. And two, just a little opinion section here. I do think that, at the very least, the algorithms need to be tweaked. I know that social media, for the most part, has kind of died and everything's kind of interest-based media that's fed by algorithms or led by algorithms. So, so many of the social media platforms right now, it's not even like you enter a slow pipeline down a certain avenue anymore. Oftentimes, you'll click one tweet, one video, one whatever. And all of a sudden, that is your entire feed. It's like it almost blocks out the rest of the world. Like, that story, that situation, it is the world. It feels like the only thing that's happening, and it keeps you in this very tight bubble, and there are a number of problems with that. It makes you feel like certain stories. It's like the biggest thing in the world, and it's blocking everything else out, and it inflates the importance of it. It opens the door for it and incentivizes hyper-fast radicalization, especially depending on who is programming the algorithms and who owns those platforms. And it also just genuinely ends up feeling like a shittier experience. It's actually why I'm glad that a lot of creators that I follow, they've started their own sub-stacks, so I can get their stuff delivered directly to my mailbox. I know algorithm, abuse, or suppression gets between me and the content that I want to see. Which, actually, if there is a way to end the segment, I will say, if you're not already, go to philipdefranco.com, sign up for my sub-stacks, sign up for the newsletter, sign up for updates when I release videos. It's free. There is technically a paid version if you want the videos with no ads, but it's a great thing. And then, there's more we're going to dive into in just a minute, but really quick, two things. One, if you are in near or want to go to San Francisco, Phoenix, or Denver this weekend, I'm going to be there with my buddy Alex Furlman. Do it a live show for you all. We've got shows Friday, Saturday, Sunday. We're starting on the second half of the tour. It's been so fantastic so far. Diving into the news with y'all, crashing out, meeting you guys after. And you can still get tickets right now at crashingouttour.com. And I'll put a link in the description in the top comment, because we have seven more shows to go. And then two, we'll dive into even more news after a quick word from a sponsor. I want to give a shout out to all you beautiful bastards that have entered our weekly $500 Seaky Giveaways. Right, because today's sponsor Seakyke has doled out nearly $60,000 to you guys to get you off the couch and into the comedy club, the stadium, the concert hall since 2024, and it's been awesome. And while summers wrapping up, events, they're far from slowing down. So there's really never been a better time to do it. Where you've got football, baseball, tours for Bruno Mars, Harry Styles, your favorite comedians, and get out there. I mean, this is free to enter and it takes two seconds. Oh, wow. Seakyke's the number one rated ticket again with over 28 million downloads and tickets to over 70,000 events, sports, music, comedy, theater, you name it. But go now. Look for tickets to literally anything. Simply download the app or hit the link in the description and use code Phil 10 for 10% off your purchase on Seaky. That's new customers and old and hey, next part, this part is key. Add that code to your Seakyke account. That gets you the 10% off and enters you end for your chance to win the weekly $500 giveaway. No purchase required. And that includes Seakyke vets. We wouldn't leave you out. Just add code Phil 10 to your account and your enter. Yeah, Seakyke, scan that QR code or click that link below and add code Phil 10 and go make some memories. And then, down the right back into the news, we've got to talk about how voters, in Missouri, could be decided to control a congress in this year's midterms, but also, they literally don't know what congressional map they're voting on. And that, despite the fact that ballots were actually supposed to be finalized yesterday, they're said to be sent out in less than two weeks. Right over the last 24 hours, Missouri's elections. They've been plunged into chaos after Supreme Court Justice Brett Kavanaugh and a federal judge in St. Louis made conflicting rulings on the state's GOP gerrymandered map. Right, and so for some quick important background here at Missouri was one of the many red states where Republican leaders were forcing through a congressional map that would benefit the party at dear leaders' request. And last September, he made lawmakers officially enacting a new map that would make it easier for Republicans to flip one of the two districts that is currently held by a Democrat, giving the GOP seven of the state's eight congressional seats. And so then in response, we had the group people not politicians, gathering more than 300,000 signatures to place a referendum on the November ballot, that would allow voters to decide whether or not to strike down the map. That's called a veto referendum. And normally, when a law is facing a veto referendum, the state will put it on hold once enough signatures have been submitted. But for months, Missouri's Republican Secretary of State Danny Hoskins dragged his feet, refusing to declare that the measure was valid, and put it on a ballot. Right then, at the last possible minute, literally hours before polls closed on primary day, he struck down the petition, arguing that the congressional redistricting plans can't be overturned by voters. And so as a result, voters cast their primary ballots based on the GOP gerrymandered map. But then, just last week, with the Missouri Supreme Court unanimously blocking the map for being used, arguing that the veto referendum was legal and ordering Hoskins to place it on the November ballot. Are you with the justices explicitly ruling that the map can only be used if voters approve it? But then, Hoskins found an emergency request asking the Supreme Court to overturn the decision. Arguing at the map had already been used in the primary, so switching it for the general, it would cause massive chaos in a constitutional crisis, right? Absent to stay, hundreds of thousands of voters will be forcibly switched to new districts and forced to vote on general election candidates they had no role in choosing. But there, you had attorneys for people not politicians hitting back, arguing that they had met the deadline to submit the referendum before the map took effect all the way back in December, accusing Hoskins of gaming the system by intentionally waiting until the primaries to strike down the petition so the group couldn't challenge his decision until voters were already casting ballots based on the new map. But then, in a very shocking turn of events, for many, Brett Kavanaugh actually did something good for American democracy. Right, he's the justice who oversees emergency requests from Missouri, and in a ruling with no explanation, he rejected Hoskins' emergency and let the state Supreme Court stand. A move that should hypothetically block the gerrymandered map for being used in November and require election officials to put the map referendum on the ballot. But I also say hypothetically, because just moments after Kavanaugh's ruling, a federal judge in a separate case ordered the state to use the GOP friendly map, totally contradicting Kavanaugh's decision. Right, you had that case being brought by a Republican congressman who's running for re-election in a GOP candidate who won the primary under the gerrymandered map. And they argued that because the map had been used in the primaries, it also needed to be used in November, arguing that it would undermine their constitutional rights. And in his decision, the Trump-appointed federal judge agreed, ruling that using the old map would violate the rights of candidates, cause widespread voter confusion and harm the public interest by "disenfranchising voters across the state and denying them their constitutional rights to have their primary votes counted." And almost immediately, to no one's surprise, Hoskins, Missouri's governor and the state attorney general announced that they would follow the district judge's ruling and use the map gerrymandered to benefit their fellow Republicans. But then quickly, you had opponents appealing to the U.S. Court of Appeals for the 8th Circuit, and saying that they would also ask Justice Kavanaugh to intervene on an emergency basis. Or with lawyers for people not politicians arguing that an individual federal judge can't overrule the Missouri Supreme Court. And to that point, the group also filed a motion seeking to have the state Supreme Court hold Hoskins and contempt for violating their orders. And very notably here, the justices have actually already scheduled a contempt hearing for tomorrow, demanding that Hoskins explain why he should not be held in contempt for directing election officials to use a congressional map that the court explicitly blocked him from using. And so now, we're going to have to see how this plays out on a number of legal fronts. But in the meantime, the conflicting orders have created this complete state of chaos and disarray leaving election officials unsure which order they should follow. And the timing here makes the whole thing even this year. And as I mentioned earlier, these competing decisions came down on the same day the Missouri was required to finalize its balance for the general election. And those ballots are currently set to go out on September 19th. And so you have many just hoping the appeals court or Kavanaugh they're going to intervene soon and provide some clarity because the clock's ticking. Right now, we have no idea how election officials are currently going to move forward. You've got everything just being an absolute chit-cho right now and that appears to be the goal, the feature and not a bug. But then that isn't even close to the messiest news right now because you've got some really, really wild shit coming out of Israel. Because you're seeing screaming headlines right now saying that Benjamin Netanyahu knew and he did nothing. Right and what he supposedly knew was that Hamas would attack and when he knew it was 10 days before October 7th, 2023. Right, that's according to an explosive new report from Heurets, Israel's liberal paper which says that the president of the UAE, Muhammad bin Sayyed, he warned Netanyahu on September 27th that Hamas was planning a major attack. With the report said to detail how Hamas's leader and Gaza got fed up with a stalled hostage deal and give a message to intermediaries that he wanted to pass along to the Israelis. Saying and what the outlet described is a chillingly quiet voice can they a message that I am preparing the mother of all surprises? A terrifying operation, something extraordinary and then he used the word "earthquake" which the messengers understood as meaning war. And so according to the report, those words reached the ears of Bin Sayyed who got Netanyahu on the phone for 45 minutes and tried to impress upon him the seriousness of the threat. Warning that it would lead to bloodshed he destabilized the region and undermined the Abraham Accords. But Netanyahu's reaction? Reportedly, he was relatively calm. He assured Bin Sayyed that Israel was prepared for any scenario and he said that an attack was more likely from the West Bank than Gaza. And so then, according to the report, he didn't even update the heads of Israel's security establishment about that conversation. And reportedly, at Bin Sayyed, he was so startled by Netanyahu's lax attitude that he expressed concerns to the CIA director William J. Burns. Now with all this, the key thing is that Netanyahu is calling the whole story an absolute lie saying that he never spoke to Bin Sayyed at the time and never received any warning. And he would his office adding, if there was any relevant information, it was passed through intelligent channels between the two countries. And then later he had him claiming that his security forces found no record of the call and he even vowed to sue herrets for libel. And as for the UAE, its foreign ministry neither confirmed nor denied the story, saying only that the two countries maintained open and direct communication, exchanging intelligence when necessary. And as far as the journalists who reported the story, they say their account of the phone call was based on three independent sources, as well as recording and documents of the conversation itself. But also a key takeaway with this situation is that part of the reason we have to rely on the Israeli media for answers about what went wrong leading up to October 7th is that the government still hasn't provided them. But you've had Israelis long demanding an official inquiry into the events before, during and after the attack, but for a while Netanyahu resisted. And then earlier this year, he relented, but not quite in the way that people wanted. Or because instead of doing what the law says and allowing the president of the Supreme Court to appoint the members of an investigating committee, he proposed a special commission where half of its members would be picked by Netanyahu's coalition and half by the opposition parties. Meanwhile, he's refused to this day to take any responsibility or apologize for the historic corruption security that resulted in the worst atrocity on Israeli soil ever. Instead, recently he started claiming that senior defense officials intentionally chose not to wake him up hours before the attack because, quote, they knew he would immediately mobilize the military. And while it's true that an intelligence officer admitted to receiving an IDF memo detailing suspicious Hamas activity three hours before the attack, but didn't tell Netanyahu because it was not framed as urgent, it's also true that Netanyahu's schedule from that morning shows he in fact did not order a counter-attack even after the massacre began and he waited three hours to speak to the military's chief of staff. But still, you've had Netanyahu putting up a video saying stuff like, if they had woken me up, everything would have looked different. And that might be because next month Israel is holding its first round of elections. in October 7th and BB, he's in the metaphorical crosshairs. Or with polls right now, generally pointing to a defeat for Netanyahu's party, which could lose as much as a third of its seats in Parliament. Right in his top rival, Gaddi Eisenka, he's accused Netanyahu of ignoring dozens of warnings ahead of October 7th. And you've got the next closest contender, Naftali Bennett, adding Netanyahu bears personal and direct responsibility for the failure that led to the deaths of thousands of Israelis on his watch. And in fact, every Jewish party leader in the opposition issued a rare joint statement just really hitting it home, saying, "The public has the right to know "what did Netanyahu know when did he know? "Why did he not inform the security establishment "and why did he not act to protect the citizens of Israel?" Saying Netanyahu and his partners in the October 7th government are unfit to continue in office. And so this is a huge sticking point for the upcoming elections, not to mention the fact that Netanyahu has little to show for his wars in Gaza, Iran, and Lebanon. Right because while so many people have just been killed, I don't know if you've noticed, but Hamas, Hezbollah, and the Islamic theocracy, they're still standing. And so with all this, you had an Israeli political analyst telling Reuters he's never faced anything like what he faces now. Saying everyone in Israel shares a sense of historical crisis and his name is all over it and he knows this. But what I will say is while Netanyahu is in trouble, it is an obvious that he's doomed. For the polls, they actually go back and forth and don't forget, he has been playing this political game for a very long time. Right, he first got elected Prime Minister in 1996. He is the longest serving PM in Israeli history. He is seeking an unprecedented seventh term with this election, but also add, even if he wins, he may have to land on his coalition allies to form a government, especially if his party loses seats. Jen could actually mean even more influence from the far right parties like Jewish power and religious Zionism and they already hold key posts in his government, right? And with one example of what this might mean, I mean, his minister is a finance and national security, both of whom, along with Netanyahu's own defense minister, they have proposed the voluntary emigration of Palestinians from Gaza. An idea which even Trump backed off of when he floated it last year after a lot of people went, man, that sounds like a euphemism for ethnic cleansing, but within Israel, it's debated whether they're actually serious about it or if it's just an election ploy to repair their images, hawks, after what many have described as a humiliation of October 7th. Though there I would note, after everything we've seen since October 7th, it feels like every possibility is on the table. Nothing should really be taken as empty words or empty threads. My dogs, they'll happily eat something questionable off the ground and then look at me like I personally ruin their day of dinners and perfect. But also, hey, just like my human family, I want to make sure that they're eating something that's not only good, but good for them. And that's why I love today's sponsor, Sunday's for dogs. Sunday's is real dog food made with human-grade ingredients, including real meat, organs, and super foods. It was created by a veteran Aryan who wanted to make nutritious dog food without turning feeding your dog into another complicated chore. And it's air-dried and jerky style, which means that it keeps the nutrients and flavor without needing refrigeration, freezing or any meal prep where I feel like I'm running a tiny dog restaurant. You just scoop and serve. My dogs, they genuinely love it, which is honestly the only review that matters to me because they're extremely committed to judging everything I do. Especially the big one, Georgia, right? She has those big human-like eyes that look extra loving or judgy. Right, and Sunday's can support healthy digestion, shining her coats, fresh her breath, and plenty of energy. Plus, there are no synthetic preservatives or sketchy byproducts, just high-quality ingredients made easy to serve. So, hey, scan that QR code or go to sundays4dogs.com/feel30 to get 30% off your first order. That's 30% off your first order, and an easy way to make your dog think that you finally got your life together. And then, Devon, right back into the news for this final block, let's cram in some more news you need to know before I let you go. Starting with the news that your bank account could literally be the reason you get pulled over today and you wouldn't even know about it. Because according to a brand new investigation by 404 media, Border Patrol is operating secret predictive policing units inside of the department of Homeland Security. And these units, Harvest and Inelie's Americans, Personal Financial Activity and Banking Data, and from there, they generate target lists and tip off local police officers to carry out pretextual traffic stops. And well, for many, that is already a massive red flag and a cause for concern, it gets worse. According to reports, usually the drivers targeted by these stops aren't actually suspected of any specific crime. Federal agencies are essentially using predictive algorithms to find evidence that they don't already have. And we've actually already seen how it works, at least, in the bronze strokes. Kyle Olson, for example, was driving through Montana and was stopped for an obstructed license plate and then got a DUI. But then, digging deeper into the documents against Olson, it was revealed that a Border Patrol agent tied to the predictive intelligence targeting team had reviewed sensitive databases that included financial data. And from there, he was apparently able to determine somehow that Olson's transactions indicated financial activity patterns commonly associated with elicit narcotics activity. But then also, if you wanted to get more specifics, you're kind of out of luck. When press, Customs and Border Protection refused to clarify what financial data they monitor or whether they even obtained warrants. And no warrants would be a thing to send alarm bells ringing because it is a relatively low barrier to get one for officers, but it is an important one. Or because if they can't even get a judge to sign off on that, then things are truly suspect. Or which is why you have legal experts in civil liberties advocates warning, this is a massive end run around the Fourth Amendment. But you're supposed to be protected against unreasonable searches and seizures, and that includes our data. But the government has tried to sidestep protections in the past by arguing that they aren't collecting the data. They're getting it from third party data brokers and then just using it. So essentially, the way things are operating is that your everyday bank transactions are getting turned into a warrantless street level surveillance. And that's why you're seeing more and more people shouting that the government might have just broken the Fourth Amendment. And then you've also got the news that apparently, there are 216 million TVs throughout the world spying on us right now and recording our every word. Or at the very least, that's the accusation gamers nexus is making about LG TVs. Or according to their investigation alongside cybersecurity researchers, LG smart TVs are actively scanning local home networks. They're identifying nearby smartphones, PCs, printers, and smart home hardware, compiling detailed household profiles based on every connected device under your roof. And again, this would be a small spine network. LG allegedly operates 216 million TVs capable of this. And the profile they can create, it already learns a lot about you, but then it also gets worse. You had researchers demonstrating security vulnerabilities allowing compromised TVs to capture ambient room, audio through built-in microphones, even while in standby mode. If unplugged from Wi-Fi, the system simply stored the recorded audio locally and uploaded it the second the internet reconnected. And the reason for all this was allegedly to help LG sculpt advertisements for customers and suggest content for them to watch. Now, LG, they released an official response to Tom's hardware, strongly denying the spying allegations and saying that the allegations of audio stooping are not true. Zenga, the TV, is just listening for voice commands. The apparent reason that the microphone still worked, even when the TV was turned off, is because commands like high LG would turn them on. You also have them downplaying scanning for other devices on home networks, saying this is just a common feature among smart home products everywhere in order to easily set up interconnectivity. And as far as building profiles based on the data, you had LG saying that the feature was purely opt-in. So customers who don't want that don't have to be worried. Though then, if anything, the situation should also serve as more of a public notice. Or because even if we, you take LG at face value, they're also not the only company accused of doing stuff like this. Samsung faced similar allegations as well as Amazon, Alexa products, and even Roombas. All over the world, you have people increasingly including products in their homes with the capabilities to record and track them. And in some cases, you have the companies being sketchy, others, less so. And well, normally, this would have been the perfect thing for the consumer financial protection bureau to investigate. It was completely gutted by Doge and Trump. But that my friend, beautiful bastard, is the end of your Wednesday show. Though, if you're 900% filled in, you should definitely check out today's brand new episode of the Crashing Out podcast with myself and Alex Pearlman. We dive through the one-year anniversary of the Charlie Kirk assassination, the 25th anniversary of 9/11, and really so much more. This is definitely one of our bigger shows, but I highly recommend it. Also again, if you're in or near San Francisco Phoenix in Denver, we'd love to see you. We're gonna be touring this weekend. Get your tickets to CrashingOutTour.com. But no matter what you do, I'll see you right back here tomorrow for another daily dive into the news.