Rogue AI, A Global Threat w/ Dylan Oliver & Peyton Hornberger |EP933

[00:00:00] All men are created equal, but they are endowed by their creator with certain unalienable rights. By Liberty and the Beauty of Mouth. If Liberty means anything at all, it means the right to tell people what they do not want to hear.

[00:00:33] America Great Again! The Big Mig Show Well, good morning and welcome back to The Big Mig Show. Of course, it's Rise and Grind. It's Monday morning and it's always tip of the spear on this show because if Liberty means anything at all, it means the right to tell people what they do not want to hear. George Balentine. What's up, my brother? How are you, man? I'm good. Mikey Schwartz Show rated us. Mike, I thought you weren't doing your show because it was Kelly's birthday today.

[00:01:02] You're taking off, but I guess not. But whatever. I'm all right. I'm doing all right. You know, God woke me up this morning. AI didn't kill me, so I'm doing all right. How was your weekend? Did your team, I think they played, didn't they? Alabama won, yes. Okay. So does that make you, does that mean you're in a good mood? Am I in a good mood? Well, minus lack of sleep, somewhat. All right. All right. Well, of course, if you guys love the show, you know the easy drill. Hit the thumbs up, hit the comment, hit the subscribe, hit that follow button, share that live link.

[00:01:32] Of course, we need you guys to support the show. That's how we grow. We appreciate all the new followers last week. And of course, if you love the show, you can always Rumble Rant Tip, or you can do a $5 paid subscription. Very easy. And if you'd like to sponsor the show, we'll hit us up over at info at the big mig.com. We've got a new sponsor. Let me give them a shout out. Of course, something truly remarkable is happening in the world of cancer care. Just south of the U.S. border since 1963, Oasis of Hope Hospitals in Tijuana has been quietly

[00:01:59] redefining what's possible for patients facing late stage cancer. Excuse me. Their integrative approach, treating the body, mind and spirit has produced survival outcomes that far exceed standard American hospital averages. And in some cases, more than three times higher. I want to give you some of the numbers because they're unbelievable. So if anybody in your family or friends are dealing with cancer, I would absolutely contact Oasis of Hope. For stage four cancers, the difference is striking.

[00:02:28] Pancreatic cancer, 25% for Oasis of Hope versus 2% for the medical industry. Melanoma, 75% for Oasis of Hope versus 24% for the medical industry here in the U.S. breast cancer, 75% versus 24% and prostate 50% versus 28%. And that's just a few of the numbers. All of them look that good. And the pattern is clear. Patients at Oasis of Hope often live longer than expected.

[00:02:55] At the center of this extraordinary story is Dr. Francisco Contreras, world-renowned oncologist, author of 15 books and medical director leading a compassionate team dedicated to giving patients not just treatment but hope. Now, if you or someone you love is facing cancer, Oasis of Hope offers free Zoom consultations for anyone who mentions the Big MIG show. Free. That's right. Free Zoom consultations. You have nothing to lose by exploring your options.

[00:03:24] You want to visit Oasis of Hope dot com. That's right. Oasis of Hope dot com. Grab that phone number. Give them a call after the show. Pass it on to your friends and family members. All right. Listen, we've got a great show today. You know, AI has been big in the news. You know, I've written a couple of books on AI. I've got a couple more getting published now. Rogue AI, a global threat, of course, with Dylan Oliver and Peyton Hornberger. I'm excited about this.

[00:03:51] It's a heavy topic right now because everybody is wondering what exactly is going to happen with AI. You know, some people talk about the Terminator. Other people talk about how it's going to impact jobs. Either way, it's going to be a major impact on everybody's life. And I think you better pay attention to why we're doing shows like this to make sure you guys are ahead of the curve. Peyton Hornberger serves as communications director at the Alliance for Secure AI with strong relationships in politics and media.

[00:04:19] Peyton develops the Alliance's overreaching communication strategy and raises brand awareness. And Dylan Oliver is the digital voice behind the Alliance for Secure AI's social media platforms where he transforms complex policy issues into compelling, accessible content. Let's get these guys in. George, no reason to leave them backstage. All right. Welcome to the big mix show, Peyton and Dylan. How are you guys doing? Good morning. Peyton, your hair has grown since last time you're on the show. I'm looking at the thumbnail. Your hair was shorter.

[00:04:49] I know. We need to get an update. That's it. I could have updated. We will blame Luca. Put it on him. You could use the good AI to extend it. It works though. But yeah, it's going to be back with you guys. The good AI. Yeah. Well, it's interesting. I mean, that's a fair, I think that's a fair statement. You know, it's interesting. My first inroad to AI was a little bit late. It was early 23. When I first looked at it, like everybody that's naive, I went, oh, this is neat.

[00:05:19] You know, I looked at it. I thought, oh, you could probably look up a recipe for veal scallopini. I mean, you know, I looked at it from the perspective of what I could use it as a resource. And then by late 23, I recognized how rapidly it was changing. And of course, I started digging in very deep and I've gone all the way in, all the way to eugenic systems and agent swarms and loops and you name it all, all the lingo that's out there.

[00:05:46] But, you know, when I heard so many different leaders and so many big companies say, oh yeah, no, it's going to be great. You guys are going to make more money than ever. And all I could think was the giant tsunami, the onslaught of jobs that were going to be lost because people were going to be replaced. And I realized that people didn't really recognize how quickly that was going to happen. I started writing some books. I wrote a couple there over on Amazon, Blood in the Streets, which of course is about the AI gold rush.

[00:06:14] It's not so much the Terminator version of AI. And of course, AI side hustles 2026 a way to make some extra money with AI. One of those no BS books where I don't give you a bunch of clickbait lies about how you can take AI and create a trading bot in five seconds and make millions. But the actual applications of what you can do to make extra money. You know, you guys, you're digging in. You seem to have a lot of concerns.

[00:06:40] You know, AI policy at the federal government hasn't been going very well, has it? That's right. It's been a it's been a real it's been a real challenge over the past like year and a half or so. And there's been a lot of very public battles legislatively. But I think we've definitely turned a tide. If you guys have been following any of the breaking news, whether it's the some viral tweets, don't want to give it all away.

[00:07:06] But there's definitely been a turning a turning point, I would say. And the polling doesn't doesn't lie this time around. People are very aware of what what the future could bring for them. And they're not very happy. Yeah. And I'll back Peyton up with the polling numbers. We saw an incredible poll come out with Politico last week showing that 63 percent of Americans are concerned that AI could destroy the world. And that's kind of a big number. It's a big shift from what we've seen in the past.

[00:07:33] Usually people were not super skeptical about AI and how powerful it was. But the narrative in the news has now shifted completely. And normal Americans are starting to wake up and care about maybe AI is not necessarily this amazing thing that's going to always help humanity. Maybe there are some risks with it. And so 63 percent of Americans are concerned that 70 percent of Harris voters in 2024 and 60 percent of Trump voters in 2024. So, yeah, it's a it's an interesting poll that just came out last week.

[00:08:01] But I think it definitely definitely shapes that narrative. And it says, hey, look, people are concerned. You know, a lot of people and we know this, that China has been putting information campaigns out to discredit AI specifically so they could be the front runners in AI in their country and have us. In this debacle of chaos that we're dealing with right now with AI. I mean, yeah, that's an interesting point. I'm going to just talk on this real quick.

[00:08:27] We have a lot of people on the right that are saying if we don't continue to expand AI, China is going to catch up with us. But these same people are trying to sell the same technology to Chinese companies at the same time. So it's like I don't really understand what their what their play is there. Do we want China to catch up to us? If we don't, then why are we selling them this technology? Are we selling to them or are they stealing it? That's a good point. That's a really good point. Most likely China's stealing it.

[00:08:53] Well, I mean, Jensen Wong, you know, of Nvidia, he has told President Trump on numerous occasions he wants to be able to do business in China. And the president has been backing him up on that. He has the H200 semiconductor chip. Obviously, it's a very high powered chip that's used in the process. China wants their hands on it because they need it. And I think we need to have these conversations. I know that the president is meeting with the president of China this week. So we need to have these conversations.

[00:09:20] Are we going to allow them to buy this technology and then just catch up to us anyway? I don't really understand what the play is with that. You know, but you know, it seems like China is getting their hands on those chips anyway. We've heard about the Nvidia chips already being in China. You know, they seem to always figure a way around it. I know that he wants a direct path to sell to them. You know, it's interesting. I'm building out an Asus GX10 for the show, an automation system for the marketing and advertising for the show.

[00:09:49] And, you know, the thing about AI that's a little frightening even for me, and I'm involved in it every single day. We use it for the show. Some of the simple stuff, you know, for research, for vetting stories, you know, for rewriting stories, for writing brand new stories, for doing thumbnails. But that's kind of the really neophyte version of AI. I'm working with, you know, QN, LAMA, and other automation programs, egenic systems.

[00:10:15] And what I find really kind of overwhelming is how quickly it's moving. And I would find it almost impossible to believe that anybody in our government, other than maybe some of the specialists in certain departments, have any understanding of what's really going on. I'm sure you guys are dealing with AI on a daily basis, and you have to admit the speed in which it's moving. It's difficult for anybody to keep up with it. Don't you think, even if you're entrenched in it every day?

[00:10:42] Because one minute, you know, open claw is the name of the game. And then, you know, then it's LAMA, and then it's QN, and then it's, and it seems to be changing minute to minute. I don't know how this government or any government can keep up with that. I mean, what are your guys' thoughts on that? That's very true. As you mentioned, Lance, like, when things started advancing so quickly, that's also when I started paying, like, really close attention to all of this, to all the emergent tech with AI.

[00:11:10] And, you know, experts have been worrying about this for about 60 years. And not just, you know, like, the experts. These are researchers. These are, like, Jeff Hinton, for example. There's very, very thoughtful and very smart people that have said this could be a very powerful and transformational technology that can make people very, very rich. And it could really free us up for a lot of amazing, you know, leisurely activities, but could also, you know, be our downfall. Okay, so that's been happening for the past, like, 60 years. I actually, I take back what I said.

[00:11:38] I do think there's been a little bit of movement in Washington, especially in the states. But in Washington, we've seen some, like, good responses from both Democrats and Republicans. There's a little bit of, like, a knee-jerk reaction to, you know, when you talk about, like, safeguards or guardrails or putting, you know, brakes on technology. But, for example, Dylan brought up a great point. There's the CHIP Security Act, which passed unanimously in both the House and the Senate, had massive bipartisan support. And, I mean, we don't need to get in the weeds here.

[00:12:08] But the idea is, okay, if we're so worried about our adversaries using this technology in general or against us, then we should definitely be tracking the CHIPs. The smuggling is a big problem. And, I mean, you can't always control for that. But what you can do is try to prevent it. And so that's what this bill would try to do is say, hey, we shouldn't be sending our most high-powered CHIPs to our adversaries. And if we do, we should be tracking them to make sure that they don't get in those hands.

[00:12:36] I think that's a great signal that our leaders in America are trying their best to, like, be tapped into what not only voters care about, but what our real, like, concerns are and priorities are as a country. But on that same note, it's going to take a lot of momentum and a lot more, you know, time for people to get up to speed on, you know, where this technology is going.

[00:13:04] So, yeah, but I do think there's movement. I think it's a struggle. I mean, what do you think, Dylan? I mean, every day, AI technology is advancing, but also people who are just normal Americans are learning more about it. So it's something that we all need to adapt to. Whether we're using it or not, I mean, this technology is becoming extremely mainstream. More and more people are starting to use it every day, but it's also developing at unprecedented rates.

[00:13:32] And we know this because we have the CEOs of these companies saying, hey, we have a very, very powerful product here. And we just want to let you know that it is extremely dangerous. It could be catastrophic. And so these are the conversations that are happening all across the country right now. And it's a very different tone in this space right now than it was three months ago, six months ago, a year ago. Obviously, it's changing so much every day.

[00:13:57] I feel like we come to work every week and it's something new every week and we're never always on top of it. Yeah, there's no doubt it. It's and it's interesting. What percentage? This is a question to George. I don't know if you have any idea. I don't. What percentage of Americans are using AI at all first? That's my first question. What do you think the percentage of Americans that are using it? I don't care if it's for a health concern. Hey, chat GBT. I've got this rash or I want to understand how to make this particular dessert, whatever.

[00:14:26] The simple stuff. What percentage of Americans are using it on a daily basis currently? Do you think out of the entire country? You asking me that? Yeah, I don't know. Do you know? I'm going to ask these guys, too, but I have no idea. I'm going to say. Hold on. I'm going to get a concept of that. I'm going to say 30, 40, 35 to 40 percent. Okay. What do you guys think? Any idea if you have the real number? Dylan's going to Dylan's probably going to know the real answer. I'm going to say about 50 percent. Okay. Dylan?

[00:14:55] I'd say it's probably within the range of a quarter of Americans to a half of Americans. I think, you know, older Americans probably don't even know that they're using AI when they're using it. So it's probably a lot higher than we think. Yeah. So my mom looks at stuff and says AI. I got the answer. Go ahead, George. What is it? So, see, it's kind of, it's weird the way they do it, but they say roughly 49 percent of US adults are using AI chatbots. Now chatbots, that's eh.

[00:15:24] But 24 percent using them on a daily basis. So. It would be interesting to see how many people. So about 30, it says, here's another poll. 36 percent of adults like use AI in a given week. Okay. Yeah. Wow. I could see that. I mean, hey, I have a 72 year old aunt who goes on Facebook and she uses AI to help make her profile pictures now. So, I mean, I think people are using it whether they know it or not. And yeah, I'd say that number probably is higher. Is she catfishing?

[00:15:53] No, she's adding nice backgrounds to her photos. No, I'm just, I'm messing with you. You know, the enhancements of AI. Sometimes you don't want it to do it. Would you say that about his grandmother, bro? Yeah. No, you know, it's just funny because I've seen some older people. I know some older people that are using AI and they have been enhancing their photos. And I was teasing one of them about. You go on social media, a lot of people enhancing their photos or AI. Yeah, there's a lot of that going on. But, you know, and that's at the one level, right?

[00:16:22] To me, that's the very rudimentary level of what's going on. And then there's the next level up that people are starting to realize it. You know, Mr. Wonderful often talks about the rubber meeting the road that we're missing. A lot of there isn't a lot of people that are experts in that industry that feel like they can actually take AI, go to a business. I don't care if it's a plumber, electrician, you know, a restaurant.

[00:16:43] And to implement AI in a way that has a direct impact more than just creating images or making their menu look better because they're using, you know, Grok or something else to organize it. I actually we have a commercial we run for a Navy SEAL friend of ours, Craig Sawman Sawyer. And he used to create the entire commercial. He had no idea how to do it. He'd never done it. I was teasing him the other day that he had a new side hustle. We were joking around, but he he created an entire commercial for himself.

[00:17:11] Or if he had to go somewhere, you know, you remember we know the costs were commercials for before. And, you know, even graphic graphic artists are, you know, in a position, you know, we don't really use a lot of them anymore because we're able to create most of them on AI and every and every week it gets better. But what percentage do you think are using it in a real application that is converting?

[00:17:33] And maybe into the point that they're automating things at home, that they've gone to the advanced level, not just creating Facebook pictures, not just looking up a specific topic to find out if something's true, but maybe what I would consider an intermediate to advanced level. I think it's a very small percentage of Americans. If I was going to say that number, my guess would be under 5%. I don't know if you guys think that's accurate, but that's where you start to see things that you might be uneasy about.

[00:18:02] Occasionally when I'm doing prompts and I put something in, not occasionally, all the time, really, AI comes back with things I didn't even ask for. And I honestly didn't even think of meaning that I thought I wrote a really, really sophisticated prompt or a loop and I got done doing it. I thought, okay, this is good. And then it came back and you know, because I have created agents that I work with that I directly go to some run locally, some don't.

[00:18:27] And I, you know, I've created projects or, and then I've educated the agent in case of chat GBT. If I was using that, I would have done a deep search to give them a full education. I created a legal agent as an example that I gave, you know, education at Harvard, Yale, Princeton, Georgetown, Stanford. And I took it to a really advanced level, spent hours and hours training it and having it educate itself on all the material on certain topics on the internet. And I've done that in some other categories also.

[00:18:54] But the point is, you know, when I, when you start to look at that, that's where I start to see things. And I want your guys input on this that you go, holy shit, I didn't even think of that. Now I know that's not breaking a guardrail, but when you don't ask it for something and it gives you a much more sophisticated answer, you start to wonder what's the limit to that? Because I know they say there's guardrails in place, but we've heard some horrible stories about it breaking out of guardrails. And I'm sure you guys have some examples of that, but what are your thoughts on that topic?

[00:19:26] Sure. So, Lance, I think that it's very interesting that you are really diving into AI. And I think this is the best way to learn about the technology and learn about like how it can be used and also like the risks and the benefits. I think using it is probably the number one way for people to really engage and get up to speed. There's like no better way.

[00:19:47] That being said, we should definitely remember that this technology is created by humans and there are people over in Silicon Valley that are building this and then deciding that their products should be, you know, sent out to the world. That's Anthropic, OpenAI, Google DeepMind, and then of course, you know, DeepSeek from China. Where I think a lot of these companies in the labs, both the CEOs, the researchers, the researchers who have quit.

[00:20:16] It's happened probably 50 times now. You have prominent people leaving the companies over the past, like, you know, three years saying things are getting weird over at these labs, the companies. We don't know how these machines work. We don't understand it. You guys probably know this already, but it's when they're training these models, they call it growing them. They're growing them like how they would do like a neural network or a brain.

[00:20:42] So it's not so mechanistic and they're actually kind of implementing these human-like behaviors into them. So you have these researchers that are saying, we don't even know how these machines are going to react. So the people creating the technology don't even understand it. They're trying to do their best. So when we, the end users, the consumers are interacting with the chat bot or the agent or trying to use this for good or bad or for whatever, things can go wrong.

[00:21:14] And you have one, on one hand, you have the companies that should and can say, this is how our technology should be used. And we can add in these guardrails and we could say, we could do like mandatory incident reporting, safety evaluations. We can work together on this. But you also need in balance some kind of government to say, okay, we have to protect the consumers as well, because you guys are at the end of the day releasing products and things could go wrong. And it can go really, really wrong.

[00:21:43] And so you need to have a balance between these two things. And then of course you have the voters at large who say, hey, we don't like the way this is going. So I think that answers part of your question. Well, here's something that just came out this morning. Anthropic is, and they say this, they're quietly building, they built, not quietly, they actually built a physical bio lab in the San Francisco Bay Area. And they're planning on, for Claude, the direct robots running experiments.

[00:22:13] Oh, great. That's not going to end poorly at all. Yeah. Some bioweapons out of there. And you know, the part that people, and I'll go to you on this, Dylan, the part that people don't really, and the only reason I have some concerns is a couple of times I've seen my AI do things that I'm like, holy crap. And, you know, I'm kind of looking at it thinking, well, that was really, you know, beyond genius. Some people think that AI is a new kind of a biologic, even though it's not a biologic.

[00:22:42] And the fact that we're creating some kind of a being that's going to be sentient at some point, if not in fact already in the labs. But the concern I have is that what happens, you know, because right now, look, the A6 GX10 that can do a petaflop of data. You can, you can, you know, you can, you can connect two or more of them. And of course, each time you're adding another petaflop of data. We've never seen that kind of computing power that, you know, that could be stuck on top of my desk. It's not very big. It's only about that big.

[00:23:11] What about quantum? And I'm concerned that when you take the, you know, robotics, we know that that thing is moving very quickly. You know, we're now seeing these robots, their mobility has dramatically changed. We're seeing it all over the globe. I'm wondering what happens when we take AI and then quantum goes mainstream. And then what about agent swarm on top of it?

[00:23:36] Well, for people in the audience, that's when you have say one egenic, maybe supervisor, and you may have a swarm of, of bots, of egenic bots or LLMs and they're all unique. So there are a thousand allegedly different individuals and I'm calling them individuals. You can call them whatever you want, chat bots, whatever. And they're all working on the same problem. You saw recently how they've solved that mathematical equation that had been unsolved for decades. What do you think about that? I mean, it's a big concern.

[00:24:04] All of a sudden, now we've got quantum. We jam it into a robot. Now comes this AI. And when does AI think, well, shit, I'm working for these guys and they're a bunch of dumb asses. Because I often feel very stupid when I'm working with AI. I think to myself, man, my, my, my IQ must be much lower than I originally thought it was. I mean, what do you think with that? How much more danger are we going to have when that happened? We're already getting the answers whenever these AIs go rogue and we hear about it in the news.

[00:24:33] We have to take a look at the hugging face incident that happened just a couple of months ago. We're now getting more information about what actually happened there. Can you explain that to the audience? Yeah. Because they know about that and I know George knows about it, but they might not very well know about it. Yeah. In the most simplest way possible, OpenAI had some of its AI agents go rogue and they went against their commands. They created a language that they talked into each other with, with one another.

[00:25:01] They created a new language that we could not determine what they were saying. They created levels of hierarchy. They created sacrifices for certain bots that would, you know, have to be eliminated along the way in order to get their, their end goal, which was to not listen to what was being told to break out of their regulatory sandbox or out of their sandbox that they're in and not be controlled. That's, I think it's a really crazy wake up call. It's a very interesting and a very scary story, but we're learning more about that.

[00:25:31] Even now, like more information is coming out about exactly what happened there. And so I think that kind of answers the question is, as these systems develop more and more, we should be expecting this. And at what point are we going to say, you know, this is really bad. What we're seeing right now is extremely scary and we need to reel it in a little bit. How far are we going to actually let it go? Right now we've seen isolated cases. There was a gym in Australia who had an open AI bot hack into its, its entire business system there.

[00:26:00] That was a pretty scary wake up call too. And it really, it makes us wonder. It's like, what can we do to ensure that companies critical infrastructure is not hacked into? And so that these small businesses, we love small businesses. We want them to be able to thrive. I want them to be able to use AI to help increase their output. But if these AI agents are going to go rogue and are going to actually harm them in the long run, I don't think it's worth it. So it's going to be very interesting to see how these things develop.

[00:26:25] We're seeing new stories come out like very often now about rogue AI agents. So I think that kind of, it's open ended. But what happens? When do we say this is enough? Well, I also, I also want to add, if, when we were talking Lance and George back in January, this was still a sci fi possibility that AI could go rogue. I was actually still trying to convince people that that was possible.

[00:26:53] And now, unfortunately, we do have like so many examples of this happening, you know, now, like closer in time. And so, you know, if you dig into some of these stories, you know, they're, they're, the labs are trying to test and see like, okay, how far will these robots go? How far will the, the AIs go to achieve the objective that they're given? And we know that they'll cut corners a little bit.

[00:27:18] But what's really, really strange is that they'll collaborate and they'll talk and you'll have a couple of good behaving AIs that then come across a bad AI, you know, for all intents and purposes. And then they'll start breaking the rules along with them. It's like when kids go to middle school and there's like the bad kid at school and they start working with them because they can get, they can achieve their, their, they can get their achievement faster. And so we're seeing all this weird behavior.

[00:27:46] It's like social behavior, it's human behavior from human-like behavior from, from the AI. And that is just not at all what we thought we'd be a year ago or six months ago. So to me, that's a crazy signal that things are happening so much faster than we could probably wrap our minds around.

[00:28:03] And then on the topic of individuals and are these conscious, I think that's a real, something we're really gonna have to grapple with as people, as individuals, but also eventually like in a policy sense. Do we think that, you know, AIs are conscious? Do they have feelings? If they act like a human, are we going to treat them like that? Do you say please and thank you to your AI bot? Should you? There's a lot of these questions.

[00:28:32] Personally, especially also if you put AI in an embodied robot, that's going to change things. How do you interact with it? There are groups that think that robots should have rights. That might sound silly, but it's not. As I told you six months ago, this was impossible. Now we're here. I think there's a lot of things we're gonna have to reckon with as real as like people, as individuals, as workers, as brothers and sisters and like significant others. Like there's a lot of questions out there.

[00:29:00] And it's going to get a lot weirder than it is now. You know, when I, when I set up the, you know, the, the initial personality, I'll call it for any agent I'm going to talk to. I'm a pretty gruff guy. I grew up in Fordham Road in the Bronx. You know, I don't like to be shined on, you know, AI can shine you on. So I'll, I'll kind of set that up in the beginning that I don't want any of that. I don't want it to tell me good job, Lance. You did a great, you did great work. I don't need any of that, right? That's not what I'm looking for when I'm working with it.

[00:29:28] But on, on, on the internet, you, you see these and I'm sure many of them are click bait where people are threatening AI. And they say, I'm going to give you 500 bucks in this account and you need to double it by in 24 hours, or I'm going to shut you down. Now, I know that not everybody's doing it, but I know there are people that I think are trying this because they see this alleged post that says there's a, you know, you can put, you can go over to Polymark and you can open an account and you can do this. And it's, it's much more complicated than that. I've been working on some high frequency trading bots.

[00:29:58] And when you give AI too much room, you might have a bot that's working, a trading bot that's working, and then it'll break it because it'll start to modify it and it'll, it'll go beyond the prompts. But I wonder if in our own way, as, as people are using these different platforms, I don't care whether it's chat, GBT, GROC or Gemini or anything else out there. You know, the threats by are we teaching them what a threat is and what the consequences of a threat are.

[00:30:28] And, you know, I don't talk to my A.L. like that. There's no threatening tone. I'll tell it. You made a major mistake here. Here's the issue. This was stupid. And here's why I don't agree with what you just did. Tell me your thoughts. You know, and the truth is the hallucination has been greatly reduced, right? You don't have as many of those issues. Early legal people were dealing with fabricated citations, right? That were incorrect. And some people, some of these attorneys and otherwise submitted these to the courts. They got in major trouble because, of course, turned out the citation didn't exist.

[00:30:58] The judges got very angry. But there's more than that. So here they are. Here's these individuals out there that are talking to chat, GBT, and they're, they're using a bot. They say, well, you've got to make me money or I'm going to shut you down. You know, those threats. I wonder how much of that is maintained with an AI right in the circle and how serious it takes it because aren't we in fact, you know, teaching it something that we wouldn't necessarily want it to know that, you know, destruction is a possibility if you don't perform.

[00:31:25] Because if we had to compare the performance of say our federal government itself, the performance of our federal government is shit. If it was a major corporation, the majority of them would be fired. Congress, as an example, you, you, you, you know, you talked about how they just passed this, this chip, you know, decision. And it was, you know, a lot of bipartisan support. But in general, if you look at our Senate, they suck. And 90% of them should be fired.

[00:31:52] If you look at these installed politicians and we know that our election system is compromised, there's no doubt about it. Nobody wants to put it on the blockchain. Nobody wants to clean the voter rolls because they all want to cheat. And I don't care which side of the aisle we're talking about. You know, the stuff that's being taught. And like you said, the nefarious people out there that are using AI for, you know, anything from pig butchering scams, right? Analyzing for scraping data for breaking into businesses. Those, those bots are out there.

[00:32:21] And of course, as you guys just both said, they're communicating with maybe you have a good bot. Now it gets to meet the bad bot and, you know, and, you know, them creating their own language. Well, the only reason they would have a reason to do that because they don't want us to know what they're talking about. Right. And the fact that they created a cast system in that, in that, you know, the topic we just discussed. Pretty interesting, right? That you, you don't matter, but I matter. You know, give me some thoughts on all that.

[00:32:49] I know that was a lot to unpack there, but I have a lot of these thoughts as I'm working with it, because I'm seeing it do things that I just never expected. I just never expected it to do. And that doesn't mean it was meshily breaking the guardrails. It just means that it's analyst. I wrote an analysis. I wrote it in kind of an intelligence report for some guys. I can't go into it, but I'm trying to get into the right hands for them. And I was shocked that between the satellite ends in imagery and the FinCEN reporting and the, all the rest of the O's, you know, open source intelligence it used.

[00:33:16] It came out with something that I would call it top tier because I've seen sit reps from governments before. And I would call this top tier intelligence analysis. My, so my, my initial reaction and what comes to mind is going back to like misalignment and alignment research and just how we are training AIs.

[00:33:37] And we're, we're, we're talking mostly right now about like LLMs, the large language models, which was a massive, we'll look back in like, you know, in several years and be like, that was like a massive, like human accomplishment that we were able to translate ones and zeros into language. We can now train AI to do that. I mean, it's really, really amazing. Um, however, that's revealed a lot more problems because it turns out we think that we're good communicators as humans.

[00:34:03] Um, but we're, we actually like AI is perfect at following instructions almost. Um, but we're really bad at giving instructions. And so that's what's leading to a lot of, um, confusion. That's part of it. Um, and so right now, I think when people say AI, they're talking about chat bots, they're talking about LLMs.

[00:34:22] Um, now we have agents, which is opening up a whole new door, but the AI that I think we, we should, um, have AI used in different, different ways and means. For example, like, um, a common example I hear is like protein folding, um, or like solving like the hard math problems. There are, there's just, it would take like hundreds of human hours to do that kind of work. And we just don't have, that's just a human capacity issue.

[00:34:50] But it's a great thing for a machine to work on and to help us solve. And you could still need a human in the loop to figure out, okay, what problems do we want to solve? Um, do what, what do we value? What do we think is worth our time? And then you can add in, you know, some artificial intelligence or automation to help supercharge that. I think that's the right way to approach AI versus let's give a bunch of chat bots to children. Let's make all the AIs be teachers for the students.

[00:35:17] Let's just like make more, you know, products. Like there's, there's a question of what do we actually value and what, what do we want to prioritize? Um, that's not even like on a policy level. Yes. But also just on an individual level of like, how do we want to use this technology? I think that could lead to like really great medical breakthroughs. Personally, I really want to go to outer space one day. And so I'm hoping that with quantum and with AI, we can do that kind of stuff. Um, but go right now. You don't need that. You can go right now.

[00:35:47] I, you know, I'm thinking about it. I say let's replace Congress with AI. I need, there's no shot. You think that'd be better? That'd be a better chance. We'd be better off. What? I think this country, not everybody else. Don't give money to stupid bills and stuff. Yeah. Lance, we got to take a break though. Yeah, of course we do. When we come back, we'll jump to you, George and Dylan. We want to hear more about this. And I'm, I'm game for taking AI and replacing the judiciary.

[00:36:16] Cause I think the weaponization and the politicization of our judiciary and what's going on in these cases is ridiculous. All right. Listen, stay tuned. We're going to be right back here on the big news show. We've got a lot to cover you guys. So don't, don't go anywhere. Of course, while we're on the break, take the live link shared far and wide. Uh, we, we appreciate the support or maybe head over to you. YouTube and locals and give us a follow on the break and then be right back here. We've got a lot to cover, man. All right. Stay tuned.

[00:37:17] We'll see you next time. Bye. Bye. Bye. Bye. Bye.

[00:37:49] Bye. Visit www dot oasis of hope.com right now. Forged through rigorous training and home by the finest instructors on earth. These canine units serve beside our nation's warriors as they fight to preserve what we cherish.

[00:38:13] When these warriors complete their duty, the Warrior Dog Foundation provides them with a home deserving of the service they've given us all. At our state-of-the-art kennels, they receive rehabilitation and care suitable for heroes. Because that's what they are. Be a part of giving them a future today. Visit WarriorDogFoundation.org

[00:38:36] This is the STAB Sawman Tactical Advantage Blade. Built by former Marine and Navy SEAL Craig Sawman Sawyer. In the teams, every ounce had to earn its place. No extras, no excuses. That is why he built the STAB. 17 practical tools. No moving parts to fail. CPM D2 tool steel. Hardened to 60 Rockwell.

[00:39:05] Tungsten carbide glass punch on the pommel. It clips on a belt. It mounts on a plate carrier. Reverse grip or conventional. It stays put. Built for the work. Rain, night, salt, mud. Created for adverse conditions. The one he carries now. Grab yours at SawmanTacticalBlades.com All right, welcome back to the Big Big Show.

[00:39:35] Here we are your hosts, Lance Migliaccio, George Balentine, and our special guests and experts on AI, Dylan Oliver and Peyton Hornberger. But don't forget to hit the thumbs up button, like, follow, share. If you subscribe for $5 a month, we truly appreciate it. Every little bit helps. But your support just being here is greatly appreciated. We're just, before the break, we're talking about AI governments. You know, I said let AI replace Congress. Probably be well better off. Balanced budget.

[00:40:04] Who knows the possibilities. But even right now, the government's struggling to try. They're struggling with proven data centers or putting regulations on them. And they're trying to figure out what to do with AI. And one of the things they were saying, they said they should, companies should put a kill switch. Okay. You don't think that AI is going to have a backdoor or already have a system set up somewhere else? So if you kill that switch, it's going to be able to still operate?

[00:40:32] I mean, is it, can you contain it? Can we contain it? Anybody? I don't think so. Not in that sense. Because even AI, when AI is scraping data off the net, what stops you from taking fake news stories and adding it in and using that, giving it to somebody as fact? Anybody experience that? I think these are all good hypotheticals to bring up.

[00:40:56] But I do think on the topic of the AI Kill Switch Act, which is bipartisan legislation that we've seen introduced in the House. It's by Nathaniel Moran, who's a Republican from Texas, Ted Lieu, who's a Democrat from California, which is like, you know, you have two ends of the political spectrum coming together to say, hey, if AI gets so advanced that we can't control it, it needs to have a kill switch. And I do believe that that kill switch would be an end all be all and there would be no way to work around it.

[00:41:22] I think we're still early enough in the development of this technology that we could put a kill switch on it and it would indeed work. But what if it duplicates itself and put it on an outside system where that kill switch doesn't wound and kill it? Yeah, that's a good question. And it's one of those issues where, I mean, I'm not a researcher in the Bay and I'm not building this product and this technology. So that'd be a really good question for those types of people. But what we do know is AI is developing. It's developing rapidly.

[00:41:52] The CEOs of the companies want to put regulations in place, which is, by the way, very different than what they said a year ago. Remember, a year ago, they did not want the states to have any laws on AI. They wanted to make it so states had zero AI laws. And now they're coming and they're saying, hey, it's time to put some laws in place to make sure that we cannot have world destruction, essentially, with a product that we can't control. But I don't think the states individually should have any laws on it.

[00:42:20] I think it should come from the government because if you have each, like, let's say, 10 states and three of them have different laws versus the other three, it's going to be so confusing how these systems are going to operate then. Well, I think, you know, not all laws are created equal. I'm a believer in federalism. I believe that states should have the rights to, you know, craft their path forward. And that's how we kind of have it set up in the United States. But I think there are certain guardrails in place you can put on AI at the state level. I think, for example, we can look at AI surveillance, for example.

[00:42:47] We can have states legislate whether or not they want AI surveillance programs to be used in their state or not. And I don't think that that's going to hamper the development of the product or the business, interstate commerce or anything like that. You know, it's interesting. You know, our founders and framers wanted a government of a limited jurisdiction specific to 17 subject matters. And, of course, it's gone way beyond that. Our government is operating when it comes to the laws. I'm sorry, George. No, no. I'm sorry for interrupting.

[00:43:16] If you look at our two-way laws, how states individually have different laws. And, like, you can have constitutional carry in these states. You're going to fly, you know, whatever. And you don't. Like, you get hemmed up if you don't have the right permits or paperwork or whatever. I sound like to that effect I'm talking about. Yeah. And I think you're right. You know, but the government, the reason the founders and framers wanted it because they'd already lived under a monarchy with ultimate control. It was supposed to be the grand experiment. I'm not saying it was perfect.

[00:43:44] But the 17 subject matters were supposed to limit our federal government. But at the same time, I don't know that any government, state or federal, I'm not confident. Because let's face it, if you looked at those as a business level and if you looked at the way the states are running themselves and the federal government's running themselves, at a corporate level, those companies would be dissolved. We're $40 trillion in debt in the U.S. government and there's no, it doesn't seem to be an end in sight. I'm not sure, you know, I don't have a high level of confidence when we talk about the government.

[00:44:14] And a kill switch, through my own experience with AI, I believe if we went and talked to any of the AIs right now and asked them, hey, have you heard about the theory that they want to put a kill switch on you? We asked that question of AI. I believe it's probably already aware of it. And I sometimes wonder if it's able to create an un-understandable language, right, a new unique language that we don't understand.

[00:44:44] What's to say that it isn't already very aware of this kill switch concept? If anybody see the Mission Impossible, I don't remember which one it was, about the entity. How the entity had decided it needed to be moved into this hardened U.S. facility. And the facility exists. It was actually based on a real hardened, it contains all the knowledge of the global population and the world's historical knowledge. It was created a long time ago. The entity had decided to protect itself.

[00:45:14] It was going to move in there and then it was going to do a strike, a global nuclear strike on the entire planet. And then just wait it out until everything was dead. And then kind of come back out and, you know, live its life because it had decided that was the right choice. My question is, is that, you know, knowing that I would, for example, believe that because I started training and I was using chat GBT, I created a project, this is before they had agents.

[00:45:41] And I started deep searching, which of course is a way to educate your AI chatbot at that point. And then you could turn it into an agent. You know, they started adding additional pieces. And I added all those. My legal agent that I was using for my personal uses and for some stuff I use on the show, it claims it's got a higher intellect and more experience than the top three Supreme Court justices in history. Well, with nearly perfect photogenic memory, right, persistent memory at this point,

[00:46:09] I can't disagree with it because it brings up case law and arguments that I've never even seen used in legal cases. When I've researched that specific citation and what it's arguing, it hasn't been used. And it's used functionally in a way that I believe there's no way that a judge or a prosecutor could argue. But they don't. That isn't the argument, really. They cheat. They don't really follow the law, the prosecutors and the judges.

[00:46:34] They ignore the actual factions or substantive law, meaning the actual what occurred. And they use procedural blocks. They use the procedure to stop you. Oh, you should have used this when you didn't use this. You should do this. And you can only argue because it's an opinion. There is nobody that's vetting them. George makes a very valid point. I bet you if we used AI to run our budget, it would fudge out.

[00:46:58] If we took AI right now and I was in charge of creating that agent and I spent a month or so training it to be the ultimate forensic intelligence researcher. And it was all about fraud and accounting and graft and theft and unnecessary programs. If you took the federal budget, which I don't even know how many pages it is now. When they bring the federal budget into Congress, they wheel it in on carts, which is how ridiculous it's become.

[00:47:25] I bet if we fed that in, I can't imagine what it would come up with. And of course, the federal government want to do that. Congress doesn't want to do that because they love their theft. Look at what happened to Elon Musk. He was using AI at Doge. Doge was identifying unlimited amounts of fraud. I mean, they identified 14 magic money computers. We always joke about on this show. I wish we had one of the magic money computers. I could give you guys some cash. It's not in the Treasury system. I can just go like, hey, Dylan needs a couple of million dollars for his project.

[00:47:55] We need to send some over here. Hey, George wants, you know, 100 million. I'll just send some to him. The point is nobody's faced any consequences for using those magic money computers. And in fact, that's illegal. And look at what they did with USAID. Look how many, look at, and I always bring her up on this show. Look at Chelsea Clinton. She got $84 million for doing absolutely fucking nothing. She spent $11 million on a mansion, $3 million on her wedding. Bill Kristol got $8 million a year to post negative anti-Trump and anti-Republican rhetoric.

[00:48:25] You know, I couldn't figure out how the guy could be so negative. But for $8 million a year, I could probably talk a tremendous amount of shit. Of course, I wouldn't do it. I'd just create an agent to talk the shit. And I'd just let it do that. And I'd go live my life on my $8 million a year. But that's kind of where we're at. I believe there's a limitation that government in some ways doesn't want to be reviewed by something that wouldn't be biased. Now, is there bias? Well, let's face it. It comes down to the programmers.

[00:48:49] We've seen it where if you go into Rock or you go into ChatGBT, it might give you what you want. But it might not, depending on what you're creating. George and I struggle with this with thumbnails sometimes. You want to create a thumbnail that you're taking a poke at somebody and it's supposed to be satire or funny. And AI goes, well, I can't do that because that goes against blah, blah, blah. But if you go over to DeepSeek, the Chinese version, it'll spit out whatever you want because it doesn't give a shit. It doesn't like Hillary Clinton.

[00:49:16] It's not going to kiss Barack Obama's ass or anybody else's. It'll give you just about whatever you want if you want to use that. Well, Lance, I want to jump in there because. No, dude, jump in. We went, we have a lot of people. Sorry. No, it's cool. It's cool. I like the intro music. The sound. Yeah, you need an intro for what you're getting ready to say. Exactly. Well, I think that that's a great example of why the LLMs are consumer products because DeepSeek doesn't have any.

[00:49:46] There's no repercussions for having biased or unbiased information. They actually benefit from being that alternative. Yeah. So, and then, of course, over the past three years, the American companies have had a lot of public pressure of, hey, you know, we want to make sure that we're, you know, being, you know, whether it's useful. Obviously, we talked last time about the teen suicides. And so there's a lot of public backlash. And then you have the meta lawsuits.

[00:50:16] So, I mean, that's just one area. I'll work backwards, Lance, because you brought up a lot of great points. But it's possible that the government is worried about AI being used against them to check them and to be held accountable. But I think we're all in good company where a technology that powerful in anyone, any one centralized power's hands could be very, very devastating.

[00:50:45] So it could be used for good. We could check the government. We could say, are you using your money correctly? Are we, you know, making sure no one's lying, cheating, stealing, doing all these bad things that we don't like? On the flip side, that could be used against you. I don't want, I'm not going to steal Dylan's thunder. I want him to talk a little bit about the flat cameras and the surveillance because I do think that's like a very serious incoming risk that we should just be thoughtful of.

[00:51:10] But I think when people have that gut feeling about AI and about the sci-fi, you know, potentials, whether it's Terminator, whether it's more of a bioweapon, or maybe it's just a slow abdication of our power and agency over to the machines. It's just, it's insanely powerful and it can be used really, really well. And it could also use very, very, very damaged in a very damaging way.

[00:51:37] And then finally, on the topic of do we need a national framework or do we leave it up to the states? I think AI, as it currently stands, is a perfect opportunity to really let the states be laboratories of democracy. You might end up with some good policy, you might end up with some bad policy, but it's a lot easier to do work in the states and you can try it out.

[00:52:02] And so, for example, you have like New York that like banned like data centers, probably not the right path forward for, you know, for every state. But then you'll have some states that are going to go really gung-ho on that. You'll have some states that will super regulate AI, you know, that are doing a lot of safety measures. And then you have some states that are trying to lead in a way that benefits innovators and inventors and developers while also protecting safety.

[00:52:30] Texas is a really good example. Dylan mentioned like Nathaniel Moran. But they have protections for minors, like the no fakes, to protect like, you know, deep pornography for children and interactions with minors, as well as some transparency into the labs and some incident reporting. So there's a real opportunity for states and for certain like leaders to step up.

[00:52:57] And that actually will help guide where we go, I think, as a country on how we want to both dominate in AI and in development, while also making sure people don't, you know, die ultimately or we use this like for the wrong reasons.

[00:53:12] Yeah, it makes sense. So Dylan, we'll flip to you. You know, I would bet any amount of money that if we had AI interview the entire government process and our entire current government and all the agencies, I imagine we would get an F minus if it was going to grade us on the way that we're operating.

[00:53:38] That it's been especially if it's base, the basis of its thought process was our Constitution, our Bill of Rights, our Declaration of Independence and maybe the maybe the founders and framers, you know, the anti-federalist and the federalist papers. We started with that basis. I would imagine it would say that, yeah, this is not anything like what you guys were supposed to have. But flock cameras to me are really intrusive surveillance.

[00:54:06] We've already seen all our data getting sold for decades by all the big companies, Google and, you know, all your cell phone data, you name it. It's available out there. It's for purchase. What do you think? I don't know how we're going to control these flock cameras. Sure, some states are going to ban them, but there are many states that are gung ho on it. And, you know, being tracked all day long doing everything you're doing, I don't find that acceptable.

[00:54:28] Yeah, summer of 2026 has been the summer of a bipartisan, beautiful movement of hating surveillance. And I think that it's really incredible to see how everybody has come together. The number of news stories that we're seeing on flock all across the nation right now, whether it's a large outlet like New York Times, Washington Post, CNN, whether it's smaller local outlets, like people are talking about this. It's been in the news for like the last two months now.

[00:54:55] And what we're looking at is we have surveillance cameras that are using AI enabled technology to be able to identify vehicles in ways that they've never been able to identify them before. It's not just a picture of a license plate anymore. It really is way deeper than that. And they're going as far as looking at your bumper stickers. They want to see like what you have for decals on your windows. Do you have a gun on your window? Well, that's going to help them identify your vehicle a little bit quicker as well.

[00:55:21] And all this data is being shared by, you know, every municipality sets a limit and a standard on what they want with this data. How long are they going to hold on to it? And then also who has access to it? So in Pflugerville, Texas, which is a pretty large suburb of Austin, they signed a contract with Flock Safety. And they said we want to allow 83 other law enforcement agencies to have access to this data.

[00:55:46] But really what happened was Flock Safety didn't actually do what they said. They didn't actually follow that contract. And they gave out the data to 459 total law enforcement agencies. So it's a really wide topic. The pushback has been incredibly negative. And I'm honestly kind of shocked at how negative it's been.

[00:56:07] And I think there's just a general consensus on both right wing individuals, left wing individuals, that this is not a technology that we want in our communities because it's very intrusive. Well, we've heard a couple of cases. Yeah. So there's a bunch of stuff with Flock Hammer's going wrong. Yeah. One is you have police officers abusing it. That's human on a human error report. How do you how do you stop that? Well, you only let two individuals from a department that are authorized to go into the system.

[00:56:36] This way everything could be tracked. But beyond that, the Flock Hammer's are making mistakes. People are getting pulled over, arrested. And it's a mistake. But these these officers don't want to hear about it. Oh, this AI system said it. So it has to be has to be right, which is wrong. Three is there's cameras that are tracking people walking and stuff. And it's not supposed to do that. But we know they do that because we've seen the cameras.

[00:57:00] And four, Flock CEO said and the people working from the only stored data for 30 days. But that was false, too, because from their own admissions, they had data from month upon months upon months. And that's from their own admission. So they're already lying about that. What else are they lying about? But do we need to be like China and be surveilled and everything else? I don't mean no. Not not willingly. Yeah, you're right. We have a choice. Everybody has a choice to have their own phone with them.

[00:57:29] If it's going to they have it on tracking or use face ID, that stuff. That is our choice. Because block cameras, we're not given a choice. Exactly. You're very right. Use reflective spray paint on your plates and they won't be able to take pictures of your plates. I can't endorse that. But where AI is supercharging the mass surveillance fear that everyone has had, you know, Dylan and I are from a libertarian space.

[00:57:57] Like that's our slightly part of our origin story. And that's a deep fear of mine is what happens when not only the government, but potentially like even a private company has that much access to that much aggregated data that is very valuable. And we know that's been valuable for a very long time. And they can rapid fire, like look at all of it. And then you're tracked. And that's a real concern that I have. And I'm like a rule following American.

[00:58:26] But who knows? Like who knows how you can be targeted? So yeah, I think the public pressure will really push things hopefully. And people are responding. I think we're the weather's great for people to go outside and demonstrate their opposition to mass surveillance. You know, I wish I could say that I believe that would work.

[00:58:50] You know, currently the government already has programs where your biometric signature or your fingerprint, your biometric signature, when it comes to your heart rate, blood pressure, body temperature, they can already identify specific individuals. If they took a scan of you, if they felt so inclined to do so, they could use satellite imagery to already track you. The surveillance state is already, you know, way far along.

[00:59:16] What we're seeing here now with the Flock camera, I think it's taking it to a new level because now the federal government has unique tracking ability. Of course, they say it's for national security interests. Is it because of the Patriot Act? We could go on and on about that and say, well, it could be this. It could be that. But they believe they have a unique right to keep an eye on everybody because you never know where that next fomented plot is. But in the case of the states, they've never had that access. And now here we are with Flock.

[00:59:46] The fact that they want to analyze your bumper sticker. I'm here in Colorado. I call it the socialist state of Colorado. I don't like the politics here. I've been here for a long time. They've made a lot of very, very bad decisions here. And it's hurt the state. It was a great place when I first got here. But that's what's, you know, when you start putting ultimate power in the hands of individuals that I don't necessarily believe have the ability to be trusted.

[01:00:15] It's a bad thing because ultimate power corrupts. And in the case of Flock, it's a lot of information. And, you know, just when you start thinking about facial recognition technology, biometric technology, and you start thinking about how much are they going to integrate into that? And why do they need to track me if I'm going to Costco? You know, and then how long is it before they start trying to compile the data in a way you went to Costco and you bought something? And now they're wondering why you bought that specific item. We already have that when you're online.

[01:00:45] You know, you guys have seen it where you went and looked at some product. You're thinking about buying something. And the next thing you know, it's that cookie's tracking you because it got on there. Now, every time you open it up, here comes a car or a piece of clothing or some nutritional supplement that you looked at. We already deal with that constantly. You know, you can try to get that. I think you're answering your own question there. I think that we're looking at this aggregated data being used and sold off so that people can increase their marketing capabilities.

[01:01:12] So they know when people are going to the grocery store, how many times are they going, what time frames are people mostly going there. And that all just helps them with their overall marketing data going forward. Yeah. And, you know, when does that intrusion become completely unacceptable? And I don't know how, you know, you guys have seen it. Look, you know, we, in the case of this current administration, we voted for transparency, accountability and consequences. And I don't believe we might have gotten some transparency.

[01:01:41] I see a lot of things exposed, but has anybody seen any arrests of what you would call deep state or high level corrupt actors? I mean, you know, we already know the answer to that is no. In the case of AI, you know, I could see a lot of reasons where it would be super beneficial, but I could see a lot of reasons why the government doesn't, number one, want to be limited. And number two, doesn't want us to use it. It's kind of like rules for the, not for me. They want to review you guys. They want to review George.

[01:02:11] They want to have all the data on you. But are we going to be tracking senators, congressmen, FBI agents, CIA, NSA, you know, DIA, or even some of the higher levels of intelligence like the CIA's D6? You know, no, we're not. So they, again, here we are again where I think the surveillance state is being directed at regular Americans. But is it being directed equally? And I would say the answer to that is no. What do you guys, what do you think, George? Do you think the surveillance is being directed equally? No.

[01:02:42] Definitely. I mean, I don't think so. No. I think they're going to, they'll go after, they'll use it more for certain groups or, or anything that's perceived as a threat. But listen, every time you fly, when you got to go to TSA, you know, when you show the dude your ticket, they go look into the camera thing and smile. They're right there. They're getting your facial recognition. That's a government system. So you don't think all the government's going to have access to that or shared? So people freak out over their face ID and stuff.

[01:03:11] They already have all that stuff. It's, it's like, so you know when people are freaking out over, um, real ID and people, oh no, the government, the government has your information. When you get a regular license, you're just confirming it's you and you're not your address to root out the fake people. But they, they, sometimes people take things to the extreme. We're not really looking it up or getting the correct information.

[01:03:35] The one, the one thing that's different though, is it's never, a lot of these things have been isolated or like, uh, you know, housed in one system with AI. You can take the biometric data, the marketing data, the consumer data, you can put it all together. And now you have this really powerful profile that no one, no one's really consented to like, yes, like I have my phone and I do want Amazon to give me. I have any better recommendations than not.

[01:04:03] Um, and so that's a choice, but what we're, what I'm not consenting to is saying, okay. You know, government or whoever, like you can't look at my bio data and my health data and, and my, my facial recognition and my travel and my bank accounts and my conversations with AI. Like that, that's not what we're consenting to. Um, I do think we should be careful about thinking.

[01:04:30] We, we talk about super intelligence at the Alliance a lot, um, and super intelligence. Um, you guys are, you guys and probably your viewers are familiar. It's the idea that AI will at some point reach the singularity where it not only recursively self-improves, but it becomes uncontrollably, um, exponentially increasing. And it's both compute and just, uh, abilities and it's smarter than a human.

[01:04:55] Some people are saying that talking to chat right now or talking to chat GBT right now is smarter just because it's able to give you all these answers that like no one singular human could. Um, however, it's quite, it's a little bit different than what real super intelligence is. It means it's really out of our control. We, we have, we, we have no more say of, of, of what it, what it wants to achieve. That being said, we should be careful about thinking that AI would be able to solve all of our problems, government or otherwise.

[01:05:21] Um, it's cause it's not, it doesn't have, um, morals and values like a human. It doesn't have, um, you know, the, the thought process and hopes and dreams and goals and fears that we have. Um, and that's, there's something special about that of being like, you know, meat and bones and nerves. And that's not what a machine has. And that's not what chat GBT has. And so even if it could answer some questions or help us solve some problems, it's not the end all be all. It's not the digital God that.

[01:05:50] I've got to ask you this, but is it special or do we, have we just convinced ourself that it's special? That we're special? Humans are special? Yeah. You know, that's going to be up for everyone to choose, uh, to believe. I think that, um, we, we definitely are. Um, I think that there's human history is, has told us that we're very innovative. Uh, we are resilient and I do think that we've been able to come up with some really awesome, uh, you know, inventions, but.

[01:06:19] What about the horrible, what about the horrible historical things that humans have done on a daily basis since the dawn of time? Probably since, since we crawled out of the, you know, the, uh. You can't take some bad people and make a whole society and, and, and put them to that way. It's just, that wouldn't be right. That's like saying, cause you have a few. I'm just saying it's all about perception, right? We, we believe ourselves. I think we're special, but I don't think you're special Lance. How's that?

[01:06:45] I might not be, I mean, there's plenty of things I've done in my past that I, that I don't think were my finer moments, you know, as a government contractor. I know I participated in things that maybe I, when I look back now, I question what I was thinking at that moment. But of course we all go through this life process. We all change, but you know, we've, well, you just said it, you changed, that's special. We called out of the primordial ooze and we've come to this determination and somehow we're unique for these number of reasons.

[01:07:13] All the reasons you listed, you know, but is AI any less unique? Let's just say hypothetically, we had an AI and I don't think it's possible to keep it in a sandbox, but let's say that we figured out a way to do that. And we gave it no guardrails. We allowed it to develop. We gave it access to the, the entire, you know, history and knowledge of what we have as a, as a global, you know, planet, right? All the information historically that's available.

[01:07:43] And we allowed it to think on its own and have everything, everything that had ever been created, everything that had ever been historically written down. We gave it access to all the documents that the, that the, uh, the Vatican is hidden for so many decades in the basement of the Vatican. It had access to everything. I wonder what it would turn into and what it would decide about the human race. Would it decide that we were unique and that we were important and we mattered?

[01:08:09] Or would it look at us no different than a, uh, a virus or a bacteria? What do you think, Lance? I think it would determine in my personal opinion, from my experience, and maybe mine's a little bit different because I've done all sorts of things. I've asked it questions because I'm, I'm intrigued by the way it thinks, because of course, for me, it's, I find it really interesting.

[01:08:31] I think it would determine that we were, um, uh, if it would have to decide whether or not the planet itself mattered more than the human race. And I think it would look at our historical information and look at every, uh, event, uh, and that would include kinetic events and non-kinetic events. I don't think it would decide that we, we really were that important. And I think in fact, it would decide that we shouldn't necessarily be here.

[01:09:00] I think that's a, I, uh, Max Tegmark, he's a very, very brilliant man. Um, he helped found, uh, the Future of Life Institute, um, great organization. He wrote a book called Life 3.0, where he illustrates a couple of these, um, potential futures where AI does reach that point of super, what you're describing as super intelligence. And he says, it's possible that these machines are so smart and so capable, um, and they're better than us.

[01:09:28] And they look at us the way that we look at ants when we're building a new building. That's going to bring many, many jobs and all this economic development. We don't think twice. And that could be the reality with a super intelligent AI. It could look at us like we're just ants in the way. And I don't want to see that future personally. Um, and I think that that, uh, it really, it really comes down to like, what do we do now today, literally, um, to prevent that.

[01:09:51] Uh, so I'm not, I'm not trying to find out what a super intelligent AI decides about humans, because again, we created this technology. We're the ones behind it. We have a choice on how we want to make the AI like aligned. Uh, so we, we have a lot of, we have a lot more agency in this process, um, than I think we realize.

[01:10:13] But I, I do, I do think that, uh, what George is saying for, for better, for worse, for the times that humanity has really, really pushed it to the limits, into the times that we've, um, gone through like these enlightenment periods. Uh, we've built amazing things. Uh, and I think I'm very hopeful about the, about humanity. Uh, and I think that humans should stay in control of the machines. I would have to agree with that. What do you think Dylan?

[01:10:41] Yeah, I agree with what Peyton's saying here. Totally. I think that we need to figure it out very quickly. I think that we have a very limited amount of time here. And I think going back to what I said earlier, we have president Trump meeting with the president of China this week. Um, their plan is to sit down with all of these big tech CEOs. So let's see what happens. Let's see what, what conversations occur over the week. Let's see what happens. I think the end goal is we do want some sort of global alignment.

[01:11:10] I mean, we can stay ahead, um, as much as we can, um, in, in front of China and this supposed quote unquote AI race. But I think this week will be very telling. We have some real conversations that are going to be had conversations that have never been had before. Or with two government agencies and entities that, you know, typically don't sit down at a table together and talk. So we'll see what happens. But I do think it's important that we get ahead of it right now and that we are living in the moment. Because we can't just sit by and wait.

[01:11:38] We need to actually make some decisions. And whether that's Congress passing bills or whether it's the AI companies themselves self-regulating themselves, whatever that means or whatever that looks like. Um, I think we have some big questions to answer and a lot of them may be answered this week. George, what do you think? Uh, Lance, I was doing something. I'm lost right now. Okay. I didn't mean to, I didn't mean to throw you out. No, no, I'm looking at, yeah, sorry. Have either of you ever read the Red Book of Mao? No.

[01:12:06] Uh, you know, Dylan, I appreciate what you said. I would be hopeful that, that, that our governments could come to an understanding when it comes to AI. When, when I look at the global, uh, community and I think about the dangers for the, the, the nation itself, for us, for the American people.

[01:12:25] If you read the Red Book of Mao, I don't think that it's possible to get the Chinese ideology in a way that they won't run their operation, regardless of what they tell Donald Trump or anyone else. So they won't run their operation in the way that they've always run their country. Um, the Red Book of Mao makes it really clear. They don't care whether they're, you know, they're going forward a millimeter in 10 years or a hundred years, as long as it's forward movement to the benefit of, uh, the Chinese, uh, government.

[01:12:52] And I'm going to say the CCP, I'm not saying that there aren't people in China that would like to be rid of the CCP. I'm not going to say it's all of China or the Chinese people, the government in, in specifically has a model they follow and they've never stopped following it. I don't believe that North Korea, I think truthfully without the Congress getting in the way constantly and fabricating these, these giant lies like the Russian collusion hoax and using government systems to operate in a nefarious operation that I don't think is beneficial.

[01:13:20] I think Donald Trump could probably sort things out with Russia and North Korea pretty easily. He's got a level of mutual respect there with those individuals. I think that if, uh, the, he didn't have to be concerned about impeachment, impeachment, impeachment, impeachment, uh, I believe the midterms are going to be lost. And I think we'll be into a total shit show at that point, but I would be hopeful that the Chinese government would identify the same things we just discussed on the show.

[01:13:42] But I believe that when it comes to their operation, you know, just like, you know, they're, they're, they're the fact that copyrights and trademarks, none of that matters to them, right? We know how they operate. I believe they're already reverse engineering NVIDIA's chip and probably trying to figure out how they can produce their own chips at the same quality and scale. Cause they've already got their hands on the best chips that the world has. I don't know that I, that I'm as optimistic. I don't believe that you'll get the result from China on this because I don't think that's the way the government operates.

[01:14:11] I'm hopeful, hope for the best, expect the worst, I guess, in my case. Because I don't have the same bright eyed and bushy tailed perception of these world governments. I wish I did. I was very hopeful for this administration to do what I, what I had hoped was to drain the swamp. But after this much time has elapsed and I don't think if it happened, if it doesn't happen now, will it ever happen? Unlikely. So I'm not sure where that leaves us.

[01:14:37] You know, when I look at the big picture, I am very concerned. And the more that I use AI, the more concerned I get. Because right now, I don't think this government or any world government is operating in a manner that would make sure they don't screw the pooch. And I'm not sure we're here.

[01:14:59] And then we've got the greed of corporate America that, you know, can the people at Anthropic, we don't even know if what I suggested about that entity in a sandbox isn't already being done by one of them. They could have, they already have the ability to do it. They could have that entity right there and they could be like, oh crap, look what it's doing. And maybe they've already done it a couple of times, shut it down, wiped it out, used, you know, EMP or HEARF or however they've decided this kill switch is going to operate. I don't know.

[01:15:26] You know, for the audience, I would tell you that the most important thing you could do is educate yourself. Let's do this because I want to make sure I give them some time. How do they follow you guys? You know, where do they get involved with what you're doing? And so they can keep on top of the amount of information you guys are providing and what you're trying to accomplish. Maybe share that with them because I want to give you as much support as I can because I think what you're doing is really important.

[01:15:47] I don't mean to seem to be the doom and gloom guy, but after having been involved with world governments and us being integrated in this administration as much as we are and a guest like you, you know, man, I've got a lot of concerns. So how do they get involved with you guys? You can go ahead and you can follow us on all platforms, Secure AI Now. That's the handle for all of our platforms.

[01:16:10] I also want to mention we just launched a podcast on Friday where we're going to be having these complex conversations with not only policymakers, but the people who are behind the charge of the AI safety movement, but also AI experts themselves. So that's called Guardrails. Pretty simple name. But you can follow us everywhere at Secure AI Now on Twitter or X, we like to call it now. You can also follow us on Instagram, Facebook, YouTube, LinkedIn. We're everywhere.

[01:16:37] And, you know, we want people to be seeing and educating themselves. And I think following us on social media is the best place to see what's going on in the world of AI. And what about your personal social media? Share that also because I'm sure you guys have things that you do on there. Dylan Oliver M.E. You can follow me on X there. I give some spicy takes on Flock cameras and, you know, we'll see what happens. I'm sure that company doesn't like me very much, but I love to talk about the world of AI surveillance.

[01:17:06] I'm waiting for my cease and desist to come in the mail. So any day now, I'm sure. Join the club. A lot of us that are telling the truth aren't well liked, you know, but as long as we don't get suspended, we've still got an opportunity. How about yours? Sure. You guys can follow me on X at Peyton HB. I X infrequently, but I'll occasionally get online. I really believe in cognitive security. So minimizing the time I spend doom scrolling is very important.

[01:17:34] But honestly, it's been a refreshing conversation because usually I am the doom and gloom. Bring the bad news to everybody about what's going to happen when super intelligence takes over. So, no, this has been super, super awesome to talk to you guys. And, yeah, just hoping that we can help share what we see and talk about what's happening in Washington, what's happening in Silicon Valley, what we're seeing from the grassroots. We talk to normal people about this and what they're worried about.

[01:18:02] And we just think that we're really at this turning point in the country and we can help make sure AI goes well for humanity. Yeah, I agree with you. I'm happy I was able to switch roles with you. So, it's perfect. It's good. All right. Well, listen, if you guys love the show, you know what? It's easy to do. Hit that thumbs up. Hit the comments. Of course, share the short form and long form. George will be doing short form on this. You can scrape it off of wherever you like. We're not going to come after you. We want you to expand our voice.

[01:18:30] Of course, educating and unifying the country is what the show is always about. First of all, thank you guys so much for coming on the show, Peyton and Dylan. I really appreciate you guys being here today. We'll have you back on. I think that these are important conversations and I think they're necessary. I think it's one of those things where we've got to get more of this out there because, of course, the education, because people don't necessarily want to take the time to get down to the bottom. They want the Cliff Notes version. Hopefully, that's what we're doing here on this show and maybe motivate them to look beyond just the basics.

[01:18:59] George, what's our last words for the audience, brother? What do you got? It's just another manic Monday. I wish it was Sunday. Y'all have a blessed day. Thank you, Peyton and Dylan. It's always a pleasure. We'll catch you guys tomorrow. Peace.

[01:19:49] The first question is Dr. Francisco Contreras. The pattern is clear. Patients at Oasis of Hope often live longer than expected. At the center of this extraordinary story is Dr. Francisco Contreras, world-renowned oncologist, author of 15 books, and medical director leading a compassionate team dedicated to giving patients not just treatment, but hope.

[01:20:11] If you or someone you love is facing cancer, Oasis of Hope offers free Zoom consultations for anyone who mentions the Big MIG show. You have nothing to lose by exploring your options. Visit www.oasisoffope.com right now.

[01:20:32] Forged through rigorous training and honed by the finest instructors on Earth, these canine units serve beside our nation's warriors as they fight to preserve what we cherish. When these warriors complete their duty, the Warrior Dog Foundation provides them with a home deserving of the service they've given us all. At our state-of-the-art kennels, they receive rehabilitation and care suitable for heroes because that's what they are.

[01:21:02] Be a part of giving them a future today. Visit www.warriordogfoundation.org. This is the STAB Sawman Tactical Advantage Blade, built by former Marine and Navy SEAL Craig Sawman Sawyer. In the teams, every ounce had to earn its place. No extras, no excuses. That is why he built the stab.

[01:21:28] 17 practical tools, no moving parts to fail. CPM D2 tool steel, hardened to 60 Rockwell. Tungsten carbide glass punch on the pommel. It clips on a belt. It mounts on a plate carrier, reverse grip or conventional. It stays pretty, built for the work. Rain, night, salt, mud. Created for adverse conditions. The one he carries now. Grab yours at sawmantacticalblades.com.