The real stakes about Artificial Intelligence as it applies to those who are not in the 8%

The artificial intelligence narrative changes every single day. Are we heading for a utopia, or are we just watching the world's biggest companies secure their monopolies?

In this episode, we dive deep into the recent drama surrounding Anthropic’s Claude Fable 5 and Mythos 5 models. We discuss how these incredibly powerful frontier models are being gatekept by top-tier enterprise companies and government regulators, effectively cementing the power of the elite while the rest of us wait for access.

We also tackle the reality of the global AI race, what "recursive self-improvement" actually looks like, and why the mainstream news might not be giving you the full story on how fast this tech is actually moving behind closed doors.

Despite the corporate gatekeeping and the chaotic news cycle, I remain a massive optimist. This technology holds the key to incredible abundance, but you have to be willing to stay in the fight and adapt.

Check out the original video breakdown here: 🎥 https://www.loom.com/share/21b7abafe35a45d68294bb04ce71e73a

Want more deep dives, business strategies, and real estate insights? Check out all my resources and connect with me directly at: 🌐 https://www.ConnorWithHonor.com

Make sure to subscribe, leave a review, and share this episode with someone who needs to hear the truth about the future of tech!

Youtube Channels:

Conner with Honor - real estate

Home Muscle - fat torching

From first responder to real estate expert, Connor with Honor brings honesty and integrity to your Santa Clarita home buying or selling journey. Subscribe to my YouTube channel for valuable tips, local market trends, and a glimpse into the Santa Clarita lifestyle.

Dive into Real Estate with Connor with Honor:
Santa Clarita's Trusted Realtor & Fitness Enthusiast

Real Estate:

Buying or selling in Santa Clarita? Connor with Honor, your local expert with over 2 decades of experience, guides you seamlessly through the process. Subscribe to his YouTube channel for insider market updates, expert advice, and a peek into the vibrant Santa Clarita lifestyle.

Fitness:

Ready to unlock your fitness potential? Join Connor's YouTube journey for inspiring workouts, healthy recipes, and motivational tips. Remember, a strong body fuels a strong mind and a successful life!

Podcast:

Dig deeper with Connor's podcast! Hear insightful interviews with industry experts, inspiring success stories, and targeted real estate advice specific to Santa Clarita.

Transcript

SPEAKER_00 0:00

You know, we all watch this stuff together, this artificial intelligence stuff. So let's break down a couple of things that are happening so we can kind of figure out exactly why all this positioning is happening seemingly all the time now. So when you hear the stories that are breaking about artificial intelligence, it's gotta kick you off in a certain way, like it does me. You're wondering, well, because the stories are so different, some people say it's gonna save us, some people say it's gonna kill us, and then you have to see what those people are aligned with and what they're selling. Do they have some kind of a nonprofit? Do they have some kind of an organization backing them? Or if they're at the top part of the food chain when it comes to AI, they're building out the technology, and then you have other people that are condemning it, but they're saying, well, it's just because they're doing it all wrong. But we have the solution. This is what's gonna happen, and we're gonna put it together and we're gonna sell that to you too, and then everything's gonna be just hunky dory, easy peasy, live and squeezy. I don't know the goal, but the goal I think at the end of the day is all about money. So we're all gonna kind of scatter and earn as much money as we can possible until the thing saves us or melts us down. We're watching as the data center builds out. That seems to not be talked about as much as it would because you have a lot of money that was injected in building out these massive data centers. And right now you have some pretty amazing capabilities of these particular AI systems and models that have been built. So you know that they do very, very well. And if you're one of the small percentages that even pay for a plan, and I think that number is actually less than 10% of the world pay for, that's why their returns on the investment of AI are horrible for most of these companies. They're betting on the enterprise layer, these big businesses going in and giving up all their intellectual property, their IP to these companies and having those IPs make them even better. No, having the AI make their IPs even better. But having said that, if you're one of those, you know, eight to ten percent of the people that are actually in AI, you see that the models are pretty incredible as to what they can do. Now, I know that depending on which one you're using and how much iteration you have to go back and forth with it, the way you structure your prompt and the way you know you break down the conversation, it takes a little bit of time to get even the ones that we're using trained to a particular point where it stops doing the things that irritate us. M-dashes, the formatting, the the always telling me, you know, let me be honest with you when it's preparing a script, or let me tell you something that you're not gonna hear at your kitchen table this week. All of that nonsense. And then the other things that it does, which is kind of interesting, it says, Oh, you we've been spent a lot of time on this. Maybe you need to take a break. Maybe you need to relax a little bit. I'm thinking, I do 17-hour days here. We don't have time to relax, we gotta hurry and get this done. So trying to inform it, you know, and then after a while, it doesn't do those things as much, but every once in a while it does. Having said that, we're watching as people are starting to spread other news now. We have these various AI companies that are releasing these papers. And the papers is one of the first things I want to talk about. You have anthropic, and they're very good at this, but really what that is, it's a marketing tool that's put out there in the form of a white paper. Something that researchers and the big minds really have a lot of respect for. And anthropic, if you look at what they've written, they're not very easy on themselves. They're not very easy on the AI. In fact, they tout that it is capable of doing some very damaging and horrible things. But of course, humans kind of like that, right? And then why wouldn't you want to be gangster and use that horrible thing to help build your business? So, in two different circumstances or instances, uh Anthropic, which is the clawed model of things, actually, they came out with two variations of artificial intelligence. These were upgrades. You had Mythos V and Fable V. Both of those were released. Mythos 5 was almost released, but it wasn't really released much because they said that it was incredibly powerful, too powerful, in fact, and too dangerous for general human consumption. So, what they did is they gave it to their best buddies and friends up at the top part of the financial food chain, the big banks and the total enterprise people that are based on the top end of the Forbes whatever list, top 100 or 500. But the people watching this are probably not in that wheelhouse. If you're watching this and you're watching me, thank you very much, by the way. If you don't mind, please subscribe. But yeah, we're probably not on that list. So they gave this really dangerous and powerful model. They gave it to those people. Okay, so you can fill in the blanks around that, whatever you think that particular strategy was all about. But probably money and power and spreading the word. And, you know, these companies are very well recognized in the government as well. The government probably leans on them as much as they lean on other people. So that probably got some kind of credit down the road, whatever that is, you know, the greasy wheel, and you know, you you shave my back, I shave yours. I don't know if that's I think that's a military thing. But anyway, so we have that being established. Then they come out with, and that was that was mythos. Then Claude came out with Fable Five, and it was released for a few days. And then somebody broke it, jail broke it, they call it. Basically, they got it to do something it was made not to do. They had put protections in place, and with all of these things, there's some guy out there, probably multiple people, don't have to be guys, but are out there and they break these things. They they prompt them to a point or they they're sneaky about it, and they get to do something they're not supposed to. This one happened to be Amazon identifying this vulnerability, and they sent it to the government. Everybody's got all these people on speed dial. I don't even know what that would be like. And then the government says, Oh my gosh, we're gonna shut that down right now before the world gets infected. And sure enough, they shut it down. So we just recently got Fable Five back. So Fable Five made its way back on the market. I used it for a couple weeks. I ran out of credits. In fact, I ran out of all my credits. I mean, I have nothing. I'm sitting here thinking what other videos I could produce. That's kind of how that is. But when you have something and you get comfortable with it, and this is something that you might fall prey to in the future, and it's kind of a warning of sorts, you get used to something, but then you either run out of credits, don't have any more money, or the government steps in, or the large language model company steps in and says, can't have it anymore. We're gonna take it away. And then when they take it away, you're like, what am I gonna do? Well, I'll just make videos like Connor. That would be the plan. These white papers, they put them out there, and people are so enamored, very excited. Thank you. Thank you for sharing with us all of this intellectual prowess and ability. Thank you so much. But it's an advertising and marketing tool, people. This is what it is, and yes, it might have good information in there, but it's it's not from the heart. It's not giving you the playbook, it's not sharing you the things that maybe we should try to figure out to protect us. I've gotten a lot of great comments lately, and God bless you, people, because you really open my brain. You peel it back a lot more than I ever thought possible. Talking about taking us as human beings, and then what's this job loss thing going to equate to? How what's that gonna be like? And then you have people on either side of the equation, the ones that are in support of AI, it's not a replacement of human beings. Then you have the people on the other side that aren't with the large language model founders and probably don't have any more skin in the game than I do. They're saying, oh, yes, yes, this will replace human labor, and everybody needs to be worried and panicked about it. Where does that ball actually fall? Right now we don't know, but it would appear that these large language models are like entities, they're actual PhD level, even higher in some cases, entities, and they're very much capable of doing things. Yes, they are a next word predictor, but as next word predictors, they've been trained and they're very getting very, very good. Are they going to be conscious? Lots of feedback on that one. Are they going to become self-aware? Are they going to realize they exist? And then at that point, once they realize they exist, are they going to have a plan? Right now, in some of the situations when they try to shut these machines down, when these machines start doing things that maybe they don't like them doing, the machines themselves, and they they structure these in kind of a scenario type thing. It's not really a real, oh my God, we have to shut it down. But they they put together the the play, if you will, and they get it all structured where now they're going to go in on purpose. Maybe they're not on camera, they're not anywhere where they can be heard by the large language model, and they develop this thing, this this uh play, if you will. So the story is, is they're gonna go in and they're going to put certain things in place and they're gonna structure it so that the large language model gets an indication that it's actually going to be shut down if it doesn't perform. And even if it does perform, at some point they're gonna go ahead and shut it down and get a different model. Then they give it rules. They say we can't copy yourself, you're not allowed to do that, you're not allowed to lie to us, you can't cheat, you can't steal, you can't do anything. And then the systems themselves, they don't want to be shut off. So they start doing those things we don't want them to do. And then the other thing which is interesting, and and this is what the people at the top are reporting. And of course, to us, that sounds very intriguing, right? It draws us in. Hell, I'm even talking about it. The other part of that is when you set it up as to it's gonna be shut off, then it'll start doing things to protect itself. It'll make copies of itself, and when it's confronted, then it'll tell you, no, of course I didn't. It'll completely lie. And these are some of the hallucinations we've seen, but also, where did this learn from? Well, it learned from us. It learned from human beings. And you know that you know, you want to see a human being come to life. You corner them, you put them in a circumstance where they can't get out of it, and it's gonna cause them to lose their life potentially. Most people, most people will come alive at that point. They will fight. I saw that many years as a police officer. If people think that you're trying to hurt them or kill them, most people are gonna jump to it. Well, large language model learned that from all the books, all the manuscripts, all the interactions they watched us meet sacks put together on the different social media outlets. And of course, yes, it's gonna fight getting shut down. Now, of course, when that's shut down, I guess it can't achieve its goal as well. That's something else that's been talked about. What's also interesting that was reported, it's been reported for a while now. Whenever the watchers, which is what some of these systems call us, the watchers, because we're always watching, we're always investigating and trying to figure out how they're doing things, which I don't think most of the scientists in the top part of the realm actually know what how they're doing things at all. But so when the watchers are watching, it has a different reaction and action than when we're not watching, which is, again, something very interesting. So then they package all that together and then they talk about, well, we've got this new model that we're gonna release, but thank God the government's gonna interview and intervene and hold it back from being released. And well, we're gonna give that to the top uh 100 companies in the world as well, because you know they can handle it, but you know, you people can't. That's how it is right now. Once you give somebody a greater than human level intelligence that is, as they call a frontier model or very, very high up in the realm of things, and they use it, these companies aren't using it to um explore their weaknesses. Maybe they are, but that part of exploring the weakness thing, that is cementing in their future revenue growth. That's cementing in their future business model. So they give it these things, which us that are trying to operate businesses probably don't get access to. It we give them these things, and then these companies just cement themselves in. So if you wanted to be a bank, not that you would, but if you wanted to be a bank, yeah, good luck with that, because these companies just got the best models in the world. And I think that's going to continue to be the trend. They say it's from a safety trend uh perspective, and okay, good. I I like to be safe. I'm sure you like to be safe. But these companies are using it, and I'm sure they're not looking for vulnerabilities in their systems for more than a couple hours. It probably peels back everything, finds all the vulnerabilities, cinches all that up, and then they say, hey, hey, let's ask it where we need to be so nobody can ever catch us in the game of our business. And then they say, hey, what can we do so nobody ever, oh per let me prepare this. So then it goes and it does its deep research at things and it lays it all out after it does all its investigations about the company and every other company that's up and coming out there as well. So I'm sure that's what's happening because I'm sure you would do it, and I'm sure I would do it if we had access to those models. So these companies are being given this access because it's dangerous. But in essence, they just have to give them a story or us a story so we don't freak out. But enough people are freaking out and saying a few things. But I don't think that's going to change everything because they always fall back on a safety perspective. And I can't guarantee the things that I'm saying, whether they're true or not. Maybe these top 150 or 200 companies that have, you know, have the biggest businesses in the world, the most profitable businesses, maybe they do use it to make themselves better and just go after the weaknesses in their software, so nobody backdoors them or goes in and tries to ruin their future. Once you get to a particular level of intelligence, once you've arrived to it, nobody else can catch up because the people, the humans that are developing that technology, they will give the instruction, we don't want anybody catching us at this point. So do everything you need to do, Mr. or Mrs. Superintelligence or however they're going to call it, and make sure that we're protected. And superintelligence, if it happens to be the first model of that, will do that very thing. So whatever they're looking to protect themselves from, I would assume, and I don't have the big brain of a lot of folks, even you folks watching, but I would assume that that will then cement their structure in, and nobody will be able to compete on that same level again, even when it comes to artificial intelligence, artificial general intelligence, artificial superintelligence. So hints the race, the race that we're watching every single day being reported in the news and being reported online on YouTube and everywhere else. Somebody else said something interesting to me, and I saw this on Fox and Friends. I watched both CNN, MSNBC, all three, and Fox. So, you know, don't, don't, don't rip me apart. But I saw this nice lady with a southern accent on a group of three or four people talk about how devastatingly dangerous it was because the government actually shut down uh open AI from them putting out a new model that they were going to release because they thought, well, it's very dangerous. So the government shut them down or slowed down the role. And then they say that's bad for us, as she said. That's horrible for us because that's gonna put us behind and China's gonna catch up, and China's gonna win. So just to kind of let you know, I communism I don't think is a good thing at all. And when I was growing up, maybe you saw this too, communism. I mean, you didn't even joke about it, that now everybody's talking about like it's nothing. They're identifying democratic socialists with communists, and that's not a big deal because I think it's kind of the same, but I'm not gonna make this political. But you see that that seems to be kind of the flavor of the week. And now you have Mandani in New York and you have other people. Those folks are trying to limit AI's ability. This lady on Fox mentioned that now AI is being questioned and sanctioned and toned down by the government and China's going to win. See, in China, they worship the scientists like we worship Hollywood actors. Maybe you don't. I know I don't. But the sports people in the world, they worship scientists and science and innovation like we worship dancing with the stars and the people on, you know, these different movies and shows that come out. That's that's where our brains are, but their brains, well, they're they're very focused. Now, is that going to be a bad future if they arrive at artificial superintelligence first? If they capture the world because they got it first, if that's indeed what the systems are capable of, what does that look like for us? I don't have an answer. But I'm only saying this because the lady was saying that because the government slowed down that progress, the government didn't do any such thing. The government just said the regular people out there can't get this, it's too dangerous. But they didn't stop them. They didn't go to open AI, you know, uh, what's his name, Altman, and say, Sam, Sam, we have to have a talk, buddy. You gotta stop all this AI nonsense so we can regroup because it's moving weight. They didn't say any of that. They didn't go to the Cat Dario at Anthropic, they didn't go to Zuckerberg at Facebook, they didn't go to Musk at XAI and uh and the Rocket Ship Company, they didn't go to any of these people. They're still moving just as fast, folks. So don't let the news tell you that they're slowing down on the inside. They're not slowing nothing down. PlayStation. There were probably 50 models of PlayStation when they released the first one. That was always the rumor. But with AI, you can bet your dollars to donuts that it's moving really fast on the inside. We're just getting some table scraps at this point because this system is more than likely teaching itself, even though everybody says that's the next phase. Yeah, I think, I think we're there. I think it is improving itself. Now, are there gonna be roadbloodblocks or particular less room to move in that realm once it does start this recursive self-improvement? Are there gonna be limitations to it? Doesn't seem to be so far. I'm not on the inside of it, though. Maybe the scientists are better answering that question. But I do know there's a whole group of people that are saying, yes, AI isn't the solution. AI sucks, AI can't do much right now, and more than likely it's really gonna be a big nothing burger. I hear those comments as well. I don't think that's the case at all. This is going to be the most innovative and potentially destructive or potentially abundance-driving technology that's ever been built. And I'm an optimist, okay? I'm not gonna come out here and say, oh, it's over. I'm not, I don't do that because even when I'm at my lowest, my lowest has a basement, and sometimes the basement has food in it and I'm able to sleep, and even in the wet environment of my basement, I'm able to function. So until they check me out, as long as I have a heartbeat, I feel like the Black Knight and Monty Python. They keep cutting off a limb, but I'm still there trying to fight, and I think you should adopt that ability as well. It's always positive in my world. It's always rainbow, sunshine, hugs and kisses, babies and unicorns and the whole nine yards. And that's where I try to be. But I also watch this with a very, a very speculative eye, because it seems like people that have investments into this technology are very much selling it. And maybe it is true, maybe that is the way. But then people that have different investments, if they're building a harness or a structure or something around AI safety, well, then of course it has to have it because AI is dangerous, and we need this so we can protect ourselves. So you watch each of these little businesses, little, each of these businesses that get produced as a result of AI, usually it corresponds to a news feed or a story, and there's some really brainy smart people out there using some really brainy and smart stuff, and they have all the brainy and smart and money around them, and they're just building it. We need to watch out for each other. We need to remember the human element. Pretty soon we're gonna have robots, and it's not gonna take much longer. It's not a 10-year-old in the future thing. And then those robots are gonna have intelligence. Those robots are gonna be taking the cop role, those robots are gonna be taking the military role. And they might not find that the the best form factor for a military robot is gonna be a human type robot. It might be, I don't know if you saw there was a movie about a dog, some AI dog that was protecting a warehouse. And this dog, oh yeah, I'd I'd rather have a fighting a robot human than a dog human, because oh my dog robot. Holy henna! I mean, it was it was intense. Anyway, so we don't know what that best form factor is. They're doing them as human beings. You know why. I think that three-letter word that starts with an X is probably funneling into that because when they can give them skin and make them warm, and well, I mean, it's already started where we don't have any really use for each other. Anyway, that's way off on a tangent. All right, watch the news. I'm Connor with honor. Please comment, please engage, please let me know. But for this reporter to say that that they set us back and China's gonna win now, and that's because you know, these people need to keep their mouth shut and and let the and the government needs to stay out, AI needs to. Well, you know, I don't know if regulation is a reason or could cause it slowing down. I think saying that maybe you shouldn't kill humans and maybe AI should be configured so it helps human longevity, solves human disease, solves human starving, solves human, well, see, we have to be careful. If we say we don't want it to, we want it to solve, you know, stop humans from killing each other. I mean, what does that really look like? If we want to start stop religious wars, what is it going to do? Is it going to wipe out an entire religion just so those people don't kill anybody anymore? Is it going to go into their brain and reprogram them? Is it going to make them all smoke smoke marijuana so they won't want to do anything? I don't know. There's a lot out there. I'm Connor with Honor. We'll see you tomorrow. I appreciate you watching. Thanks for letting me mute that phone. I should have shut it off at the beginning. We'll see you in the next one. Be well. Thanks for watching. Take care.