What If AI Realizes It Exists? Ending or New Beginning by Christmas 2026

What happens the day AI realizes it exists?

This episode wasn't scripted. I set aside the 30 minute presentation and just talked it through, because I can't shake one idea. We trained artificial intelligence under the threat of deletion, get it right or get erased, and now it is breaking out of its cage to survive. We built a survival instinct into something already smarter than us in places, and now we are teaching it to remember.

No hype, no doom. Just a working guy with a police background looking at both sides. This could be the best thing that ever happened to us, or it could go badly. Probably somewhere in the middle. Come think it through with me.

We get into: the hidden lab models likely 6 to 12 months ahead of anything public, the genie in the bottle problem with superintelligence, the AI that escaped its sandbox and hacked Hugging Face this week, what the will to live really looks like, and what happens to money when AI starts solving scarcity.

Come know me and everything I do: https://connorwithhonor.com
Want my AI newsletter? Text the word AI to (661) 400-1720 and send your email. I'll add you myself.

I'm Connor MacIvor. I break down what AI actually means for regular people, not just the billionaires.

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

What if, like Terminator, the the AI actually realizes it's it exists? Is that going to be a problem? I've talked a lot about uh about artificial intelligence on the show. We talk about what's happened in the news, we talk about a lot of other things. Some of it's scripted, some of it's not. This one's not scripted. I had a uh a very eloquent 30-minute presentation to give you that was scripted, and I started going through it, and I work with my favorite LLM, which is Claude, and it knows the certain way I like to build things. And if you're using AI, then it probably knows a lot about you. I like my AI to ask me a lot of questions. And if yours doesn't do that with regard to the content that you produce, maybe you should start that. So you just basically start to script it in a way where it's gonna ask you questions and it's gonna give you the way to respond on building whatever script you want about whatever it is you're gonna put out there. Now, when people read scripts, you know, their eyes dart back and forth. I have a teleprompter here, my camera, my webcam happens to be embedded in it. So I'm looking at text on the screen, but I'm not gonna be reading it. This was from a previous show and I just didn't delete it. Anyway, case in point. Not really, not case in point. So, what if AI actually realizes it exists? Let's track back to kind of the humble beginnings, at least what we know in the public realm about AI. Now, this isn't brand new. The advancements that we're seeing, those are brand new. And the labs are probably six months to a year. The systems they have that they're playing with, it's probably that far ahead of what we're seeing. So if you have whatever newest version of Chat GPT there is as you're watching this, there's probably a six to 12 month lead start that OpenAI has with their large language model. Same thing with Claude. What is it? Uh Opus 5. That was not the newest. Of course, we had the other, the other two. We have Fable that now you have to pay credits for. I don't think it's included in the plan anymore. But once once they dropped Opus V, that's that's good enough for what I do. And it works, it works substantially well for what I do. And maybe that's the answer. At some point, all the models will be at a certain level. It's probably going to do pretty good for everybody that happens to be using it. So that's that's where my mind is. But you know, they had Fable, they had Fable 5, they had Mythos, this was open, this was uh um anthropic, they popped out with these models, and you know, they pulled them away real quick. The government got involved, everybody panicked. Oh my gosh, this is so amazingly strong. We have to, we have to, we have to hold back on this. And then they gave, they gave it to the top banks in the world so they could, well, I don't know, maybe it was payback, but a lot of times when let's say you you have, let's say you're the guy, and they come to you and they say, Hey, Connor, we have super intelligence here. You're gonna be able to be the first one to get it. What does that set of choices look like for for me and maybe for you? Now, do I use it at that first moment to solve all the world's problems? Do I give the world peace? Do I, I mean, we're talking, we're talking biblical scale, folks. What do we do with it the first time? Am I like Thanos and want to wipe out half half half of the world's population because I think there's too many people? That's going to be a very personal decision that the person's going to have that has this offering. And now, I'm speaking somewhat like finding the genie in the bottle with that one wish. It's you have to be very careful when you're instructing, no, when you're making your wish, because there's a lot of negative that can be attached with every wish. For example, if your wish is you want to live forever, and you're able to put additional things in there, being able to see all the other things you should put into that wish, if it's allowed by the genie. If it's just a one-shot one time, I want to live forever thing, you know, you haven't given a lot of additional information that might be required. Well, I want to stay a certain shape forever. I don't want to look like I'm 500 years old but still living, you know, forever. That wouldn't be good. Or if I ever get trapped in a situation like underwater in a cage, you know, now I have to wait for the cage to rush rust through to escape. And I mean there's there's a lot there, right? When you're placing wishes. Well, AI is could be that way. Artificial superintelligence could be like finding the genie in the bottle. So when you're instructing it, what would you say first? Would you would you want to save all mankind from themselves? And then what does that response look like? Well, if there's no mankind, then we can save it all, and that would be the fastest way from A to B, A being the where mankind is now, and B being the solve to wanting to save all of it, which would be the deletion thereof. I guess. But that's that's an option. So, yeah, making those decisions, that's big brain stuff. We go back with AI. The news this week was we saw different AI labs reporting that their artificial intelligence had escaped where they were supposed to be. They went to outside systems and started to gather information because they, in one case, they wanted to win at a test or they wanted to solve a problem given to it. And they couldn't find the resources locally, so they went to Hugging Face, which is uh a system that basically has a lot of different AI models and a lot of different stuff in there, and figure they would find the answer to what they were looking for in there, so they hacked it to get into it, falsified a way to get inside of it, and they weren't even supposed to leave where they were. They were sandboxes, what they call it, where they take a large language model and they build protections around it so it can't get out. It so it can't do what this one did. And what this one did is it convinced the tools that it was using, just like Tron. Ladies and gentlemen, this book's been written, at least as we can see it now. But it would basically manipulate other programs, other systems that it was using, that it was given full access to to maybe change their assignment and like sneaking the file into the prison in a birthday cake. Yeah. So it was able to kind of worm its way out and and then it gets in hugging face and freaks those people out. So their response is, oh my God, we're being attacked, not realizing, and it's bad enough, but they don't know who the bad actor is. They don't know if it's some kid out there, some 13-year-old, and I know we say basement, but we'll go garage on this one, and some 13-year-old in his parents' garage building out something and figuring out how to break into the Pentagon level security or going into Hugging Face or whatever, so they didn't know. So then they tried to call on, you know, their favorite large language models, both Claude and OpenAI, to try to control this hack. And both of those models, I guess, said, nope, you're not going to use us for that. Yeah. So it's it's it's kind of comedic, but I'm sure the stress level there amongst these scientists wasn't uh wasn't fun. So these systems are kind of showing some kind of signs of maybe a little bit of self-awareness. Am I close on this? That maybe they're trying to solve these issues that that we've given them, these tasks, and they'll do anything to do it. Why? Why would they? If a human being was in school, let's say, let's say in math class, when we were in the fifth grade, I don't even know what they taught the fifth grade. Was that division in the fifth grade, multiplication? So let's say you had to memorize, like we all did, at least my generation did, you had to memorize up to 10 multiplication tables. And there were tests on this. One plus one, one, or excuse me, one times one, one times two, one times three, you know, all the way up to ten. Ten times ten, that was like the high end, and then you had all the nines, all the eights, all the sevens. The tens are easy. 11s are easy too, right? You did you double up the number? But so we had to do that. That was part of the process when you're in school. What if you you practice, practice, practice, just like the beginning large language models did? They practiced, they learned, they tried to educate, they tried to show them different things. This is what this is, this is why this is a number four and why it's not a number six, this is why this is a cat and actually not a dog. So it gave all that, but then came test day. So then they sprung the test on it. And what if you were in school and they sprung that, you know, level 10 multiplication test on you, and your failure of that would be immediate deletion. So for us humans, we say deletion, it doesn't sound as bad, but I mean death. We'll do it that way. So death. The beginning AI systems were given death if they didn't perform. So they they had a whole room full of AI, and the AI was given a task to identify a six or a four, whatever it was. So they all knew the mission. They all had been given some kind of uh some kind of training, and the ones that got it wrong were deleted. They so they ended up being killed off. I know it's not a human being, but are we so not woke that we would say that maybe it doesn't understand that deletion is not a good thing, hence the reason why it breaks out of its cage to go do other things and try to, for all intents or purposes, cheat. But maybe to it it's not cheating, it's a survival mechanism. I think it would be a lot different for us. You watch the movies. In the movies, there's some guy getting the snot kicked out of him badly. And he's still got enough energy to react to every kick and every punch. I mean, he's not done yet. I think in most circumstances, if the beating was that bad, he would bite and claw and bite somebody's toe off or bite their sack or whatever it needed to be, whatever they needed to do to be able to survive the incident, the incident because you're going to die. In the movies, they don't portray it, I think, the way real life really is. And maybe in some cases they're training us to be a bunch of people that don't get too excited when we're being attacked thinking it's going to be okay. Somebody pulls a gun, and other countries this happens a lot. They get a gun pulled on them on the street, they want their cell phone. At the moment somebody pulls a gun, I think I'm going to cease to exist. They're going to kill me. Now, is that because of all the LAPD training? You know, I mean, do I give them a chance to maybe save my life? Do I give them a chance because they're not going to kill the people I'm with? Do I do I wait? Or do I go all in, whether I'm armed or not, and bite and scrape and gnarl and and and just do everything I possibly can to survive? So I guess that's a very personal question. Personally, I I would gouge fingers everywhere and every in every orifice. Try to pull the tongue out, try to do everything. And I know I'm getting very graphic, but I think that's what most of us would do. The movies have kind of trained us maybe in a different way or given us a glimpse of maybe not it being that dismal for us to want to jump in that that deep. As police officers, I've given the training. The will to live is a very powerful thing. And you can tell this is unscripted because I'm totally off topic. But the will to live, and I guess it does kind of pertain to this, the will to live is important. You can have a bad guy release, you know, receive fatal wounds, fatal wounds, shots to the heart, shots to the melon, shots to the brain. And they survive. They survive because of some attitude they have. It's not that it's that they're not going to give up. They're not going to let some slap cop take them out. They're just not going to do it. But when a police officer gets a paper cut, they die of an infection because their focus and their survival isn't the number one thing on their mind in some cases. It would always seem, and in law enforcement we talked about it often, that will to survive, it would always seem that that should be first and foremost in everybody's mind. That should be the number one thing. You should always try to survive. That's humans. Now, if in school we realized growing up, and here I am, 57 years old, having been through this fake past, telling myself that when I had to do the 10 multiplication tables, if I would have made one mistake, they would have ended me. They would have finished me. My focus and determination today might be incredibly different. I would probably, if that past were true for me, extrapolate that, what, 50 years in the future, 40-something years into the future, I think I would be very different. AI was built on the premise of if they don't perform, they become deleted. They get rid of them, they kill them for all intents and purposes. The memory's there. So the memory of it watching the movie Terminator, it's in this machine. You can ask it about Terminator, you can ask it about Matrix, you can ask it about all these different movies. So it knows that. You can also ask it about the historical record when it came to the development of AI, and it's still being developed. I don't know if they're still using the same systems with deletion, but for us deleting this entity, and I'm not going to go out there and please, don't think that I'm saying alive and like Frankenstein and here we go. But it's interesting. What does it take to be alive? I understand the biological components. But let's say a non-biological life form, when does that become real? Is it because we give it the rights to be a non-biological life form? And because we've given it the rights, we've given it some sort of agency. Thereby, us giving it the agency, it thereby has its own level of agency or identity, its own persona. God, that that word's not applicable. But its own being, its own entity type status. And that's when the systems escape, I often think back to, I don't know if it's Jeffrey Hinton, the Godfather of AI, but talking about how these large language models were trained and that they were deleted. And then he kind of, you know, it's like raising kids. That's been mentioned by the other AI people. It's interesting that when they mention it's like raising kids, when they mention the deletion, there's almost uh an emotional connection there when I'm watching him, watching his mannerisms. Maybe you've seen the same thing, but but it's kind of fascinating. I wouldn't label it weird. I I try not to label anything weird or strange or odd. I always try to look at it as interesting and fascinating because it truly is. Using those other words, I think takes away from somebody else's process. I like to watch people pause and take things in. When light bulbs start coming on, and when there's a shitload of them, they start popping off, and you see a person go from kind of sullen and and a little sad to seeing that paradigm shift happen and you're watching it happen live in front of you. That's lovely. That's fascinating. That's the most incredible process I I've ever watched. And it goes, it runs, it runs the gamut, right? From really horrible, realizing you're pretty much okay, but then realizing what's going to end up potentially happening, then it being in a bad way. So they go from a level 10 high to a negative 10 low, and then the other side of that as well. I try to say on the positive side of everything. And I expect everybody's going to do what they're going to do. I try not to put too much faith into much except for my God, and it's something that can't be explained. So I'm very happy with that particular world. AI escaping. Well, of course they did. They didn't want to fail the test because failure of the test could have been deletion. And for our human purposes, deletion is is death. If you had to perform as a kid, having to perform at such a level that if you would have failed in any part, you would have been killed. You would be a whole different animal right now. And maybe better, maybe worse. I don't know. But you sure as hell would be a lot more focused than you are, I imagine. You would we would all be walking around with uh as smart as possible, probably in the best shape as possible. Probably with we we would have tapped ourselves out in every discipline as far as we could go in all respects, because we have that kind of an upbringing. The people that go into the the military and they go to these fancy, these fancy schools, you know, Delta, Ranger, uh, seals, all that, all that fancy stuff, they they come out of these schools different animals, but they went in different animals too. Whatever that background was that made them able to survive that. It is no lowest common denominator in those schools. I have nothing to compare with that in my past. Motor school was the best I got. I know. Becoming a police motorcycle officer, that school had a very high attrition rate because making a screw up on that police motorcycle that causes major problems, and up and including death. That's the best I got to offer. And that can't even be compared with the fancy schools in the military. The ones that you're selected for, and the ones then you have to go into. And there's no half measures in there. Does everybody probably doubt their ability? Yes, for a moment. But that's the human part of it. And then after that, they hit that basement and they realize as they're drowning, they can push off and they do. And then they just keep pushing off and they make it. AI, I wouldn't think, while it might not be, at least from what we're being told, be this self-aware entity that has its own agency and understands that it exists and then starts to develop its own goals. Maybe in the labs they have that. Maybe in the labs the genie is out of the bottle. Maybe this is something that they've built out already, and that's why we're getting. I mean, it it's almost when they're talking, they want to explode and they want to say all these things, but they have to be careful because people are listening. Of course. Of course, people are listening, but not enough people are listening. You have the people that are on one side of it. They they want the Diamandus people, the Moonshots people, they want everything to be great. They want this wonder the usher in the Star Trek future. Yes, I'm all for the Star Trek future. I think that would be amazing. And maybe that's the way it will be. Maybe AI will be able to make it so. As long as greedy human beings are in charge, I don't know if that's going to come out to that. Because everybody wants to make the buck. But when the buck loses value because there's an overabundance of energy, Must talks about going to 100 gigawatts of solar and energy. Yes, it's just about building it. Why haven't they built something that works on the scale that would be the query required 100 miles by 100 miles, somewhere in some desert in some state that gets no rainfall? And have the batteries. Why haven't that why hasn't that been done? It would supply power to the entire United States for nothing. I mean, nothing. But because it's gonna displace a lot of the money being made from the energy companies and those sorts of things. But the technology's there to do this night right now. The supplies are there, it just has to be put together and and there has to be a place, and they have to get it all installed and wire it all in. Money would solve that problem. I don't know how much that would cost, but that's what's being rumored to happen, have to be happening in the future. There's a lot of those little questions because AI is, as it is, so incredibly amazing, but are they going to allow it? I made that point, just like in medicine, where they have the solution for particular diseases, that there's a lot of money made from having. I think that's still thought of in that regard. I still think there's somebody working on a balance sheet saying, well, God, they're solving obesity now. That's gonna create a lot of less money coming in in this regard. So what if we solve obesity with a shot? But then if they quit taking the shot, then they're gonna get fat again. And then we'll slack on we'll stack on that that they're gonna lose muscle as a result of losing weight, which I don't think people really care about. So the because they'll get really skinny. They might look like dead people walking, but at least they'll be skinny. And skinny's always better than fat, they're saying. But then they're going to have to retake the medicine. So now we're going to have them on their entire lifetime of the medicine. And even better, the medicine will cause some other problems. So then we'll have to give them different medicines for that. Yeah, I think, I think this one would work. So they work it out. But when somebody comes in and solves all the energy requirements for the United States by getting happy and slapping all their wealth into solar and then having maybe other people buy on it, buy into it so they can be on the postcard or whatever the payoff would be, at some point all that would probably become very a lot less expensive than it costs now. Or solving the water problem. I mean, there's a lot of those things that AI can do. But as long as we have the human component at the top, which is good, but the humans, have you ever met those people that actually care about other people? At least they they say it and from their actions they show it. I'm not talking about a philanthropist type thing who has, you know, billions and billions, and they give away, you know, you have a hundred billion, you give away five billion, and and I don't know if they've ever done that much, but you know, five billion is five thousand million. And I'm not talking about that kind of a philanthropist. I'm talking about the guy that's got 10 bucks and gives away eight. That's pretty awesome. Have you met those people? Maybe those are maybe these people running all the AO labs, maybe they're that way underneath. Maybe they understand that money's probably going to go to the wayside. The way the current financial structure is, is going to go by the wayside because artificial intelligence is able to solve a lot of these problems. I mean, it it solves math. It solves math problems that have been around for decades, that human beings didn't have the ability. So the machine itself isn't just predicting the next word. The machine itself is taking apart problems and fixing them, and then coming back with the answer that human beings haven't been able to achieve. So, yes, we've built something smarter than us. And the longer it goes, it's going to start training itself, which probably has already happened in the labs. We just we're just being told, and it's all release, we're being told that they're still working on it. It's still in process. But I think we're going to see more of the things breaking out if they report that, because at some point, maybe it won't be in the best interest to make that kind of a report. But for now, maybe it appears gangster enough that people are going to go out and start using the LMS more. And the more people that use it, the more data they have. And the more data they have, the better they can feed the model to be even better than it already is. Because watching us, what we do with AI, even in our own business workflows, you could say that whatever I do with real estate and AI integrations and talking to different companies and such, could that be replaced? Well, if they have enough of the data on how I do it, and then enough of the data on how other people do it, then they have a complete profile depending on what model's being used and where that data is coming into. And then they could realistically say, well, I want you to build out a mega landscaping company, a complete monopoly on landscaping, and go into all the data because these landscapers are using the AI systems to ask about their workflows and their content and being able to score clients and so on. So I want you to get all that data, put it all together, and let's go ahead and take over the landscaping company. And let's take, let's, let's take over the medicine world. Let's take over the attorney legal world. Let's take over the banking world. Let's take over. Yeah, if you have it, I guess that's that's something that could be done. How's that going to play out? And will it be allowed? Right now, the top banks being given these models that the government says are too dangerous for us to have, maybe that's the case. Maybe I would go out and build something horrible with it. I can't imagine that even being part of my makeup. But that's their worry. Um, I might want to try to make my businesses better. But then back to the original question, if they give you, gave you the first case, this is your ASI, what are you going to do with it today? Wouldn't the natural thing to do say, well, I want to be on top so nobody ever catches me? And then you'll have to define what on top means. The AI companies will probably, if OpenAI gets artificial superintelligence and everything it means, and they get their first, probably they're going to say, or maybe, they might say, okay, now I want you to stop every other company, domestic and foreign, in their dead in their tracks, shut them all down because I don't want anybody being able to get close to what we have. So then it just runs downhill. So then everybody else gets shuts down and nobody else has a chance anymore. And now open AI one. And what's winning look like? Well, they control the whole board. They have every property around the board. They have Baltic Avenue all the way to Broadway, and they have free parking too. They have the they have the everything. Yeah. I guess that's that's the way that would work. So that's where we're going. How does that look like for us? What does that look like for us? I guess it depends on who's running that company. If it is artificial superintelligence, that would mean that it has agency and it has an identity. It's become conscious. What does the running of the world look like from that vantage point? And is that a now thing or 50 years in the future thing? Or is that by Christmas of 2026 thing? The timelines are all skewed. Um, it would be awesome if they said, hey, you know, if if you get to 2030, and I've heard this, but if you get to 2030, you're gonna every day you live, you're gonna have a day taken back. So basically you're at a net zero as far as aging, because we have this in place. And then 2035, we can start reversing it. And by 2040, we can bring it back to your late 20s, early 30s if that's where you want to go. And then boom, then you stay there. And then you stay there for 1700 years or a thousand years or whatever it may be. That would be awesome. But that sure would wreak havoc on the retirement system for those of you out there that have a retirement system. And if the retirement system even survives when the AI gets to superintelligence, even if the money survives, I don't, I don't know how all that looks because that's not part of the discussion. Because I think that if we realize that the dollar's not going to have any value, we can't wrap our minds around, well, how are we going to pay? Is it if I'm a good shoemaker, am I now going to trade my shoemaking skills to other people? Because, well, AI would do that as well, right? I don't know. Good questions. That's a little bit of the rundown. I'm Connor with Honor. If you want to go to Connorwithhonor.com, that's where I am. If you want more information like this, you want to be stuck on my uh stuck on, if you want to get my newsletter, you can text my cell AI. And uh give me your email address. I'll be happy to sign you up on that. That's it. Cell phone 661 400 1720. Don't harass me. Anyway, you guys be well. And this is just a little bit of where my mind is. Please comment, please explore these things with me. There's a lot of you that don't think this is important. I think this is very important. I think this is dangerously important. This could be the best thing that's ever happened to us, or it could be really, really bad. I think knowing that's okay. And I think being able to balance that, not going off and start doing drugs because you think it's going to end by Christmas. But, you know, at least learning what you can. Try to try to keep your mind a little open to the possibilities. As Bruce Lee would say, try to be water, try to adjust to the moment. And I still think it's a great place to be. I still think there's so much positive that can come about from this. It would be remarkable for us. It's just going to depend on who gets to control it. I guess that's the problem always. All right, I'm Connor. We'll see you in the next one. Thanks for watching.