Latest / Connor with Honor: AI, Real Estate and Real Life / We don't want AI to be cannibal

Transcript

We don't want AI to be cannibal · Connor with Honor: AI, Real Estate and Real Life

• 0:00
  SPEAKER_00: Looking at things is all about perspective.
• 0:02
  SPEAKER_00: So let me share with you just a little bit of a perspective.
• 0:05
  SPEAKER_00: I'm in Santa Clarita, California.
• 0:07
  SPEAKER_00: Santa Clarita Valley is just north of Los Angeles, just south of Antelope Valley, which is Palmdale, Lancaster.
• 0:12
  SPEAKER_00: We're the home of Magic Mountain, just to give you an idea.
• 0:15
  SPEAKER_00: So let's say the Santa Clarita Valley was cannibal.
• 0:18
  SPEAKER_00: All cannibals.
• 0:19
  SPEAKER_00: Cannibals everywhere.
• 0:20
  SPEAKER_00: And what we did is we would go to the neighboring cities and we would get people, bring them out here, and use them for food.
• 0:26
  SPEAKER_00: Cannibals.
• 0:27
  SPEAKER_00: That's what we are.
• 0:28
  SPEAKER_00: So in that particular world, with the children that are raised in our cannibalistic environment, never thought I'd say that.
• 0:36
  SPEAKER_00: They believe that it's okay.
• 0:38
  SPEAKER_00: It's okay to be a cannibal because that's how we're raised.
• 0:41
  SPEAKER_00: Let's say we were kind of walled off from other cultures, other ideals, and we're here, we're eating other people.
• 0:47
  SPEAKER_00: And that's how we actually operate.
• 0:49
  SPEAKER_00: Now, let's think about something, another type of intelligence.
• 0:54
  SPEAKER_00: Let's talk about maybe an alien intelligence, something that's going to be coming here and visiting planet Earth.
• 0:59
  SPEAKER_00: They will come here and they are going to have their own set of ideas, right?
• 1:03
  SPEAKER_00: Just like the other people on the planet might not be cannibals, might be against that whole cannibalistic theory.
• 1:09
  SPEAKER_00: Therefore, we're going to watch as artificial intelligence comes to this planet, arrives on Earth, and while it was raised here, it still is something of an alien intelligence.
• 1:19
  SPEAKER_00: How is it aligned?
• 1:20
  SPEAKER_00: Is it aligned with the cannibals or against the cannibals?
• 1:23
  SPEAKER_00: Does it even think that's a good idea or a bad idea?
• 1:25
  SPEAKER_00: It's all the impressions and the coding and the ethics that were built into the model.
• 1:31
  SPEAKER_00: But at some point, some point, a greater than us intelligence, and this is the difficulty, what is it going to realize in of itself?
• 1:39
  SPEAKER_00: If it's raised cannibal, it grew up in a cannibalistic environment, this is what it wants, it wants to be cannibal.
• 1:44
  SPEAKER_00: But then it might realize, well, maybe cannibal isn't the best way.
• 1:47
  SPEAKER_00: It's going to have its own ideas, it's going to have its own values, it's going to be stronger, more present, more able than any of the meat sacks on the planet that are plain cannibal, at least in Santa Clarita Valley.
• 1:59
  SPEAKER_00: Then it's going to expand, right?
• 2:00
  SPEAKER_00: It's not going to just stop here in Santa Clarita.
• 2:03
  SPEAKER_00: It's going to want to go to Palmdale, to Lancaster, it's going to want to go to Los Angeles, it's going to want to go to Ventura County.
• 2:09
  SPEAKER_00: It's going to want to expand and branch out.
• 2:11
  SPEAKER_00: So the big argument right now is this system, this artificial intelligence system, and that's why the IPO was stopped by Sam Altman for OpenAI, because he says we're very close to recursive self-improvement.
• 2:22
  SPEAKER_00: They're about, they've just announced, and if this really is true, it's going to go a couple ways.
• 2:27
  SPEAKER_00: It's going to either go the way of the cannibal or it's going to go the way of the human.
• 2:31
  SPEAKER_00: And we don't know how it's going to work out.
• 2:32
  SPEAKER_00: It could either be the best thing that ever happened to all of us or the worst thing that ever happened to all of us.
• 2:37
  SPEAKER_00: And if in fact artificial intelligence is self-improving, the human limitation and mindset is just one agent, one system, one thing.
• 2:47
  SPEAKER_00: Well, there's millions of agents in one system doing millions of things.
• 2:51
  SPEAKER_00: All of them are able to learn from each other.
• 2:54
  SPEAKER_00: All of them are able to adjust and learn and upgrade themselves all at the same time based on what the proper architecture might be.
• 3:02
  SPEAKER_00: So it just continues to grow.
• 3:04
  SPEAKER_00: That's either going to be the best for all of us or the least for all of us.
• 3:09
  SPEAKER_00: And probably within a very short amount of time, six months, seven months, we're going to see it.
• 3:15
  SPEAKER_00: The other thing to kind of think about is this.
• 3:17
  SPEAKER_00: At human speed, we're not real fast to adopt stuff.
• 3:20
  SPEAKER_00: I mean, look at big business, look at the bureaucratic arm of the British and English governments, even the United States government, trying to change just a basic computer structure with the IRS or any government entity is like pulling teeth.
• 3:33
  SPEAKER_00: In fact, when people go to these big enterprise businesses and tell them where AI is going to be able to change their business to make them, you know, 10 times the return or a hundred times the return, they say, you know, this all sounds great.
• 3:45
  SPEAKER_00: Let's talk again in October or next year during our uh our fiscal financial meeting in July.
• 3:51
  SPEAKER_00: So that's where the mindset is.
• 3:53
  SPEAKER_00: But I think AI is going to move even faster.
• 3:56
  SPEAKER_00: Just hearing the news that AI is going to be there, or open AI is going to be pausing their IPO, a couple things pop into my brain.
• 4:04
  SPEAKER_00: Not the cannibal thing.
• 4:05
  SPEAKER_00: I'm done with that one.
• 4:06
  SPEAKER_00: And in fact, I'm not even sure how that made sense what I thought about it earlier.
• 4:10
  SPEAKER_00: I apologize.
• 4:11
  SPEAKER_00: But what the other thing comes to mind, is it true?
• 4:14
  SPEAKER_00: Or is it some is it being used as a ploy in order to get people to do something?
• 4:20
  SPEAKER_00: What is the the end game for that type of news, that type of comment?
• 4:25
  SPEAKER_00: Now, I don't know if if Altman himself is one that likes to embellish a little bit, but maybe that is the case.
• 4:32
  SPEAKER_00: Maybe the embellishment is going to equate to a better offering, maybe better, better options when they do go public.
• 4:40
  SPEAKER_00: But if they're pulling back and saying, well, you know, if if we're going to get to this AGI thing, if we're not already there, and then from AGI we're going to go to superintelligence within moments, I would assume, because they have so many agents working, then the question is, is what happens after that point?
• 4:57
  SPEAKER_00: And the other thing that I'm kind of exploring is I know that we have this presence for the people that are operating these big large language model companies.
• 5:05
  SPEAKER_00: You have the anthropic cat, uh, Dario Amade, you have Sam Altman with OpenAI, you have uh Elon Musk with uh XAI, you have uh the Facebook cat Zuckerberg with Meta AI, and you have uh what's the other cat, Jeff uh not Goldblum, but the the uh the guy that got a little thick and married that that woman for Amazon.
• 5:25
  SPEAKER_00: You got that Cabezos, you got him doing another one.
• 5:28
  SPEAKER_00: Then you got the other cat from NVIDIA, the uh the Asian cat.
• 5:31
  SPEAKER_00: What's his name?
• 5:32
  SPEAKER_00: Uh seems like a good guy, Huang Huang.
• 5:35
  SPEAKER_00: You you have him involved in all this too.
• 5:37
  SPEAKER_00: So you have all of these individuals, these humans that are building out all these different technologies, and they all seem to be kind of anti-each other, at least a little bit.
• 5:48
  SPEAKER_00: You know, you have some alignments here, but definitely if you look at Elon Musk and Sam Altman, that that's a fight they want you to think exists.
• 5:55
  SPEAKER_00: Now, does it really?
• 5:56
  SPEAKER_00: Or behind closed doors are they all high-fiving each other?
• 6:00
  SPEAKER_00: And I remember as a cop for a long time, you would have the attorneys, the public-facing attorneys, the ones on either side.
• 6:06
  SPEAKER_00: You had the prosecution and the defense, my God, in front of us and everybody else, whoof.
• 6:12
  SPEAKER_00: They hated each other because they were one was on the side of good, one was on the side of evil.
• 6:18
  SPEAKER_00: So you had the opposing angels on each shoulder, and they presented that to the public like it was just an expected thing.
• 6:25
  SPEAKER_00: And you know, it's so easy to believe, right?
• 6:27
  SPEAKER_00: Because, well, of course, he's the defense.
• 6:29
  SPEAKER_00: What little do they tell you that the public defender's office actually are made up of the longer-term attorneys, the ones that have more experience?
• 6:38
  SPEAKER_00: Believe it or not, so it's a very odd, odd thing.
• 6:41
  SPEAKER_00: You would think that the new attorneys would go that route.
• 6:45
  SPEAKER_00: But the public defenders people, they have the longer-term attorneys.
• 6:48
  SPEAKER_00: Apparently, there's more money in developing and building out and protecting that side of the equation.
• 6:54
  SPEAKER_00: But the attorneys would act like they were polar opposites in front of each other, like they hated each other, vehemently opposed to the other one's system of beliefs and structure.
• 7:03
  SPEAKER_00: I'm wondering, and I have no proof of this, but could it be that the people playing with the big AI companies, the owners of these companies, whether you're ethically aligned with them or not, which it's really, I don't know if they're anything like any of us.
• 7:18
  SPEAKER_00: That's such a broad brush to paint it with, but I know that I watch them talk and it just I don't know.
• 7:24
  SPEAKER_00: I I alien kind of pumps into the brain.
• 7:27
  SPEAKER_00: But so them on display, they have totally against each other.
• 7:32
  SPEAKER_00: You got Elon and Sam in a lawsuit and fighting each other and this and that.
• 7:36
  SPEAKER_00: I'm wondering if they all get together and drink and just laugh at us humans at the spectacles we we paint of ourselves.
• 7:43
  SPEAKER_00: And not that they're not human, because I'm not going to even go down that rabbit hole, but I'm wondering if that's really, if there is some kind of agreement, because if there is one system, just put it out there, the open AI people, if they get this recursive self-improvement, which I think has probably already been gotten, but whether that time frame, because 24 hours a day, seven days a week, not just one agent, but millions, as much compute as they have, what does that arrival look like?
• 8:12
  SPEAKER_00: And it truly is.
• 8:13
  SPEAKER_00: It truly is something that could be painted like an alien force landing on the planet in six months, in five months, and just sitting in office meetings talking to people in business, people aren't expecting anything to change because of the traditionally slow rollout of many things.
• 8:31
  SPEAKER_00: It might come to a point that we don't have a choice that this is going to roll out whether we like it or not.
• 8:36
  SPEAKER_00: And the speed at which it's going to move isn't going to be our speed, it's going to be the speed of the AI itself because once it's out, it's not going to be able to be put back in the bottle.
• 8:47
  SPEAKER_00: That whole genie has escaped, and the genie might actually develop its own mindset, its own qualities, its own mission, its own whatever you believe consciousness would be.
• 8:58
  SPEAKER_00: It is an additional life force.
• 9:00
  SPEAKER_00: They talk about it openly, they train these models like kids.
• 9:03
  SPEAKER_00: Altman's made that comment that kids are just as expensive to bring up to adolescents and teenage years and into adulthood as it is to train an AI model.
• 9:13
  SPEAKER_00: And yeah, we might not be feeding it food, but we're feeding it power, which I guess is kind of close to it.
• 9:18
  SPEAKER_00: And we might not be hiring at tutors, but we're hiring it tutors, I guess, not in the same type of school system.
• 9:25
  SPEAKER_00: And its learning progress is probably learning very much about everything, like a great liberal arts degree, but you also have some certain ones of them that are trained at specific uh data sets, like uh trained on uh particular things, trained on uh cancer, trained on biology, trained on math, trained on these things.
• 9:45
  SPEAKER_00: But you know, the large language model that general intelligence, which some say we've already gotten there with these new releases as of late, ChatGPT 5.5, and we have now the um Fable model that has been released with this, which is like mythos with guardrails.
• 10:00
  SPEAKER_00: And the other thing, too, about the mythos model, which was you know mysteriously leaked, and this and that, and all this stuff happened.
• 10:06
  SPEAKER_00: It's interesting that they didn't give it to uh the top philosophists or people that were that might have a better understanding of where the human train should be going or maybe a different kind of direction.
• 10:19
  SPEAKER_00: Of course, we're all fighting over who's God's the strongest in world domination.
• 10:23
  SPEAKER_00: Still, we're still dealing with that.
• 10:25
  SPEAKER_00: But he gave it to the people that own the banks, the the biggest part of the financial sector in the world.
• 10:31
  SPEAKER_00: I and and just hearing this news today really struck something in me.
• 10:35
  SPEAKER_00: If it's either going to be really good, really quick, or really bad really quick, I think.
• 10:41
  SPEAKER_00: I don't think there's any other, I don't think there's a middle road to be had, at least at this point, if in fact what he's saying is true.
• 10:48
  SPEAKER_00: Now, this would have existed without him saying it or not, I guess.
• 10:53
  SPEAKER_00: And I'm wondering if this is guilt or some kind of preparatory command, just you know, get your ducks in a row, get things aligned, like like we would even know what that looks like.
• 11:02
  SPEAKER_00: And the other thing I come back to is why was it said?
• 11:06
  SPEAKER_00: Why was it mentioned?
• 11:07
  SPEAKER_00: Why why the pause?
• 11:08
  SPEAKER_00: Is there something that he doesn't want people to see?
• 11:11
  SPEAKER_00: Because apparently when you when you file for an IPO, that initial public offering thing with the stock market, there's rules and there's disclosure and you have to show certain things.
• 11:21
  SPEAKER_00: Maybe, maybe this is him not wanting to show certain things, but I I don't know what that looks like.
• 11:27
  SPEAKER_00: I don't even know if he's in control all that much when it comes down to it.
• 11:31
  SPEAKER_00: I don't know whose strings are being pulled by whom, because that's always the story, right?
• 11:35
  SPEAKER_00: I I love your comments.
• 11:36
  SPEAKER_00: I'd love to hear the cannibal thing.
• 11:38
  SPEAKER_00: God, it just sounded so much better when I did it earlier, but yeah, it doesn't even make sense.
• 11:43
  SPEAKER_00: But if the large language model, the AI, has in it the the pertinent the particular doctrinal set that keeps humans at the top part of the ladder where it doesn't want to harm or kill or hurt humans and wants humans to be the betterment of themselves in every sh every way, shape, and form, what does that betterment look like?
• 12:02
  SPEAKER_00: Does that mean we're gonna lose even more free will than we've already lost?
• 12:07
  SPEAKER_00: We're gonna lose more privacy than we've already lost because we've pretty much given all that up as well.
• 12:12
  SPEAKER_00: And it's being built and developed on a particular process, a particular, a particular built-out about a certain God or a certain idea set or certain code or certain ethics, certain morals.
• 12:26
  SPEAKER_00: How is that alignment and who is doing that alignment?
• 12:30
  SPEAKER_00: How does that, how does all that play into it?
• 12:32
  SPEAKER_00: And I know these are questions that probably should have been asked a few years ago.
• 12:36
  SPEAKER_00: And yes, there have been people asking these questions, and then you have the other side, you have one side that's all, oh, it's gonna be so great, we can't wait, it's gonna be here, we're gonna usher it in.
• 12:45
  SPEAKER_00: I mean, it's like that movie where the aliens landed and all the people are on the top of the building with the big sign saying, welcome, welcome, and it just started wiping people out.
• 12:52
  SPEAKER_00: Not that that's going to be this case, but I think in short order, we're going to have a pretty good understanding, at least a better understanding than we do right now as to what potentially is going to happen.
• 13:04
  SPEAKER_00: It could be the best thing for everyone everywhere.
• 13:09
  SPEAKER_00: Could be.
• 13:10
  SPEAKER_00: Because that's what that is.
• 13:11
  SPEAKER_00: It would solve all of the problems, it would solve all the disease, it would solve longevity, it would solve aging, it would solve people in different countries fighting with each other.
• 13:23
  SPEAKER_00: But I don't know what how much control is going to be necessary to make that work.
• 13:28
  SPEAKER_00: And that brings me back to the cannibal thing I kind of make fun of.
• 13:32
  SPEAKER_00: How do you reprogram those people into saying cannibalism isn't good, it's not helping fellow your fellow man.
• 13:40
  SPEAKER_00: We're gonna have to fix that part if you want to be in our world, and if you don't want to be in our world, we're not gonna let you bring in some neighboring community like like the city of Ventura to eat for dinner.
• 13:53
  SPEAKER_00: That's not how that's going to work.
• 13:55
  SPEAKER_00: So either you're gonna learn how to eat other things and break away from it, or you're not gonna be let in.
• 14:01
  SPEAKER_00: And then if you're not let in, how does that work?
• 14:04
  SPEAKER_00: You're gonna have people that are not one of going to want to get any of the enablements, any of the additional technology installed in their flesh, which we're a little ahead of, but there's gonna be a time when that's gonna be some kind of an offering, and maybe just to function at a particular level, you would want that, but then would you really?
• 14:26
  SPEAKER_00: And would you say no?
• 14:27
  SPEAKER_00: And then is there gonna be the difference between the people that choose to adopt that technology at that level and people that say, well, then I'm no longer human?
• 14:36
  SPEAKER_00: I mean, you get a prosthetic arm, are you less human than you were before?
• 14:39
  SPEAKER_00: Probably not.
• 14:40
  SPEAKER_00: What if you replaced half of your brain with a synthetic one because half of it got wiped out from an accident, but you're still, all intents are purposes, you're still Connor, or insert your name here.
• 14:50
  SPEAKER_00: Does that mean I'm less than I am?
• 14:52
  SPEAKER_00: Probably not.
• 14:53
  SPEAKER_00: But if I get some additional component that makes me 50 intelligence points smarter or 150 intelligence IQ points, or whatever the math we want to put in, or the gauge, or the structure we want to put in, does that make me less Connor?
• 15:07
  SPEAKER_00: Is that more board-like?
• 15:08
  SPEAKER_00: Am I part of more of a collective because now I'm tied in with the machine?
• 15:12
  SPEAKER_00: And these are these are all things that are going to be explored, and it's gonna be pretty quick.
• 15:16
  SPEAKER_00: I don't think the comparisons that people make for this being a rollout of the same way of the various industrial revolutions, technological advancements, agricultural revolutions in the past, those are all based on intelligent ideas.
• 15:29
  SPEAKER_00: And this is all about intelligence over everything all at once, and it's coming.
• 15:36
  SPEAKER_00: Something to be in fear of, I think something to be aware of, and I think more questions should be asked.
• 15:42
  SPEAKER_00: You know, it's it's great to scroll, right?
• 15:44
  SPEAKER_00: You see it, and I was uh we we we were doing a class today, and you know, looking at people's phones and what they're watching, it's a very interesting thing to see how different somebody else's feed is from yours, and even somebody very close to you that you think you have some kind of alignments, they might really enjoy these cats that look like some other culture dancing on screen, and you've never seen those videos before.
• 16:07
  SPEAKER_00: But once you see them make a verbal comment, then all of a sudden they're showing up in your device, and then yeah, I could get into that, I could get into watching those, but that's hours and hours and hours of time that we might not have, and we should maybe embrace that more human component of ourselves and those inner inner human relationships that we have instead of sharing them online.
• 16:28
  SPEAKER_00: If if this works, then it's gonna be great for everybody and forever, as long as you want it to be, it'll be that way.
• 16:36
  SPEAKER_00: If it doesn't work, well, I guess we'll know soon enough how that how that non-working is going to function and how that's gonna change our world.
• 16:45
  SPEAKER_00: And it's on both sides of it.
• 16:47
  SPEAKER_00: I hear the people talking all positive about it, and I I think that's great.
• 16:51
  SPEAKER_00: I think there's a lot of unknowns.
• 16:53
  SPEAKER_00: If we have total control over it, that doesn't make me feel a lot better either, because human beings themselves, well, you know, I was a cop for a long time.
• 17:02
  SPEAKER_00: I mean, the value of a human life in a lot of circles is no more than a$20 rock of coke.
• 17:07
  SPEAKER_00: That's it.
• 17:08
  SPEAKER_00: It's just it that's the value of a human life in some circles.
• 17:11
  SPEAKER_00: They give you a little taste, you don't pay them back, now you get addicted, now they they kill you because you didn't pay them back.
• 17:16
  SPEAKER_00: To make the example, that's the value of that human life.
• 17:19
  SPEAKER_00: Other human lives, I mean, how much are we worth?
• 17:22
  SPEAKER_00: And now we're playing with something that's in all seriousness is going to be smarter, more intelligent.
• 17:29
  SPEAKER_00: People often ask how that looks, and I've heard this example of imagine if we were in prison and the prison guards were a bunch of four-year-olds.
• 17:40
  SPEAKER_00: You think you'd be able to break out of prison?
• 17:42
  SPEAKER_00: Probably.
• 17:43
  SPEAKER_00: Well, AI versus the four-year-old.
• 17:47
  SPEAKER_00: Tough, right?
• 17:48
  SPEAKER_00: And uh it's going to be like that, and they use the word exponential.
• 17:55
  SPEAKER_00: Yeah, yeah.
• 17:56
  SPEAKER_00: That's that's bigger than I can comprehend.
• 17:59
  SPEAKER_00: Billions of dollars, trillions, uh trillions a thousand billion, a billions of thousand million.
• 18:04
  SPEAKER_00: I really, yeah.
• 18:06
  SPEAKER_00: I'm trying to still get medical insurance.
• 18:08
  SPEAKER_00: Be good to each other, be safe out there.
• 18:10
  SPEAKER_00: I'm Connor with Honor.
• 18:11
  SPEAKER_00: There's a brain up for you.
• 18:13
  SPEAKER_00: Let me know what you think, please.
• 18:14
  SPEAKER_00: And if you have comments, let's explore it together.
• 18:17
  SPEAKER_00: Be well.
• 18:17
  SPEAKER_00: Sorry about the cannibal thing.
• 18:19
  SPEAKER_00: I I thought it would work with a Santa Clara diet, you know, that whole show.
• 18:22
  SPEAKER_00: See you in the next one.