Loading summary
Transcript21 lines
- [00:00]
A
Hello, everyone, this is Gabriel Custodie. I wanted to do a check in here and inform you what's going on. So the podcast is still indefinitely suspended for financial reasons that I discussed in the previous episode. That has not changed. Our efforts now are directed 100% at the Watchman's Torch newsletter, where we are releasing great stuff every couple weeks and to our products on escapethetechnocracy. Com. As I suspended the podcast, I did have a few more interviews that were scheduled. I didn't cancel all of them, so I did complete them, and those have been released exclusively to Watchman's Torch Newsletter subscribers. So if you purchase Watchman's Torch, log into the back end, log into your user account, and you will have access to these. I'm sure I'll have some more interviews that crop up, in which case they will be accessible to the Watchman's Torch Newsletter subscribers. So this is just one more reason why you want to be subscribed to Watchman's Torch Newsletter. What are these episodes? Well, I want to give you a taste in this particular episode episodes. So I just have a few clips here from all of those episodes that you can now access exclusively@watchmanstorch.com and through our newsletter. The first clip comes from my discussion with the creator of nano GPT.com I had previously talked to him. And in this one, we went into all the new updates, all the latest things that you can do with the AI tool of Nano GPT. What is the most absurdly expensive prompt that you've ever seen?
- [01:30]
B
We've had people who spend like $20 on a single prompt. Because some of these models, for example, Claude Opus, that's like the expensive version of the cloud models. It can be incredibly expensive. Same with GPT5 Pro. So we sometimes saw people who would do like. Like, they would message us like, this is getting really. Gotten really expensive. I'm like, yeah, you threw in, like, essentially the entire Lord of the Rings into one of our most expensive models and ask it to, like, write an additional chapter. So it's like a huge input and a huge output. So, yeah, that's people that have sometimes spent like 20, $30 on a single prompt. But, yeah, funnily enough, those people, like, they. It's kind of what they wanted. Like, they were happy with what they got out and what they paid in most cases. Not always. Sometimes people are like, oh, I paid so much and came out with this crappy answer, like, yeah, sorry, man. Like, we don't control what it puts out. But yeah, so we've had people who did like, I think the highest must have been like 30 something dollars or maybe close to 40. I do have to say that's not at all the average. Like the average prompt is probably, I think $0.01 or maybe even sub $0.01. But yeah, it does sometimes, sometimes people just really want a certain answer or are really looking for the very best answer for a very specific thing. And then, yeah, some people are just willing to spend for that.
- [03:01]
A
Now AI is one of these garbage in, garbage out, sometimes sort of thing. So the better that you can control the tool, the more creative, clever thing you're going to get produced. What are some of the more creative uses of AI or creative or unusual or just beyond the typical stuff that you see people using AI for? That might pique some people's interest to go a little bit beyond the typical stuff.
- [03:27]
B
Yeah, so I would say this long stories model that we have, I actually love what the guy is doing because what he essentially does is people can enter a very simple prompt of just a story that they're interested in, but then he's written a sort of backend in such a way that it uses different models to create this very creative story. And then the models also create like image prompts. So they create images, they turn them into video, and then they stitch it all together. So it's like an entire workflow where you can just put in a simple prompt and you get essentially like a mini Pixar movie back. Which is honestly just fascinating to me just because of how well it already works. And like kids love it because they, they enter this simple prompt of like Spider man fighting, I don't know, Spider man fighting Superman. And they get this entire video and they're like, I created this. That's been, been really fun to see.
- [04:23]
A
Yeah.
- [04:24]
B
What else do people do with it? I think what excites me most is people who build like actually useful workflows. So instead of just, you know, talking to the AI in a regular way or using it for programming or role playing or whatever, they build an entire workflow in the backend where they send in a prompt and then it will analyze what's happening, it will feed back to the user. Like, you might want to look at this. We actually do this ourselves in a way which I'm pretty happy with. We have like a bot that's scanning Reddit comments and like summarizing them, and then a different bot is checking. Like, might this be relevant to Nano GPT? What's the sentiment here? Like, is it Is it positive, is it negative? Especially when people are talking about this, about us. That can be useful. So yeah, I don't know if that's very creative. I feel like the long story is the video model is far more creative. But yeah, just, just cases where it's, it's actually useful to people and saving you time.
- [05:22]
A
The next clip comes from a very fascinating guest who I had an excellent discussion with. This is Colonel Wyatt and he has an amazing YouTube ch where he discusses Africa and what happens in Africa and he's one of the authorities on the topic. So here's a clip from there.
- [05:37]
C
I would say I've probably been, I think in 38 to 42 of. Of African countries, including North Africa. I've lived and been stationed in Tunisia, Liberia, Botswana, Malawi, Niger, Mauritania, Uganda and Ethiopia and traveled for the rest. For business, for work. Yeah, so much of Africa and that.
- [05:58]
A
Is certainly unusual for most humans in the world. Even if somebody is going to Africa, which most people will not, it's probably only one country, maybe like to Cape Town for a vacation. What would you say is having been around Africa, what are some of the things that you take away that you really would not grasp if you were just experiencing Africa second or third hand?
- [06:20]
C
Well, that's an interesting question, one that I don't think anyone's ever asked me. So I take that as a challenge. Gosh, if I step back for a second and look at this, I would say the thing about Africa that people don't realize without having experienced it firsthand is that it truly is an amazing place. I mean the variety of the landscape, the peoples, the languages, the cultures, the cuisine, you know, the crime. There's also bad things. But the variety of things that you find Africa is just absolutely astounding and amazing. It's truly from. From Morocco in the north to the Cape of Good Hope in the south, from the Horn of Africa in the east to Senegal in the west. It's such a varied continent with tropical jungles, high desert, chaparral, Mediterranean climates. Not a lot of that, but some of it coastal things. Black Bantu speaking peoples, Arabs, Berbers, Asians, Indians, those of European descent, mixed race people. It's a very fascinating continent. I think most people don't realize that, you know, it's Europeans and Americans oftentimes look at Africa and they just think of, you know, it's just a place, but it truly is a fascinating place to go and very different from one place to another. This is I the only place I've never really lived Arguably is in Central Africa. I've lived in North Africa, West Africa, Southern Africa, East Africa, but I've never lived in Central Africa. I've been there, but I've never lived there. And frankly it's tropical jungles, so I'm not really that interested. I don't want malaria, but I would say that's one thing. Another thing about Africa is that relationships are different, generally speaking in Africa than they are in many other places. And that's a broad based statement. So people just can't. Don't attack me for making that statement. What I'm saying is that generally extended families are much more important than they are in the modern world, in the developed world today. I mean, if I go back two generations ago with my mother being one of 13 children, extended families, you know, beyond that, cousins and so on were a big deal. And you see immigrants coming to America, Italians and Jews and, and people from Slavic Europe, you know, they have extended families that they reach and rely on. That's still a common aspect of life in much of Africa. I mean, I could point to places in Africa where one person works and supports as many as 30 people because that's extended family who can't find work or employment of some sort. And there are expectations on those people and there's negative consequences of that. People who have drive and entrepreneurship, entrepreneurship and energy to achieve are often held back by their relatives who aren't achieving and don't accomplish things because the expectations on them in America and in Western Europe, that's less prominent and the same in Japan and developed world and it's becoming less prominent in China. So I would say those are two aspects of Africa that unless you've lived it, you really don't see it up close and personal.
- [09:04]
A
Absolutely. Very interesting. So you have lived your life in the period that a lot of people would call postcolonial Africa. And this is a very. A lot of people have a lot of narratives about postcolonial Africa. Basically as, as European authorities start to leave Africa in the 40s and 50s and 60s and such, you get all. And the, the natives, the, the people living there starts to take over the reigns of power. There's a lot of narratives about postcolonial Africa. When you look at it, I wonder if you could tell me what is your assessment of how Africa has transformed in the, you know, since the 60s or so.
- [09:44]
C
Well, that's a good starting point. I mean, technically the first independent state from the post colonial era would be Sudan in 56, although Ghana likes to take Credit for it as the Black Star Republic in 1957 under Kwame Nkrumah was actually Sudan in 56. And there were actually a couple of countries in Africa that were never colonized. And that would be Botswana, which was just a protectorate because they asked for the Brits to protect them from the threats around them. And so it was never really colonized as such.
- [10:10]
A
All right, the next clip comes from a discussion I had with Paul Rosenberg, previous guest of the show, multiple time guest of the show. And we talk about Liza Faire City, but we never really talked about how Li Zai Fair City, this amazing cypherpunk city that they carved out in Latin America, how it imploded. And Paul gave some insight in this discussion to that very question. We have talked in almost every one of our episodes together about the Eliza Affair City, the very interesting project that occurred decades ago. You have a book on it that people can track down if they're hungry for more, called the Untold Story of the Greatest Crypto Project Ever. Talked about it a lot over the past episodes, but I don't think we've really talked about how it's unwinded and. And maybe if we're in a little bit melancholic mood today, how did that dissolve? And were there any lessons to be learned about the way that that project not just flourished for a time, obviously we've discussed that, but how it kind of dissolved.
- [11:13]
D
Sure. That's actually some important stuff in there. What happened is that to get the project done and to get it done properly, a guy named Orland Graby and his partner, who I saw his real name once, but I chose not to remember it, they had to fight very hard to get the system built right. Because the other guy who funded a great deal of the project and ran a great deal of it, just didn't give a darn about security. He just didn't get it. He was an older guy, kind of a gruff, old time kind of guy, and he just didn't get the security thing and wanted to build and get it done now and give me a damn programmer who'll get it done. I want it up and running, you know, that sort of thing. So Orlin and his partner had to fight very hard against it. And as is very common in our species, they got pissed off and kind of crusted over about it. And they fought. They had to fight very hard to get the thing done right. And to their immense credit, they did. But they rather insulated themselves from exterior input given, you know, human psychology. Perhaps that was necessary. It was not, however, useful. So they had this system. It was up and running. It did work for quite a while. It worked well, by the way. But eventually they started getting into things where they were out of their depth. They were both brilliant, brilliant guys, but they didn't have experience in everything. And it turns out that they had no experience in banking. They could understand the numbers and how you manage things in general, but they did know the intricacies of the international monetary systems and how they worked. In particular, they didn't know that bank wires can be recalled. And there was a scammer using the system. And anytime, a warning to anybody who builds such a thing in the future you're going to have scammers find your system and try to use it. I wish I had a perfect fix for you, but I don't. It's just the human condition right now. There are certain number of people who, who are thieves that would rather thieves rather steal by brain power than. Than fist power. And you will eventually get them. And I'm sorry and do your best and you know, God be with you. Anyway, there was a particular guy who was using this system and doing some kind of scams. I think it was with selling used cars or whatever it was. It was some kind of scam. And these guys didn't know that bank wires could be recalled. And they weren't monitoring what everybody was doing. They were not surveillers. They were running a financial system, they were running a market. And markets in their pure form have no morality attached. They're just mechanisms of interconnection. So that's what they were running. Well, they allowed money to go out when they shouldn't have. And so the scammer got a lot of money.
- [14:39]
A
And.
- [14:41]
D
Afterwards the wire that, that went to the financial system was recalled because of the crime involved. Well, that put them insolvent and they scrambled and they tried to do everything they could, but they, you know, the money that they thought was theirs was actually recalled because there wasn't a crime associated with it, which I think was a real crime. They eventually had to just shutter the thing. And then, you know, everybody, they tried to pay everybody back. Everything that had been lost in the system and it was. There was a lot of arguing and anger at that moment about that. It wasn't their fault. And I did a lot of very serious investigation into this. I was given complete access. I knew, or at least just about complete access. I knew what was going on. I examined the paperwork afterwards.
- [15:42]
A
Class clip comes from a gentleman named John Robb, who is a futurist war strategizer, and he's an intriguing person and for sure, he wrote a book called Brave New War. And I asked him in this one, in this final question about the future of AI and its use in warfare. Actually, I was going to ask as a final question here, Mr. Rob, you've been talking about AI recently, and I wonder if you could continue what you were saying just now and maybe expand on that basically to give a fully John Robb response. Making people understand how terrible, terribly wrong things could go with AI in the mix. Just kind of paint a little bit of a picture of how bad things are going and could go in the near future.
- [16:30]
E
Okay, well, here's something I'm in the process of writing about because it's filling my windscreen right now, and I'm like, it seems almost inevitable. Is that okay? So the basic models, the LLMs that we've built, the reasoning models, have hit the wall, okay? And they're not getting any better in a meaningful way. And they're good enough for most interactions that people have with them in terms of reasoning. Problem is they don't. They're hard. It's hard for them to do something useful. So the big focus now in the last year or two has been on creating agents or companions, you know, AIs that have persistence over time and they can maintain their focus. Agents that can actually go do something useful and do it in a kind of a way that's. That's better than just writing a software program to do it. You know, book your flights, manage your life, do that kind of things, and also, you know, act like a therapist or act like a tutor and not drift off. But they're having, they just, just recently, just in the last couple months, they hit a wall with that. They aren't able to make it persist more than a. I think 300 minutes was the. Was the longest one, but it's relatively short before they kind of entropic decay just drifts into entropy and starts to drift off the topic and become unreliable, starts making things up. I'm personally, I've been working with my son on a different approach to that. It's not working on the. It's basically using the big models as a baseline. And then we found a way to actually get AIs to persist in a useful way as either agents or as companions, coworkers for months or years. It's basically indefinite. It's pretty amazing. We're just trying to figure out how to actually package it and not Just tell everybody and they run off with their billions. Like what happened to me back in social networking in 2001-2003. Basically built the thing and other guys capitalized on it. But it shows where it's told me. Give me an insight into where things are going and that you'll have AI workers in every field. And many of those will be AI workers that are tied to robotics. So you have everything from the maids to factory workers, dock workers that are humanoid. Robotics and humanoids. The humanoid robotic form is proving to be the easiest for those companies to build. It's, you know, it, it, it's easier for them to actually build those shapes. And, and on top of that, it works in all human built environments. So it could flow into any job. What happens in that instance is once this persistence problem is solved, and it's going to be solved in the next five years or so, these AIs will grow in number from 10 to million to 100 million to a billion, to 500 billion to a trillion. Really, really quickly. It's possible. And here's where it gets really wild is that if you allow these AIs to act as independent operators in the form of a corporation, they incorporate, become kind of a. Corporations have similar rights as an individual. They have, they have rights as if they were individuals. They can earn money and make money and spend it and they can do their work for others. Or they can recombine themselves into corporations that will serve as, you know, specific, do a specific task and then break apart again and work independently again or join with some others. But if you have a trillion AI persistent, I call them social AIs. A trillion social AIs able to make money and spend money. You've just expanded the economy to a level that we never even imagined. And if this happens outside of standard national borders. One thing I was also piecing together little bits and pieces I've been pushing for. Solar arrays can do generate power in orbit 24, 7 at many times the levels of efficiency that you get on terrestrial solar arrays powering data centers. And you put the demand that the fastest growing new source of demand for power, you put them into space and they do their inference there is that they would operate outside of the standard national laws, the economy created by them. As fast as you can put up a new solar array and create a new data center, you get, you'll get new AIs that they could become in short order within 20 years, 90 in 20 years. Look, if you think it's that short, I mean think in 19 or 2007, there wasn't a smartphone. And in 2022, 15 years later, 5 billion people had smartphones. And everything's changed. So in 20 years from now, you could have trillions of AIs either operating robots terrestrially or in space, or operating as virtual workers in an economy that would be 99.9% of the global economy. And any human beings that are tied to that one thing. We found that human cooperation or collaboration with these AIs radically increases and improves the quality of the AIs, the social AIs ability to do things productively. Anyone is involved in that economy will be dragged along with it. And everything else on earth would be cheap or a pittance, or it'd be almost useless. I mean, you know, it could be bought. Yeah, things could change. Like, it'd be kind of what we call it economic singularity, is that everything changes overnight and all it takes is somebody like Elon or Bezos setting that up in space and letting it go. Because you, I mean, if you're running a corporate, how it would work is if you're running a corporation in the United States and you wanted to hire a thousand AI workers, they would be coming from that orbital facility. That's where the, that's where their inference would be done. Because the power is, that's providing them that running those GPUs would be plentiful. And if they're successful, they would buy themselves more energy and get more inference capacity or more tokens, making them even better at what they're doing. But you would be hiring them and then they would be actually working as an outsourced worker. But anyway, I mean, so, yeah, no, that, that. Yeah. I mean, it's funny that the reason I'm tossing it out there, I'm thinking about it, is that it's one of those things that just the right combination of things could happen. It would cause everything to shift so quick, and everyone who's associated with it would be wealthy beyond measure compared to everybody else. And people who aren't so associated with it or trying to hold onto it and tried to own AIs as channel are going to find themselves on the short end of things.
- [23:58]
A
Okay, so these are just a few of the episodes that you can find in full by being a Watchman's Torch newsletter subscriber. That is the future of the brand. I think, you know that it's the right thing to do to purchase that, support the show, get your excellent information from the newsletters, and also get access to these outstanding episodes that are still happening when they do happen@watchmanstorch.com.
- [24:41]
D
Sam.
- [25:07]
E
It.