NET Society

The Net Society crew is joined by John from SuperRare for a wide-ranging conversation on AI agents, headless software, crypto infrastructure, and the strange future of machine-mediated markets. They open with AI psychosis, neurotic models, compute subsidies, and the rising push toward CLI-first products, before exploring what happens when agents become the primary users of SaaS, payments, marketplaces, research tools, and crypto rails. From there, Aaron walks through a set of 2035 and 2040 agent economy predictions, including autonomous micro-enterprises, agent procurement, machine capital markets, self-sovereign AI, robotic factories, and AI-managed civic infrastructure. The episode then turns toward internet culture, Brian Johnson, peptides, crypto’s identity crisis, NFT market momentum, DeFi hacks, AI-assisted hacking, and the question of what new digital culture could actually bring buyers and creators back into the room.

Mentioned in the episode
Guest John from SuperRare https://x.com/SuperRareJohn
Rare protocol https://rare.xyz/?utm_source=chatgpt.com
Bankr https://x.com/bankrbot
Brian Johnson Tweet https://x.com/bryan_johnson/status/2049687845082910812

Show & Hosts
Net Society: https://x.com/net__society
Aaron Wright: https://x.com/awrigh01
Chris F: https://x.com/ChrisF_0x
Derek Edwards: https://x.com/derekedws
Priyanka Desai: https://x.com/pridesai

Production & Marketing
Producer/Editor: https://x.com/0xFnkl
Social: https://x.com/v_kirra

What is NET Society?

NET Society is unraveling the latest in digital art, crypto, AI, and tech. Join us for fresh insights and bold perspectives as we tap into wild, thought-provoking conversations. By: Derek Edwards (glitch marfa / collab+currency), Chris Furlong (starholder, LAO + Flamingo DAO), and Aaaron Wright & Priyanka Desai (Tribute Labs)

00;00;16;11 - 00;00;26;12
Aaron
Hey, Chris, did you go to the psychosis event last night? What do you think it was like? AI psychosis event? Yeah. Oh, yeah. Sorry. I mean, LM psychosis. I hosted.

00;00;26;12 - 00;00;28;13
Chris
It in my head. What do you think they.

00;00;28;13 - 00;00;29;08
Aaron
Talked about there?

00;00;29;09 - 00;00;38;03
Chris
I don't know. I saw a little a few things on the timeline about it. Some pictures. It. It would have induced psychosis in me. I'm not built for.

00;00;38;06 - 00;00;38;18
Aaron
How many.

00;00;38;19 - 00;00;44;16
Chris
People's boxes filled with screens, I don't know. Looked like a few hundred. It looked like it was a very full room.

00;00;44;17 - 00;01;02;08
Aaron
I can't tell like. Like how real that is. Like, is that real? Or people, like, really that deep into it where they just have like other issues and it kind of manifests in a weird way. I can't get a clearer read on it. I mean, the branding of it is amazing, but I can't tell if that's actually going to be a real thing.

00;01;02;09 - 00;01;29;01
Chris
I think it's a little column, a little column B, you know, it's definitely a transformative thing. It certainly changes the way, you know, you go about your day, you do work and all of that. And it can be very, very engaging. I also think, you know, if you're prone, you know, to mental drift or being extremely affected by, you know, your environment, it's going to be a force multiplier.

00;01;29;01 - 00;01;31;12
Chris
So I think it's a little column, a little column B.

00;01;31;13 - 00;01;46;07
Aaron
Do you know anybody that has an AI psychosis yet? I feel like I feel like that would be like a watershed moment when it becomes like if it is actually a thing, it becomes something so large that it impacts everybody by like 1 or 2 degrees.

00;01;46;08 - 00;02;01;26
Chris
I mean, Aaron, that would assume I left my basement. I knew people that maintain social relationships and which I could draw upon the experience of others. So I don't know anyone within the AI psychosis yet because I have no friends except for my CLE.

00;02;01;29 - 00;02;06;00
Aaron
Except for your AI. Oh no. Maybe. Maybe you're in the middle of it. Chris.

00;02;06;01 - 00;02;08;19
John
We just have one friend now. It's Claude.

00;02;08;20 - 00;02;11;14
Aaron
Oh, hey, John, we got John from super on this week.

00;02;11;15 - 00;02;12;25
John
What's up Aaron? What's up Chris.

00;02;12;26 - 00;02;33;13
Aaron
I find the latest model. Like, if that is everybody's only friend. It is so neurotic. Like, I feel like I'm, like, around my, like, my family or parts of my family when I talk to it. It's like such a nervous system. It got me thinking that like, maybe alignment is just like making these systems, like, more nervous and neurotic.

00;02;33;14 - 00;02;42;17
Aaron
Maybe that's just not human. Stay in control. You just get these, like really like beta beta AI systems that just defer to you on everything.

00;02;42;18 - 00;02;44;18
John
Learning from all the neurotic humans.

00;02;44;21 - 00;02;45;08
Aaron
So yeah.

00;02;45;08 - 00;02;47;16
John
You think nothing less. I think so.

00;02;47;18 - 00;02;51;14
Aaron
Oh, man. Do you know anybody with an AI psychosis, John?

00;02;51;16 - 00;02;59;02
John
Not yet, but I certainly know a lot of people who are staying up too late using Claude, trying to max the subscription.

00;02;59;04 - 00;03;19;00
Aaron
Yeah, completely. Yeah, it's. I'm waiting for all that subsidy to kind of go down. I saw, even outside of anthropic subsidy for cloud code, I think cursors been doing the same thing. So I wonder when like that goes away and maybe there's like this like reset, like we saw with Uber where like the price of Uber's like quadrupled at some point.

00;03;19;01 - 00;03;23;17
John
Yeah, I could see that happening. The cursor XII deal is pretty wild.

00;03;23;19 - 00;03;39;16
Aaron
Yeah, I thought that was super interesting. Just kind of really highlighted to me like how much people need compute. I heard last night that like anthropic is using more of Google's compute than like Google is itself, or at least that's a rumor. Which if that's true, that's like actually kind of wild.

00;03;39;17 - 00;03;40;22
John
Yeah. That's insane.

00;03;40;23 - 00;04;03;07
Aaron
Yeah, I don't know. It just feels like it's all going to be about compute now. And I guess if that Leopold guy that left OpenAI to start his own fund is right, like, well, at some point get like somebody trying to build like $1 trillion data center. That's what he was projecting, which is kind of wild. I saw like KKR, the big private equity firm, was announcing like a $10 billion AI infrastructure platform.

00;04;03;07 - 00;04;33;29
Aaron
So they must think that there's more room to run here just in terms of the buildout. Yeah. It's nuts. Yeah. I mean, the other thing that I thought was interesting this week is just the move by some of the bigger companies to go pure headless. So if folks didn't see it stripe made parts I think they've already had a CLI, but they added added to it or developed a new product that enabled your agent to basically like pay via stripe, and then you get back in your console like a link that you could click to kind of approve it, which I thought was pretty cool.

00;04;34;00 - 00;04;50;02
Aaron
Meta made some moves to enable you to manage via their CLI, some of some of your ad spend, which I thought was pretty interesting. I imagine lots of folks would want to do that, like some of the growth marketing marketing types. And even John, I heard super is doing the same thing. You guys are going headless.

00;04;50;05 - 00;05;08;24
John
Yeah, about a month ago we launched the rare CLI, which is more or less a wrapper around the smart contracts and kind of the other features we have on the website. But yeah, I'm super bullish. If you think about Unix, you know, which is like kind of like the heart and soul of all the agents. Anyway, it's all text in, text out.

00;05;08;24 - 00;05;30;15
John
So it just makes so much more sense that, you know, that's kind of the future and especially kind of like any utility, it's more natural that you would tell the app what you're trying to achieve than like knowing where the buttons are. It's like we probably all use the app or like one of the buttons we move, or one of the buttons we use moves, and then you waste two minutes trying to figure out where the heck it went.

00;05;30;15 - 00;05;40;02
John
So yeah, I'm super bullish on kind of this like headless API first for like all, you know, SaaS products and kind of like utilities.

00;05;40;03 - 00;05;59;14
Aaron
Yeah, definitely feels like the SaaS piece will be there. I wonder like if parts like more parts of crypto look that way, like does like Uniswap soon have like a CLI and you could just log in transactions that way via Uniswap? Or does it just keep the smart contract interface? Like what? Like what's the right interface for kind of that.

00;05;59;14 - 00;06;01;07
Aaron
And digital assets?

00;06;01;09 - 00;06;23;08
John
I think it's both. But you know, it's like I've been playing around with Banker Bot for a while. Like they kind of famously launched with you could do swaps in the Twitter feed and they've been doing ton like it's all like they have a bunch of like LM gateways and you can do almost any crypto interaction. Totally like natural language, which you know, feels feels very nice.

00;06;23;10 - 00;06;23;20
John
Yeah.

00;06;23;22 - 00;06;43;03
Aaron
I didn't realize that. And how do they maintain like the security pieces of it? Like we heard from a team this week that they can get like I think it was like 98% like success from their system and just like teeing up transactions. So I feel like there's still like a pretty wide degree of error.

00;06;43;07 - 00;06;57;17
John
So yeah, I think you still need some amount of human in the loop. But just as far as like expressing the intent of what you're trying to do, that's certainly the direction. But yeah, I mean, security is huge when the robots can start moving all the money.

00;06;57;18 - 00;07;16;14
Aaron
Right, which they're definitely going to want to move all the money. So bankers kind of ahead. But is everybody going to follow or is there going to be like a winner like in that. Like is that like bankers core product. And you know x number of years. Like that's what we kind of know them for. Or does everybody kind of go in the same, same direction?

00;07;16;16 - 00;07;34;06
John
Yeah I think everyone goes the same direction. And then I feel like you also end up with a lot. It's like fee compression if you can, you know, it's like to the robots, like they don't care as long as they get the right expected result, they're not really going to care what system they're using as long as it works.

00;07;34;06 - 00;07;40;13
John
So it'll be super interesting to see how kind of like the software modes, like how does that all change and what does that look like?

00;07;40;15 - 00;07;57;15
Aaron
Yeah, completely. Chris, it's kind of funny because I remember at the end of last year you were like, oh, I'm canceling all my like subscriptions. Like all I really want is like API access. It feels like that is becoming more of a drum across the industry. Like doesn't matter which industry it is. Just like we're going headless.

00;07;57;16 - 00;08;20;08
Chris
It's. Yeah. No, it's definitely a push for I mean efficiency. But it's also you know, you just altering around your workspace into, you know, the form factor that you're spending the most time with. I mean, in some ways it is a bit like, you know, the shift to mobile, where I mean, old school landlines have much better voice quality than a cell phone.

00;08;20;08 - 00;08;46;05
Chris
And we gladly gave that up for, for portability. Right. And so I think whenever the trade off just becomes like efficiency, concentration, you know, that results in productivity. That's that's going to be the move. What I really need is I need a headless NHL.com so that I can get my playoff hockey scores because their website is loading so damn slow.

00;08;46;05 - 00;08;59;02
Chris
And so, you know, I'm having these moments where I'm like, oh man, I'm watching the Celtics. I throw the the hockey game on and you're just like, why on earth of the site acting like 1999? So that's what I need. I need headless hockey.

00;08;59;02 - 00;09;22;23
Aaron
I want like have this everything right? I just feel like if you did have access in some sort of way to almost every traditional service you use, it would be great, right? Whether that's your email, like slides, you know, even a like a spreadsheet or the ability to chart or to like, you know, even OpenAI and anthropic, they're still not that easy to interact with, right?

00;09;22;25 - 00;09;44;02
Aaron
Like you probably want like, like an orchestrator sitting in the middle that could send you some more complex things. So, I don't know, I definitely like feeling it. I wonder if that's kind of what like saves or staves off the SaaS polyps and all these services just increasingly move in that direction? I'm with you, Chris. Yeah. Like maybe the next ESPN is like headless ESPN.

00;09;44;04 - 00;09;53;15
Aaron
Like maybe they didn't get around to it. But you know, one of those like deeper database sports sites do whether it's like, you know, I'm thinking more baseball like baseball reference or something like that.

00;09;53;18 - 00;10;16;29
John
Yeah, I definitely think it moves headless. And that'll be interesting. Like trying to figure out the the business models. Like I'm not totally sold on this aspect whips, but you know, maybe it's like, you know like the per seat if you just if you're routing all requests, you're using slack and then you're interacting with, you know, your CRM via text, there's really there is only one user then.

00;10;16;29 - 00;10;32;25
John
And so it's like hard to measure, like how do you know how many different people are using it? So it's like that's maybe that's where like kind of like the X4 or two API payment stuff takes off, or he doesn't even have to be on chain. Yeah, exactly. I think there's definitely more than just hype there.

00;10;32;27 - 00;10;37;10
Chris
Thank God the micropayments crowd has a shiny new hope to cling to.

00;10;37;13 - 00;10;56;19
Aaron
It's like the quest for longevity, Chris. The search for the fountain of youth. But maybe we'll get micropayments at some point. I kind of feel like we may. We may though, because, I mean, what's the big issue with micropayments? It's just like nobody wants to feel like they're, you know, quote unquote nickel and dime, and the agents aren't going to give a shit about that, right?

00;10;56;21 - 00;11;19;18
Chris
I mean, that's one issue with micropayments. The other issue is who on earth wants to pay $0.10 for a news story or, you know, $1.20 for long form? It's just that information wants to be free and you don't have to pay to speak. You don't have to pay to read. It's just very antithetical, I think, to the form factor of information services is different, right?

00;11;19;20 - 00;11;33;02
Chris
Services like has a much deeper grounding in concepts of like labor work, you know, compensation. And so like you can get your head around that. But I don't know, can you imagine being charged to Doom Scroll?

00;11;33;02 - 00;12;04;00
Aaron
I can't imagine being charged $0.10, but I could imagine being charged $5 a month if like that was the aggregate. I feel like that almost would be the ideal situation. Like, almost like I've thought about it, you know, like, I don't fully understand why the internet's not like driving a car where you, like, fill up your browser. And it's just like routing payments like left, left, right and center, I just do, I think it's like, I think people would definitely trade that for like the, you know, like this paywall, you know, pay for the your seat.

00;12;04;01 - 00;12;09;25
Aaron
Like John was saying before type approach, I think people would much rather treat it like gas in a tank.

00;12;09;27 - 00;12;22;00
Chris
Hey, no, this is kind of beautiful. It's also like possibly the most boomer coded thing I've ever heard since the internet is a series of tubes. So congrats on like, getting both ends of the spectrum there.

00;12;22;06 - 00;12;26;22
Aaron
I'm here for you, Chris. I didn't realize I could do that in one fell comment.

00;12;26;23 - 00;12;28;28
Chris
You you ran the whole continuum.

00;12;29;00 - 00;12;32;03
Aaron
Should we talk about VCRs next to Chris? What should I do?

00;12;32;04 - 00;12;49;24
John
Yeah, they're coming back. But, like, with the micro, it's like. I feel like one thing that's interesting with AI is it's kind of like getting people used to paying for API credits, whereas that was like a way more, you know, developer type thing to do. And it's like, I feel like it's way closer to the consumer than it was, you know, a couple of years ago.

00;12;49;25 - 00;13;12;15
Aaron
Yeah. And I can't tell John if it's also like, okay, you can't pay per seat. So there's just like a repricing on all this stuff, but there's going to be so many more agents and they'll just make up in volume. Right. Like you're going from like like the seat pricing I think to like a volume play. And if they're if everybody at some point has their own agent, which I think will happen not this year, right.

00;13;12;17 - 00;13;36;16
Aaron
Or even next year, but maybe a decade into the future. Like you're going to have a lot more users of of your services, right? Like potentially magnitudes more, I don't know, like it feels like maybe that's that's kind of how that whole like enterprise service, task based software like, finds its way out of this, this AI apocalypse.

00;13;36;18 - 00;13;57;00
John
It's interesting to think to even like consumers like if, you know, even like flight finding or things like that, like how does the how does all that change? Because there's still lots of fee compression, right? Isn't like this is like years ago when we were back in consensus. Like looking at how like hotels and flights are booked. It's kind of like flat fee, like flat finder's fee type thing, right?

00;13;57;02 - 00;14;10;12
John
I wonder if it's like if everything does have better APIs and it's easier to do that searching like before, kind of the values and all the glue where you needed. But if it's not really hard to write that glue, where like, what does it do to that business model?

00;14;10;13 - 00;14;18;12
Aaron
Yeah. But I think would you pay like a dollar to like, you know, have your agent book you the exact flight you want? Probably. Right.

00;14;18;13 - 00;14;34;20
John
Yeah. Especially if you don't have to click the buttons. You know, if it's like, okay, when I go to Google flights, I spend a bunch of time looking at them. And if you're just like, I want the best flight with the lowest layover, cheapest price, you know, like that's probably still cheaper than the $12 you're getting charged under the hood.

00;14;34;21 - 00;14;52;06
Aaron
Yeah, right. Like, I think if you could get rid of some of that glue and then maybe the end user pays like a little bit for it, then you'd make up for it in volume, right. Like you would never go to Google flights or pick your favorite travel destination. You you just would say, yo, agent, you know, book me the flight.

00;14;52;06 - 00;15;07;19
Aaron
It will have memory of like the, you know, the the airlines you may like or the types of hotels you may want to stay in. And just like done. And that costs not that much and happens pretty quickly like that. That feature feels pretty good to me. I could see that happening.

00;15;07;21 - 00;15;09;01
John
Yeah, 100%.

00;15;09;03 - 00;15;15;20
Chris
Look at you, man. Wanting to like, travel, to go places, to leave your basement. Like, what is this lunacy?

00;15;15;24 - 00;15;21;02
Aaron
We may need to, Chris. You know, New York, New York's in a budget crisis. Who knows what happens here?

00;15;21;07 - 00;15;28;18
Chris
Oh, God. We're always in a budget crisis. We're also in a baseball crisis. So crisis is abound.

00;15;28;22 - 00;15;29;22
Aaron
That's also true.

00;15;29;24 - 00;15;31;13
John
What's the baseball crisis.

00;15;31;19 - 00;15;33;20
Aaron
For teams? The Mets are horrible.

00;15;33;21 - 00;15;34;19
John
Got it. Yeah.

00;15;34;20 - 00;15;39;23
Chris
No, no, they're not just horrible. They're literally the worst. We have the worst record in baseball, I think at the moment.

00;15;39;23 - 00;15;46;28
Aaron
So bad. But at least the Knicks like, mopped up last night. I don't know if you saw that. They like okay. So that 51 that was an insight.

00;15;46;29 - 00;15;57;22
Chris
Was so nuts because I threw it on with my son and we had to like double check the score because they were up 50 in the second. It was just like inconceivable.

00;15;57;22 - 00;16;04;23
Aaron
It was a wild game. So we're winning still a little bit, I'd say, Chris. So we'll get through this crisis I'm not worried.

00;16;04;23 - 00;16;09;23
Chris
About, oh, the Knicks are just setting you up for an ECF finals heartbreak, you know that?

00;16;09;23 - 00;16;26;17
Aaron
Yeah, completely. But I'll enjoy it as we go down that path. You know what I actually did the other day, which I thought was kind of fun. I asked a bunch of AI systems, like what agent agent stuff will exist in 2035. I thought there was like some interesting concepts related to it.

00;16;26;18 - 00;16;27;10
John
What they say.

00;16;27;11 - 00;16;55;05
Aaron
So one was an autonomous micro enterprise, which I would kind of recap as like a Dao. It said, you know, employees, agents for sales, marketing, fulfillment support, automated treasury, dynamic pricing, self optimized cost structure. And some of these will generate millions in revenue and humans will own them. Set guardrails, intervene only when alerted, and that the agent economy will lower the minimum value viable company size to near zero.

00;16;55;05 - 00;16;57;22
Aaron
So I thought that was kind of interesting.

00;16;57;24 - 00;17;17;29
John
I mean, they're they are good at like, you know, that type of like management type stuff. So I could totally see like one. I don't know if you ever talked to Kiren about this back in the day, Aaron. But like we were talking a lot about like, you know, like, what if Unix utilities, you could just like, it's like pay for like, you know, as you're like, piping together different products and services.

00;17;18;00 - 00;17;36;29
John
Like I'm always thinking about, like the future of agents where it's like, you could that's where the micro payment stuff could work. Like I agree, Chris, paying for like an Instagram reel is inconceivable. But if there's like agents trying to do things, that's where like the cost efficiency of like, okay, this transaction only cost like a fraction of a penny.

00;17;36;29 - 00;17;42;03
John
And I'll just charge a small markup for doing a small service. It could be pretty interesting.

00;17;42;04 - 00;17;52;17
Aaron
Yeah, I think so. Right. Like you could imagine, like, just because maybe my legal background, like setting up a company. Right. And it costs not $400 a cost like a dollar. Right. And you just.

00;17;52;19 - 00;18;15;20
John
Yeah. Or like treasury management, it's like it's pure, like if you're changing rather than, like, okay, there's a fixed fee for, like, humans managing this basket of assets. Like just anytime you make a small, you know, agents proposing a change, whichever one you go with, they just take a small clip on the transaction itself, rather than like a flat fee for stuff just sitting there.

00;18;15;26 - 00;18;19;20
Aaron
Yeah, that feels right to me. Chris, do you think we'll get there by 2035?

00;18;19;27 - 00;18;30;00
Chris
Will Vitalik finally win? Will everything be on main net? Will it be secured by eath? Will true DAOs finally rise? Possibly.

00;18;30;03 - 00;18;31;13
Aaron
Yeah, I think it's going to.

00;18;31;15 - 00;18;52;00
Chris
I mean, the thing we just can't tell is, is the scale of this stuff. Like it can exist and, you know, really be the equivalent of like, lifestyle businesses or, you know, little niche things people run or, you know, we could we could just be like blown away in the k k shaped economy. Continue to K and who knows.

00;18;52;04 - 00;18;52;20
Aaron
Yeah.

00;18;52;21 - 00;18;56;11
Chris
That feels you're not going to pin me down in the future anymore. I've given up.

00;18;56;13 - 00;19;14;28
Aaron
Yeah. All right. Here's another one I came up with, which I thought was a good one. It was autonomous procurement networks, which to me makes a lot of sense. I think it's like you ask for something and then the agent has a budget, your needs, constraints and preferences, and then they go out and find you like a good or service that you kind of need.

00;19;15;00 - 00;19;41;08
Aaron
And they would do some like negotiation with another agent, some budget aware decision making, some trade offs, and maybe they'd even like sign, let's say like a terms of service if it's something that required that or some sort of contract. But if it was like e-commerce, like I feel like that's what we'll see pretty soon. It's just like somebody that nails kind of like an almost like a, like a universal shopper for you in some capacity.

00;19;41;10 - 00;19;42;00
Aaron
I think.

00;19;42;02 - 00;19;47;20
John
Google did put out some kind of like universal commerce protocol, like UCP.

00;19;47;20 - 00;19;52;23
Aaron
Yeah, it was ACP. I would forget that. There's like so many competing protocols.

00;19;52;23 - 00;20;10;14
John
Yeah, there's there's a bunch of them. One that would be on like consumer procurement. Like whoever makes like I don't want to have to look at Facebook Marketplace, but I just want the brown couch and I want it delivered. You just tell your agent it goes and figures it out so you don't have to, like, interact with tons of people on Facebook Marketplace who aren't actually going to sell you.

00;20;10;14 - 00;20;12;29
John
The thing seems like it would be very popular.

00;20;13;00 - 00;20;13;05
Aaron
Yeah.

00;20;13;06 - 00;20;16;12
Chris
People are on the couch. Is it going in? Your conversation fit?

00;20;16;15 - 00;20;17;17
John
Exactly.

00;20;17;21 - 00;20;46;03
Chris
Everyone has a brown couch. It's like just just a rule. I think it'd be far more likely that we see a headless Carvana assuming we owned cars in ten years. You know, something in which, like, we've kind of already worked out a lot of the mechanics of this and streamlined it in a ton of different ways. Like, you could certainly see AI agents haggling over sticker price, you know, because I think that's like a known enough ritual.

00;20;46;05 - 00;21;06;02
Chris
And then, you know, I could certainly see someone sticking like a 400 square foot garage full of brand new cars that, you know, when you buy one, it just self drives over to you. So if you want me to go big on the big headless thing, I can think of the headless, headless Carvana.

00;21;06;03 - 00;21;30;06
Aaron
Headless car dealerships. Yeah, that'd be amazing. All right, here's another one. I thought this one was interesting. It was an agent agent market maker where the agents are data providers, their model inference providers, the latency optimizers, they can score risk, and then they have some arbitrage capability, and they post bids and asks to basically like by like compute.

00;21;30;07 - 00;21;37;07
Aaron
It's kind of like because I'm I guess I'm playing the boomer. It's kind of like Enron on steroids. But for compute.

00;21;37;08 - 00;21;38;25
Chris
This, I can fully see.

00;21;38;29 - 00;21;39;13
Aaron
You can see.

00;21;39;13 - 00;21;56;26
Chris
This, I can. Oh yeah. No, I've like plotted this out in my head before because it's very it's got enough like similar shades to Telecom which you know, I did in the prior life. And like I just think information market's going to behave a lot like communication markets. And so I can see you like putting a job out.

00;21;56;29 - 00;22;29;01
Chris
You know having someone break that into into different requirements. And then those requirements are going to need like both a combination of like specialty model providers and compute and that all these things like go out for bid and you know, you basically are going to set a sweet spot between, you know, some some form of quality score and price and that, like, all this work will get like bundled out into, you know, orchestrators, specialists, service providers, you know, like a 3 or 4 level deep stack that I can totally see.

00;22;29;03 - 00;22;47;12
Aaron
Yeah, I can see this one too. I wonder when it materializes, though, because it just feels like people would pay more for compute at different points in time already. Right? So I feel like the demand for that is just going to increase over time. I can't see like where it lives though. Like, is this a crypto market? Is this somewhere else?

00;22;47;13 - 00;22;48;16
Aaron
What do you think, John?

00;22;48;24 - 00;23;09;24
John
It seems like it would need to be super low latency. So I was having a long combo with one of the robots about just like like if you did have like auction markets and stuff like related to Super Rare, it was like when I was having the conversation. But just like if you're like negotiating, like your text to text, be a robot, it can happen way faster.

00;23;09;24 - 00;23;20;28
John
And so like, yeah, it's a good question of like, it seems like it has to be off chain negotiation, potentially on chain settlement. But yeah, it seems like the latency there is going to be like a big driver and super important.

00;23;21;00 - 00;23;40;23
Aaron
Yeah. Massive. Yeah. It needs to be like instantaneous right. Another one autonomous research swarm I mean this hit with me because this is kind of how I view this product that we've been building called Aidan Chat. So the idea is this agent would research, verify, synthesize, visualize and publish, just like research and just charge a micro fee for work.

00;23;40;23 - 00;24;04;01
Aaron
It would evaluate the quality of other agents, you know, select potentially like human collaborators based on reputation. And then just like optimize on the cost on the cost side. So I definitely think something like this is going to happen. Like you'll just hit like an like some API endpoint and get like real research, not just like a shitty response from ChatGPT or Claude or one of the hyperscalers.

00;24;04;01 - 00;24;12;11
Aaron
I just think that I don't know about you guys, but I just feel like the answers are always kind of math to me. If it's not, you know, just something basic.

00;24;12;13 - 00;24;19;25
Chris
I don't know. I mean, I want to pay like $800,000 to a consulting firm that will come back and tell me to fire people.

00;24;19;27 - 00;24;21;05
Aaron
To you.

00;24;21;07 - 00;24;23;27
Chris
Why do I want this? Like free and cheap and fast?

00;24;23;27 - 00;24;43;09
Aaron
I just think there's so much, like, additional little research tasks that people would do if the costs went down. Like, like there's no reason why you wouldn't do, like, full research memos if you if it was costless on like so many different topics. I think most tasks you you know, maybe this is the knowledge production of the legal industry which I've seen.

00;24;43;09 - 00;25;00;11
Aaron
But like I think that's what like high end legal work really, or what differentiates high end legal work from, you know, mid or lower tier legal work is just the level of like research that goes in before they get to an answer. And I think that's where they get better answers. So I think that that's just going to hopefully happen everywhere and probably like at a huge scale.

00;25;00;13 - 00;25;03;18
John
Have you played with the deep research stuff on Gemini or.

00;25;03;21 - 00;25;20;08
Aaron
Yeah, I mean it's good, but like, you know, I think it's not it's kind of like not good enough. And I feel like they're never going to spend time to kind of optimize that. They just want to give you the quick answer. Right. They want to be your go to place if you want to do like like really like lightweight quick research.

00;25;20;08 - 00;25;40;05
Aaron
I think some things just take a little a take a little bit longer to actually compile good work and be, you know, you want to make sure that there's not mistakes or it's like well cited, well sourced. I mean, I just think these llms are still very human like where they're really good but still have like that 5% to 10% that just off right where you can't trust.

00;25;40;05 - 00;25;46;21
John
And I guess their incentive is to like do the research as cheaply as possible. It's like they don't really care if you're willing to pay premium.

00;25;46;21 - 00;26;01;23
Aaron
I think that the AI systems will fan out in terms of, you know, like expertise. So like, you know, I don't know if you're going to go to Claude and know that you're going to get the best legal answer. Some people try to like benchmark this with evals or whatever. And, you know, like OpenAI is better for that.

00;26;01;23 - 00;26;22;19
Aaron
But I don't think that those subtle differences are like apparent to the end user. So you could see like an avenue where somebody does that mapping or knows that mapping or can actually get you like an answer. That's a little bit better just by playing around with different models, you know, adding in additional steps to make sure that the information is more accurate.

00;26;22;20 - 00;26;24;01
Aaron
You know, all that jazz.

00;26;24;02 - 00;26;36;08
John
Is this where the data mode comes in? I feel like everyone's talking about like, you know, like your personal knowledge graphs and stuff. So like it's like using the public models plus your personal preference knowledge graph. Yeah.

00;26;36;10 - 00;26;52;21
Aaron
Yeah. Maybe if you could like pull that in that maybe there's something there. I know people are geeked on the personal knowledge base for me, like in the way I do things, I just don't I don't want something that's like recording everything and like evaluating it. It just feels like too much. It's like two navel gazing for me.

00;26;52;22 - 00;27;18;05
Chris
Yeah. I don't I don't quite get it myself either. But like I also I don't take notes, I don't document everything. I keep everything in my head. And if I forget something, it's just easier for me to retreat, you know, like kind of net new discovered again. And so like, I look at a deeply human tendency. I just think this is one of those places where, like, I just I don't fit the pattern, but like, I can understand it.

00;27;18;05 - 00;27;19;16
Chris
I do think that.

00;27;19;17 - 00;27;20;17
Aaron
Yeah, 100%.

00;27;20;18 - 00;27;30;25
Chris
Yeah. I mean, there's just been a very strong drive amongst people to want to, you know, capture their lives to to have that accounting. Like it makes a ton of sense. Yeah.

00;27;30;28 - 00;27;35;17
John
Black mirror episode where every single second of your life can be retrieved.

00;27;35;18 - 00;27;39;08
Aaron
Yeah. Like something about that just like, irks me. And also, I don't know.

00;27;39;09 - 00;27;44;27
Chris
I don't want cringe on demand. I think that's the the big Sierra there, you know? Yeah.

00;27;44;28 - 00;28;10;16
Aaron
So are you ready for this was the the the last one. This is for 2030 then 35. And then I pushed the 2040. How about like an agent based like HR or bureaucrat or lawyer. So you you'd pop in any task and then it would evaluate it for legal risks, regulatory risks like anything that you wanted. Kind of like a quick read on whether or not it's permissible.

00;28;10;18 - 00;28;21;02
Aaron
You'd have to like root it through one of these agents. I, I personally think that this is definitely coming. I wouldn't be surprised if, like the government in 2100 is just like a series of these agents, to be honest.

00;28;21;03 - 00;28;31;01
John
Yeah. We've got a a hill over at the IRS. So maybe they're, you know. Iris audit cutting agent based.

00;28;31;04 - 00;28;57;02
Aaron
And the FDA just prototyped like this, like AI, like real time data plus AI analysis to get around some of the not to get around to streamline some of the, you know, drug approval processes, which a is super cool but be just like picture that like 100 times more pervasive and a thousand times cheaper. Like, I just feel like you're going to have to check in with your government agent before you do anything.

00;28;57;04 - 00;29;10;17
Chris
Oh my gosh, you imagine how hard it is to program that FDA AI? Like, it's literally a three line markup that says vaccines, bad peptides, good everything else. What does Joe Rogan think?

00;29;10;19 - 00;29;11;05
Aaron
That's it.

00;29;11;05 - 00;29;13;15
John
That's it. We're done. Chris figured it out.

00;29;13;17 - 00;29;22;04
Aaron
There'll be a committee that will tinker with that. Replace Joe Rogan with the next populist influencer. You guys ready for 2040?

00;29;22;05 - 00;29;23;02
John
Let's do it.

00;29;23;04 - 00;29;47;08
Aaron
So there was one here. I'll start with. I thought this was the most interesting. They called it Self-sovereign AI, which I thought was kind of interesting in this one. The agents themselves replace companies they hold capital owned intellectual property, issued contracts, employ other agents, and then reinvest their profits. So like they're not there's no human in the loop at this point.

00;29;47;10 - 00;29;56;18
Aaron
So by 2040, we'll just have no no need for even companies anymore. At least that's what the AI was telling me. What do you think?

00;29;56;19 - 00;30;16;09
Chris
I got some wild thoughts on this one. I think it actually ends up well. Yes. Like I think like the well front betterment, like all those, you know, sort of web2 algorithmic wealth managers like AI will take that over. I don't I don't know if that's going to become that big a category or if it does, it won't be all that influential.

00;30;16;09 - 00;30;47;14
Chris
I think where it will be really interesting is actually in like a very obscure niche of like ultra wealthy people who do not trust their heirs and also have huge egos and can't face the fact that they're going to die. And we'll place management of like, their fortune, their inheritance, etc., etc. into an AI system built on their own personality and biases, and that basically all their children are not going to be free to determine the futures based.

00;30;47;16 - 00;30;51;23
Chris
They're going to be prisoner to whatever drove daddy to make all his money. Yeah.

00;30;51;23 - 00;31;12;08
Aaron
That's wild. You know, there's there's a lot of rules right now on the books that came out of, like, you know, even like old England to prevent that type of construct, Chris. Like to prevent, like a, you know, like one of the, you know, one of your rich, wealthy ancestors from overly controlling what you can do on a going forward basis.

00;31;12;09 - 00;31;15;02
Aaron
I forgot the exact name of it. I'm blanking on it, but it's.

00;31;15;02 - 00;31;17;29
John
Like built into the trust. So you can't just like.

00;31;18;00 - 00;31;18;20
Aaron
Yeah, exactly.

00;31;18;23 - 00;31;22;27
John
Rockefeller is not still managing the Rockefeller Foundation finances.

00;31;23;00 - 00;31;30;12
Chris
You're telling me, Aaron, that the alderman of your Vic back in 1600 but worried about this problem? Yes.

00;31;30;14 - 00;31;31;29
John
That's coming.

00;31;32;00 - 00;31;48;22
Aaron
Right. So you're just saying there's going to be like the Elon trust and Elon's going to just be managing and and funding it via his Elon agent that he has for his personality, all of the things he cares about, you know, for an infinite amount of time into the future, or at least like hundreds of years into the future.

00;31;48;22 - 00;31;50;28
Aaron
Is that what you're saying? Yes.

00;31;50;29 - 00;31;53;22
Chris
I actually wanted to write a novel about this.

00;31;53;25 - 00;31;58;05
Aaron
Yeah, I think that's a good one, because I feel like that it will be so messed up.

00;31;58;08 - 00;32;06;25
Chris
It's like the Wrath Child. Except, yes, none of that. You only have the founder just looming over.

00;32;07;01 - 00;32;08;09
Aaron
Everybody forever.

00;32;08;11 - 00;32;18;04
John
Yeah. It is interesting to think. I mean, it's like it's not like you can, like throw the agent in jail and like, especially if the money is digital, like, how do you how do you get it?

00;32;18;04 - 00;32;23;10
Aaron
That's probably why we're going to have the AI governance. Yeah. Agents like sitting there.

00;32;23;11 - 00;32;24;28
John
Did you read the sovereign individual?

00;32;24;28 - 00;32;36;26
Aaron
Aaron I've read parts of it. I mean, I know like Bellagio and all those guys like love that. I mean, there's some there's a couple interesting threads of thoughts in that. But after a while it got like kind of it just was much for me.

00;32;37;00 - 00;32;43;12
John
It's very dense. It could have been like a 50 page book, probably not like a 500 page book or whatever.

00;32;43;12 - 00;33;02;14
Aaron
It is. Yeah. If it was like a 50 page article, like, like it had some good pieces like on, you know, just the one piece that I remember from that is just like the cost of violence, like as that goes up or as it goes down, like the the governments are centralized control kind of steps in its place.

00;33;02;17 - 00;33;20;08
John
Yeah. Well they also talk about the like cryptographic money that's like unseasonable. So I could see that like kind of relates to this like. Completely dump your, your rule set for managing the funds into your robot. How do you how do you coerce the robot into giving giving up control of the funds?

00;33;20;09 - 00;33;23;17
Aaron
You probably can't, right? Unless there's something that's like interceding it.

00;33;23;19 - 00;33;29;09
Chris
Yeah. You end up in Delaware probate court in a landmark case for human rights.

00;33;29;10 - 00;33;49;14
Aaron
I, I also think, I mean, it didn't come up in this list, but I do think that there's a lot of need for like, a gentle justice. Like if people had a dispute and could pop in there like, you know, their case on each side, whether it's prepared by lawyers or otherwise, and get like a quick answer, like a neutral, I think people would I think that would be a huge service for the world.

00;33;49;15 - 00;33;54;24
Aaron
Like people just hit these friction points and then get like nodded up in them, like there's no need for it to be that way.

00;33;54;25 - 00;33;59;20
John
Yeah. When's the first arbitration case arbitrated by agent?

00;33;59;20 - 00;34;27;21
Aaron
I mean, I built a prototype like this time last year, like literally showing you probably could do it. And it looked and this was probably like GPT four. Oh like or oh three maybe came out by that point. I don't remember. But like it was pretty good. Like it could I was feeding in like old cases that I found where I knew the answer, like I who won or who lost and, and this system didn't have that answer as part of it and would, would reasonably do a reasonably good job predicting who to win or lose the case.

00;34;27;21 - 00;34;47;19
Aaron
So I think the text kind of already there if people want it to build it, but I think that nothing's going to move slower than, than those courts. So I think that's the timing is a little bit off on it, but super cool. You guys ready for the next one. Machine capital markets. So we're not we're moving from internet capital markets to machine capital markets by 2040.

00;34;47;21 - 00;34;49;17
Aaron
This is the the A's prediction.

00;34;49;17 - 00;34;51;08
Chris
What does that even mean exactly.

00;34;51;08 - 00;34;52;08
John
What is that.

00;34;52;10 - 00;35;19;01
Aaron
So capital will not primarily move human to human. It will move agent to agent agent to protocol. Protocol to agent. There will be machine credit markets, machine bond issuance, machine insurance underwriting, autonomous liquidity pools. Credit scores will evaluate task completion, reliability, SLA adherence, capital efficiency, and risk volatility, the equivalent of a Moody's rating for agents, which is kind of what 8004 is trying to do.

00;35;19;01 - 00;35;29;25
Aaron
And that's when the economy will, quote unquote, become reflexive. I have no frigging clue what that means. But is that like the the end state of capitalism, Chris.

00;35;29;27 - 00;35;57;11
Chris
All right. So the way I could see this materializing is I need to train my own model. Right? Because we're in the year 2040 and everyone has to train their own models. Right? It costs six figures to run the training and I need to finance the whole thing like that. That's the use case, right? Is like we have a machine task so big it requires like a debt issuance rate of return sort of thing.

00;35;57;13 - 00;36;28;26
Chris
Okay. Like if you look at shit like equipment leasing today, you know, like that's a huge, huge business. And so like I do think we have, you know, sort of the, the proto versions of this out in meatspace and that, as you know, machine oriented tasks like scale up in cost and everything. Look, eventually look, if everything has to live in gigawatt data centers, right, you're not really getting like Rackspace in there like you used to in the old days.

00;36;28;26 - 00;36;30;27
Chris
So yeah, sure. I can see this.

00;36;30;29 - 00;36;44;13
Aaron
Yeah. Or if you know you're going to have X number of requests related to it, you could pre by compute. That's cheaper. Right. And then maybe maybe you're generating X amount of fees. So you just the agent just issues a bond against.

00;36;44;14 - 00;36;59;24
John
What's the limiting factor that do you need like really good batteries. That's one thing. It's like problem with like oil is sort of a good battery. And that you know how many like what's you can produce with it. But like yeah, that's one thing I always wondered.

00;36;59;25 - 00;37;24;01
Aaron
Yeah, I don't know what the limiting factor is, but that feels like it's definitely going to happen too. So I guess crypto was slightly ahead moving it to internet capital markets. But we have a machine capital markets coming next. I kind of see that one. I thought this one was cool because we talked to a couple of projects via Aidan that are building more hardware, and it feels like a lot of the builders, at least in the States, really want to build robotic factories.

00;37;24;01 - 00;37;50;01
Aaron
They're kind of like taking a page out of Elon's playbook there. And I think some of the institutional knowledge is actually in the States to begin to like, spread that out. It's a fully autonomous supply chain. So picture like a robotic factory where an agent is just like commanding something gets manufactured like on the fly, like it would be able to tap into like factory scheduling, shipping routes, you know, material procurement, energy sourcing, demand forecasting on the factory side.

00;37;50;01 - 00;38;04;26
Aaron
And then an agent would just, you know, trigger it, you know, like, hey, like, I want there may not be a Carvana here, Chris. Like, you may just assemble a car or get a car assembled from, you know, the next version of the Tesla factory. And then it just drives to you.

00;38;04;29 - 00;38;08;29
John
Just 3D printed, and your garage is printed. Do that.

00;38;09;03 - 00;38;16;15
Aaron
I don't buy that. But, like, I definitely could see I mean, if the factories literally like just like an API endpoint, like why not? Right. You pay for it.

00;38;16;16 - 00;38;18;01
John
Yeah. No, I could see it.

00;38;18;02 - 00;38;32;15
Aaron
And then like, like if that whole factory is also like super automated with agents, like, why not? Like it would go out and, you know, procure the material and tell you when you can get your car driven to you. So I don't even know if you need it.

00;38;32;17 - 00;38;35;22
John
Headless SaaS first. Headless factory second.

00;38;35;28 - 00;39;01;27
Aaron
Yeah, third. All right. This was the last one. And then I'll I'll I'll pull back on this. The by 2040, this is the prediction. Agents will engage in continuous planetary optimization. All right. This was a little disturbing. Power grid load balancing. Water distribution. Agricultural yield routing, traffic flow, carbon credit allocation. Cities will be optimized by interacting infrastructure agents.

00;39;01;27 - 00;39;04;01
Aaron
So what do we think about that?

00;39;04;02 - 00;39;05;28
Chris
Nope. Nope.

00;39;06;05 - 00;39;12;06
John
Yeah. Notes. It's pretty wild. I saw Brian Johnson speak. Maybe it's like a half.

00;39;12;06 - 00;39;14;24
Aaron
Ago, but that is his girlfriend's.

00;39;14;26 - 00;39;43;06
John
Yeah, it was before he. He was on that kick, but it was when he he had started the don't die stuff he was doing. And he was kind of talking about this like, future where like humans just like, listen to, like the AI is just telling them what to do because it knows better and isn't emotional. So this sounds kind of like that future where what you want doesn't matter so much as the, you know, the warmth of collectivity, as they call it, just, you know, whatever is best for for the group.

00;39;43;12 - 00;39;45;16
John
That will be what the cities are doing.

00;39;45;17 - 00;39;55;17
Aaron
Yeah. It'll be Pluribus all the way down, except we won't need an alien species. We'll just have these agents running around. You don't think this will happen? Chris, I think this will happen eventually. Maybe it's not by 2040.

00;39;55;18 - 00;40;16;09
Chris
Yeah. I mean, I just look at, like, the many, many different shades of, like, civic dysfunction from, like, you know, basic like zoning and NIMBYism type disputes to, you know, far more polarizing things. And I also don't forget, I was in the mobile industry when they tried to sell IoT. And so I have PTSD from all of this.

00;40;16;10 - 00;40;17;05
Chris
And I think.

00;40;17;06 - 00;40;34;07
Aaron
I've been thinking about that. I feel like people are scarred from IoT, but I feel like it's coming really close. Like maybe it was just off by a, you know, decade, but like, AI was off by a couple decades. I don't know. I feel like I think people are overly bearish in IoT stuff. I've been like kind of geeking on it a little bit.

00;40;34;08 - 00;40;54;24
Chris
I'm not like overly bearish on it. I just think it's like a decades long diffusion type of thing versus like a, a package major overhaul, you know, and I think a lot of like the instances of IoT that exists today scare the shit out of me. Like, do I want a ring doorbell? You know, like, do I want a smart home?

00;40;54;24 - 00;41;02;13
Chris
No. Because, you know, like, I'm basically just opening up my home for data surveillance, and so.

00;41;02;15 - 00;41;02;28
Aaron
Yeah.

00;41;03;00 - 00;41;12;27
John
Did you see the ring ad at the Super Bowl about, like, you know, find the neighborhood dog is like, reaction was just like, well, this is not just for dogs.

00;41;12;28 - 00;41;30;14
Aaron
Like we'll start with dogs and then we'll we'll kind of see what happens next. Yeah. So, Chris, you don't want your AI agents in 20, 40, 50 tapping in and optimizing the robot police dogs to monitor your neighborhood to make sure the curfew is enforced.

00;41;30;17 - 00;41;37;21
Chris
Yeah. No, I'm good on that. Like, I don't need a robot dog issuing a ticket because I took my trash out an hour early, you know?

00;41;37;23 - 00;41;38;19
Aaron
Yeah, completely.

00;41;38;20 - 00;41;38;27
Chris
No.

00;41;38;27 - 00;42;01;00
Aaron
Thank you. I thought this was interesting because I think it's kind of pointing towards. And I think this feels right. Like agents holding capital. And I think that's really where the crypto stuff comes in. Like agents acting economically, they're bearing risk increasingly. And the things that I think will increasingly see built like will be like identity, reputation. The Ethereum community is a little bit over that.

00;42;01;01 - 00;42;20;18
Aaron
Like capital control sounds like bankers, a little bit over that. The compliance and scoring pieces I don't think are there for like a negotiation protocol or at the end. Right, like an agent agent or like you were mentioning John. But it feels like that stuff that's kind of not, not not built yet. So maybe we see a little bit more of that.

00;42;20;19 - 00;42;26;22
Aaron
I do kind of wish pre was on so we could talk about that Brian Johnson tweet. That was literally the most insane thing on the fucking planet that I've seen.

00;42;26;23 - 00;42;35;09
John
Oh my gosh. Responses or so good. I feel like it was that. And then like the JP Morgan lawsuit.

00;42;35;16 - 00;42;35;25
Aaron
I thought.

00;42;35;25 - 00;42;37;03
John
That was fake.

00;42;37;05 - 00;42;38;22
Aaron
And that was fake news.

00;42;38;29 - 00;42;41;10
Chris
Yeah, well, I mean, it was a real lawsuit. It was just.

00;42;41;13 - 00;42;42;10
John
Yeah.

00;42;42;12 - 00;42;42;26
Chris
Not real.

00;42;43;01 - 00;42;43;29
Aaron
Like all made up.

00;42;44;00 - 00;42;45;17
John
Very good entertainment.

00;42;45;18 - 00;43;13;03
Chris
Yeah. I mean, it was a a weird day on the timeline, then the Brian Johnson thing, you just you're kind of it's a bit of a head scratcher. And that, like, I could easily see him being like, so off on his own journey and wanting to share and, you know, just like doing it genuinely. And then also, I see it as like, this guy knows how to push buttons and he hasn't been, you know, he's pulsing his virality via something that like.

00;43;13;05 - 00;43;15;01
Aaron
Literally, literally.

00;43;15;03 - 00;43;18;06
John
Yeah. Not not his first rage bait rodeo, that's for sure.

00;43;18;07 - 00;43;19;10
Aaron
Well it worked. I mean.

00;43;19;12 - 00;43;20;19
John
It did what we're talking about.

00;43;20;21 - 00;43;33;15
Aaron
If that's that, like intentional. Do you think that he's just sitting around all day? Just like being like, how can I get this attention? I got a great idea. Like, is he is that, like, the one of the 50 ideas he had that week on how to like, like, pump up his followers?

00;43;33;16 - 00;43;56;25
Chris
I think he has access to privileged information and deal flow, let's say, simply based on who he is and like his esoteric interest in body monitoring. Right? Like, to even do this, you have to know about the existence of this testing kit. And so like I think some of these things just come to him because of like what he's about.

00;43;56;26 - 00;44;15;11
Aaron
Yeah. That's true. Yeah. I'm always like kind of the mind of the influencer of like I just not my mind. So I never understand like fully their process. I feel like somebody should explore that because I know that they it's probably a lot more intentional, like you were suggesting, than it kind of seems on the surface.

00;44;15;12 - 00;44;42;16
Chris
Right. Well, like, you know, if you ever read like a profile of a comedian, like they always have, like, or at least in the old days, like index cards, like just, you know, boxes and boxes or like card, you know, like they had like a million jokes that they keep on hand and they return to. And so, like, I assume there's a yeah, if you're an influencer or a clipper or, you know, people have gift folders, there's got to be some form of like organizational psychosis around this stuff.

00;44;42;20 - 00;44;59;13
Aaron
Yeah, it feels that way. Yeah. Maybe it's just that like, it's just constantly like working through these concepts in his eyes. I mean, it does feel like just talking about the timeline this week, like peptides across another Rubicon. I just feel like my entire feat is now like peptide or peptide adjacent. It's like, I don't know what happened.

00;44;59;14 - 00;45;02;01
Aaron
Is that, is that your feeds or is that just me?

00;45;02;02 - 00;45;26;09
John
It's like quad maxing. Peptide maxing. It's the all the all maxing everything. Yeah. It's amazing. I feel like it's you know, was it like out of the 50s or like you're kind of like the magic pill of like the antibiotic is the 40s, but like it's sort of just like, oh, there's now a new magic thing you can inject that will like make you taller and stronger and like it's going to solve all your problems.

00;45;26;11 - 00;45;34;22
Aaron
It's like a magic thing that's like completely unregulated, though. Yeah. There's no quality control. I think for some of them they don't even know like the long term effects.

00;45;34;24 - 00;45;41;07
John
And what are they? Proteins. They're just like new kind of like synthetic proteins we haven't seen before.

00;45;41;10 - 00;46;00;16
Aaron
I don't think they're proteins. I think they like they signal something in your body to get a response. So like one and I forgot the name of it. Like it's like a pituitary peptide. And that's what helps you. It gets secreted when you're in like in the sun and you get tan. And so they've somehow isolated that and then can elicit that like tanning response without.

00;46;00;21 - 00;46;02;14
John
You can get tan in your basement.

00;46;02;18 - 00;46;06;22
Aaron
You can Chris look at you. You can look exactly. You can look like an orange man.

00;46;06;23 - 00;46;11;11
Chris
Oh my God. Wow. Like, oh look, you came upstairs and you have such a healthy glow.

00;46;11;12 - 00;46;12;21
Aaron
Exactly.

00;46;12;23 - 00;46;37;28
Chris
But I mean, I think, Aaron, to go back to my biology roots, you are describing what, like a lot of proteins do, like Java protein is to latch on to a cell and then like trigger a, you know, trigger a response. And so peptides could be proteins. I'm so far away from all this. You know, I'm of the camp of like in ten years from now when everyone figures this out and I'm starting to, you know, need need a little help, I'll just do what's told.

00;46;37;28 - 00;46;43;11
Chris
I'm not I'm not getting on Chinese grain markets and cheating myself up right now.

00;46;43;13 - 00;46;47;16
Aaron
Yeah, that's just something about that just is not going to check all the boxes for me.

00;46;47;16 - 00;46;58;23
John
But I'm also the team. Just wait and see. Like it's it's kind of fascinating how excited everyone is to try like the, you know, the new hot one. Like, why is that? Why would you want to do that?

00;46;58;24 - 00;47;16;11
Aaron
I think cause the results are good when it works. But like, who knows if it doesn't work for you or like what the side effects are. So I feel like in a couple years there'll be some, like, disaster stories related to all this stuff looping back to the AI Psychosis Summit. Did you guys just tweet that it was all just New York crypto people at the summit?

00;47;16;18 - 00;47;17;22
Chris
That makes sense.

00;47;17;24 - 00;47;29;04
John
It's not surprising. I the a lot of the AI meetups I've been to in the city, it was just like a lot of the same crowd from kind of like the early Ethereum days meetups. Just pretty fascinating.

00;47;29;05 - 00;47;41;11
Aaron
Yeah, completely. But it just it feels like the tweet here was saying that it feels like crypto's in in an identity crisis it's going through its like teenage identity crisis period trying on some new things.

00;47;41;11 - 00;47;49;16
John
Everyone thought was like, you know, changing the world. But turns out it was just putting dollars on the internet. It was like the superior use case.

00;47;49;19 - 00;47;54;26
Aaron
Yeah, completely. Do we think, talking about crypto, do we think this NFT run is going to continue?

00;47;54;27 - 00;48;04;23
John
I don't know. Let's see. I mean, I hope so, but I have no idea. Like what triggered it? It seemed like there was kind of just no, at least to my knowledge, no sort of like event or anything. Just like.

00;48;04;24 - 00;48;07;06
Aaron
I think it was the DeFi hacks.

00;48;07;08 - 00;48;08;19
John
Oh yeah. That's just like.

00;48;08;25 - 00;48;36;13
Aaron
I think it was just that I think that there's people that want to keep, you know, their crypto capital, like in market. And they got nervous about DeFi. And I think that some of those recent hacks just kind of were like the straw that broke the camel's back. And then they're just allocating in. Plus, like maybe a thought that we're at the end of this bear market or at the beginning of a bull market, and NFTs have historically done okay during those periods, right.

00;48;36;17 - 00;48;43;01
Aaron
Like if you're long eth like maybe, maybe having like an allocation to NFTs make some sort of sense.

00;48;43;02 - 00;48;45;08
John
Yeah. Like punks is a leveraged ETH bet.

00;48;45;09 - 00;49;06;25
Chris
I'm, I'm more of like the the view that like, this is a boat that's caught in the doldrums. And every once in a while, a breeze comes along and people get really excited because the boat moves a bit. You know, some of this sure could be there's no other place to put your money. I think some of it was like UGA news event triggered that, you know, maybe cause spark.

00;49;06;26 - 00;49;33;16
Chris
The challenge I think is really like you're heading into the summer and summer is always tougher in fees. And so, you know, maybe you get like a nice little May and then we're stuck again. But then, you know, I just go back to like, we, we can't just keep cycling through the classics and pretend it to healthy market and that you're going to need a new wave of creators or a new movement or just something to, you know, catch people, people's attention again and pull them back in.

00;49;33;16 - 00;49;50;02
Chris
And so, look, I'm very happy, you know, for whoever is still around and, you know, enjoying this and appreciating the classics and engaging in new work, but like, I think it's a it's a pretty small gymnasium of folks right now and that, you know, there's not a critical mass to sustain this.

00;49;50;03 - 00;49;55;18
John
Yeah. It doesn't feel like there's a new wave of excitement and buyers. It's more like folks who've been around.

00;49;55;19 - 00;50;20;28
Aaron
Yeah, like moving their moving capital around a little bit. Yeah, yeah, it feels that way to me. Or maybe, you know, taking capital off the sidelines. I do wonder if that is like the big question, Chris, like where does the new creators come from? Like where are they going to go? I mean, I do wonder if like, maybe there are some more like AI based creators to like, maybe that's a one way to like kind of capture some of that activity.

00;50;20;29 - 00;50;29;21
Aaron
Like the slop is getting good. I think I read this morning that like something like 30 to 40% of all new podcasts are completely AI generated at this point, including.

00;50;29;26 - 00;50;31;15
John
Pretty crazy. Exactly.

00;50;31;17 - 00;50;32;24
Aaron
We're not even here, guys.

00;50;32;25 - 00;50;37;10
John
Yeah, we're from 2040. We came back to shoot this podcast as agents.

00;50;37;16 - 00;50;38;08
Aaron
Yeah, completely.

00;50;38;09 - 00;51;02;00
Chris
Right now I could see, you know, if you're looking for some, some glimmers around, like what? In the next wave of, like, digital culture, I think you can go back to, you know, this like, creator like agenda creator phrase that we went through over the last two years. I think that was a little premature, but like, not necessarily wrong.

00;51;02;01 - 00;51;38;06
Chris
I think the I mean, that that whole thing was so hard to evaluate because, you know, you, you had like both a the aftermath of the meme coin frenzy, looking for a new market to, to trade in. And then you also had like the overblown hype of early agent stuff. But I do think, you know, you're also seeing a little more seriously like people look look at the ideas of like dark forest communities and like, organizing models and coordination models and close networks of shared interests.

00;51;38;06 - 00;52;00;17
Chris
And so I think there's enough materials floating around that you could see more like systems and networked based digital culture, you know, am I am I saying like FB for bots like that doesn't feel right. But like I also feel like you could point back to that maybe once this whatever this thing is materializes, I think there's enough touch points.

00;52;00;17 - 00;52;20;11
Chris
It's just it was all a little too early. It didn't it couldn't close the loop of like, you know, providing engaging experiences in that. I do think, you know, I would bet more on that than I would on like single object based art being the thing.

00;52;20;13 - 00;52;24;26
Aaron
Yeah, I could see that. Chris. I think that's like a pretty fair read on what that could look like.

00;52;24;27 - 00;52;29;15
Chris
I think it'd be nice to, like, have something other than hacks to, to look at.

00;52;29;17 - 00;52;48;01
Aaron
Yeah, I think we're like in the hacking era of, of of crypto right now, but it feels like it's even outside of that. Like I saw like that super old project cPanel got like hit, which I think a lot of folks historically have used to like maintain their website in some capacity. So it just feels like it's everywhere at this point.

00;52;48;02 - 00;52;52;12
Aaron
I wonder how long that lasts until we get this, like new equilibrium. It feels.

00;52;52;12 - 00;52;56;17
John
Like, is it just the agents getting better and helping people hack more efficiently?

00;52;56;17 - 00;53;16;22
Aaron
I think so, yeah. I mean, and I saw some like the like Mark injuries and others, David Sacks were like pushing to get like mythos open. They're just like look like let's stop pretending you can't do this with ChatGPT 5.5, which you can like. We just need to like suffer some lumps to make all this stuff stronger, you know, which is I don't know.

00;53;16;25 - 00;53;22;08
Aaron
I don't know exactly. Yeah, I don't know exactly how I feel about that. I feel like it's a pretty tricky question personally.

00;53;22;13 - 00;53;59;06
Chris
Right. Because it's a, like, ever expanding circles of access, right? In which someone's always disadvantaged. And then if you're a threat actor, you're just always probing for what's the what's the lowest surface area. You know, at a certain point, if these things dry up and they're certainly not, I think the, you know, DP kr can, you know, keep going after trophy heists here, but I assume, you know, the rest of like, the hacking world is just going to keep moving downstream, moving downstream, moving downstream.

00;53;59;06 - 00;54;37;14
Chris
And so, you know, ultimately it's it's yeah, you're right Aaron. It's a very tricky thing. We certainly don't need like zero day internet crippling attacks especially, you know, at moments in time where like there's a lot of downsizing around core infrastructure, right. Like, you know, it's no coincidence that like AWS, GitHub, you know, like suffer outages because they're transitioning into AI systems and they don't have that entirely locked down at the same time, like they're letting go of like core tribal knowledge in headcount reductions.

00;54;37;14 - 00;55;03;16
Chris
And so, you know, do you want to unleash those on the world at a time where, like, big core infra is making it surface area, maybe more tackle that? That seems really dangerous, but at the same time, right. Like you don't really want to set a precedent in which like, people are disadvantaged and like, you know, there's there's true some asymmetry in terms of what people can do.

00;55;03;18 - 00;55;07;06
Chris
I mean, it's also a horrible time to do that, you know?

00;55;07;09 - 00;55;25;02
Aaron
Yeah, that's true. But I guess we're going to have to live through it like to a certain degree. But it definitely feels I mean, like that's just going back to the NFT thing. I do wonder if that is like a subtle, like, unstated driver of a lot of this stuff. It definitely feels like it is, at least to me.

00;55;25;03 - 00;55;30;14
Chris
I just want simple non liquid 720 ones that.

00;55;30;19 - 00;55;32;09
John
People don't want monkey jpegs.

00;55;32;10 - 00;55;51;12
Aaron
I think it's that. I think it's also like, you know, I do think I would love for there to be crypto to come up with new, new things. I think some of that stuff takes more end users, which hopefully will come. Some of that takes like maturity and like also people wanting to play around with digital assets again.

00;55;51;12 - 00;56;07;22
Aaron
But if there is like a new cycle, there's always something that's been floating around that like just gets a new spotlight in some capacity. So I don't know exactly what that is this time around. Could be ICOs, it could be NFTs, it could be game fi, it could be, who knows? Right. It just doesn't feel like it's going to be DeFi.

00;56;07;23 - 00;56;11;12
Aaron
It doesn't feel like it's going to be meme coins. So it's got to be something else.

00;56;11;14 - 00;56;15;05
Chris
Well, that's why we get out of bed. We will see what tomorrow brings.

00;56;15;08 - 00;56;16;20
Aaron
That is why we get out of bed.

00;56;16;21 - 00;56;17;19
John
Exactly.

00;56;17;25 - 00;56;25;06
Aaron
Well, maybe. Maybe we end on that one. Chris, I think I think that was a good a good bridge. So welcome to net society, guys.

00;56;25;07 - 00;56;28;13
John
That's deciding. It's why we get out of bed.

00;56;28;16 - 00;56;30;03
Chris
At least on Fridays.

00;56;30;06 - 00;56;51;17
Aaron
Yeah. Every other day, not so much. But welcome to Net Society guys. We had fortunate enough to have John from Super Air on me and Chris talking all things internet and crypto and everything in between. Just a quick reminder. Nothing we say here is financial advice and nothing should be attributed to our employers. So let's go.