Subscribe
Share
Share
Embed
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;03 - 00;00;30;06
Aaron
Hey, guys. I went to this pretty, fun event last night. It was, this resonant, competing event. And was that better works? Which I thought it was great that they, hosted. Did you guys were to bump into their manifesto that was circulating around?
00;00;30;08 - 00;00;31;26
Derek
No, but I'd love to hear about it.
00;00;31;29 - 00;00;57;10
Aaron
Yeah, it was kind of interesting. You know, the the basic ideas in tech. Lots of folks are upset about it for a variety of reasons. And it's really like design decisions. Right. And we can think about rebuilding tech, particularly for the I.R.A. that's pro-social, you know, so positive for society, adaptable, plural, dedicated and private. And so the whole event was mostly about like how I should be built that way.
00;00;57;12 - 00;01;27;13
Aaron
And I kept on just thinking the whole time about how this really applies to crypto, because I feel like at the core, like when Bitcoin and Eath and related project started, it was really thinking about that, right? Like, how do we build a parallel system that's that's better and improves everybody's life, that's adaptable, that's really, you know, has like a human at the center that's social, that gets a bunch of people together to do things that are productive, that, you know, balances the tricky questions related to privacy and is really kind of dedicated to to these types of principles.
00;01;27;15 - 00;01;45;26
Aaron
And I feel like I feel like over the past couple of years, like that core purpose has kind of been lost, right? Like we took the easy path towards speculation and these, like, elements are kind of missing. And I think it I think it's a good time to kind of reground the ecosystem into it. So yeah, I've been thinking a lot about that.
00;01;45;28 - 00;02;07;18
Aaron
And I think it's incredibly important to recenter things around that, especially now that institutions are coming in. And they're not necessarily lined to that. Right. Like they're not thinking about how this technology can enable connections and coordination or be adaptable and be open ended. You know, in plural, they're no single, and they should control these spaces, right?
00;02;07;20 - 00;02;27;12
Aaron
That the software should work for us. And really, like, I think they do you think a little bit about privacy but you know, more more so just to kind of align it with, privacy norms that regulations require for them. But, you know, I think as institutions come in, it's great, but it's a double edged sword where on the one hand, you know, more capital is coming in that brings more attention.
00;02;27;14 - 00;02;50;24
Aaron
The space matures. But at the same time, you know, for many folks, like, all they're going to see is like a line item on their bank statement. Or if they log in to like a site, they're not going to feel like kind of the crypto parts of the technology. So I think it's a real moment to kind of like regroup, especially when, when Attention's kind of focused on these superhuman, synthetic beings that all the hyperscalers are building.
00;02;51;01 - 00;03;07;29
Derek
Let's, let's say we do regroup. If you could have your wish granted here, Aaron. What what does crypto look like over the next year or two? If we were to kind of rebuild into a better format or a better. Yeah. A, a better collective, what does that look like?
00;03;08;01 - 00;03;24;09
Aaron
I think that's a really great question. The way I've been thinking about it is one, we kind of know what did work with stablecoins, right? Like the one thing that's the stickiest in crypto right now is the one area where speculation kind of wasn't present at all, which I think is pretty fascinating, that that that's the case.
00;03;24;16 - 00;03;44;03
Aaron
And so I think it's really like rallying around one or a handful of protocols like these. L2 wars were great, but they they benefited the teams building it. Some like early adopters there and folks that financed it, which was reasonable bets. But it also like had a knock on effect of just fracturing the community, making it more tribal.
00;03;44;05 - 00;04;19;20
Aaron
You know, the internet one, not because people are obsessed with, you know, the Http protocol or the related email and other protocols in one, because there was kind of like a shared protocol that everybody could build on top. So I do think like a consolidation and, you know, one or 2 or 2, you know, maybe a handful more, core networks is helpful and kind of moving beyond this era of like, fracturing to I, I don't and this is a common criticism, but crypto really should be for everybody and all the, you know, complexity that sits around it, whether it's like token gating, whether it's, you know, hard to use wallets and bridging
00;04;19;20 - 00;04;38;02
Aaron
like that really can't be the North Star. It should really be to stream things out and streamline it so that anybody could use it doesn't mean you don't need to have like, you know, distributed computing background or cybersecurity background for it to work. I know lots of folks are kind of working in that vein, but I think they should be doubling down or tripling down on that.
00;04;38;04 - 00;05;03;07
Aaron
And I think it's really just remembering what blockchains actually are at the core, which is a trust machine. Right? Which I know that analogy doesn't fully hold, but their systems that make commitments make them legible, enforceable, composable without permission. So I'm hoping we see things that are more anchored in programable ownership, stuff like royalties that execute automatically, Rev shares that don't require label or platform communities that can collectively fund each other.
00;05;03;10 - 00;05;26;13
Aaron
You know, none of these ideas, I think, are are old, but I think that they're important and should be kind of the the center. And same thing for capital markets. I do think it's, you know, really thinking of them as the pipes, and transparent settlement layers and having those layers governed the same way. We've seen, you know, open source projects, governance not because it's faster, flashier, but because it's programable.
00;05;26;13 - 00;05;43;02
Aaron
And that's better for everybody, right? It lowers the cost of participation. And hopefully that like, bleeds into online organizations, something that I've thought a lot about. But I do think we need to like, empower the internet to be productive so we can get out of this era of like content slot machines, which is just like a depressing, shitty endpoint for the internet.
00;05;43;02 - 00;05;58;12
Aaron
And the only way we're going to do that is by showing everybody around the planet that you can do fun things on the internet with other people, right? And we need to make it more social, less isolated, you know, less just like yelling at the box. So I don't know if any of that resonated, Eric, but it's kind of what that is.
00;05;58;19 - 00;06;36;10
Derek
No, I totally resonates. I think it's a conundrum for me because the the whole origin of our space call it, you know, even just the origin of Bitcoin is predicated on this idea of groups of people coming together with a shared belief to to work and speculate on, you know, and in the case of Bitcoin, on the rising price of Bitcoin and, and that rising price is what pays for the shared security of, trust minimized ledger, which removes trust entirely from the system and allows you to send value peer to peer.
00;06;36;12 - 00;07;17;20
Derek
And so it's like at the very like at the DNA level of like these protocols, it is really about removing trust. And it's also there is a flavor of speculation to support, kind of like, you know, a headless collective coming together to believe in something bigger than themselves. It's like there's this there is always this inherent speculation that exists that the work you do will prove valuable and create value for this network, which then can be received via this, this, you know, token output in the in the case of like an appreciating BTC or appreciating eath and even stablecoins, which is like no question your to to call that one out as like the success
00;07;17;20 - 00;07;45;08
Derek
story of the last few years which doesn't allow speculation. But you know, it's it's also true that for stablecoins to actually be viable, they need to operate on top of a trust minimized ledger, even if the stablecoin company is centralized, itself. And there isn't really a way to speculate on that system. And so like in part, you know, stablecoins only really work because people are speculating on ether and supporting that network and, you know, staking their ether tokens.
00;07;45;08 - 00;08;26;20
Derek
And the same with Solana and other chains. And so I think, you know, it's it's it's it's so hard for me to say that we should and I don't think you're saying this at all. It's just like the I keep going, I keep swinging back and forth on this idea, you know, I think with meme tokens and in some cases, NFTs and, you know, in some cases perps and roadways and some of these other things, it's I think people have taken that inherent loop around speculation, around these other around things becoming valuable in our space and maybe taking it a bit too far or a little bit, you know, it taking it
00;08;26;20 - 00;09;00;03
Derek
so far away from, like the original thesis of, like, coordination, speculation to support coordination and instead of what we've seen recently, which is, you know, speculation for speculation sake. And so I definitely think that there are lessons to be learned there. But part of me always believes that for this economy to be trustless and permissionless and to minimize human coordination, that framing is only going to be complete if we have some level of of speculation at the heart of this thing.
00;09;00;09 - 00;09;05;28
Derek
And so I don't know I don't know what that means exactly, but it is like, like the catch 22 here.
00;09;06;04 - 00;09;36;18
Aaron
Yeah, and I agree. I mean, I agree with that completely. I think that having market based systems, that's how I think about it, Derek, are completely fine. I think it's really just that that predominated and pulled it away from something that I think has always been the most important, which is just utility. Right? Like people are building things just for the purpose of speculation as opposed to providing like real utility that like enrich people's lives, that, you know, help those coordination games that are important and meaningful.
00;09;36;20 - 00;09;55;13
Aaron
And I think it kind of like was the easy path. And now I think crypto is at a point where it will regroup. I think that's just kind of how these tech movements and ecosystems tend to to work. And as it regroups, it should really have different North stars, which are, focused, I think, on its core original knitting.
00;09;55;16 - 00;10;29;00
Aaron
You know, when the Cypherpunks were building Bitcoin or even like Satoshi, you know, he didn't mention price in the whitepaper. He wasn't thinking about price. Right? Instead, he mentioned trust, I think 14 times, or something like that. So it's kind of like re refocusing on those pieces, just so it doesn't just become another line on, you know, your, your bank statement, because I think that's the other end point kind of related to this, which would be, you know, pretty sad, in my mind, if, if that's kind of the end point, and I do think it and hopefully, like the venture community begins to stop making bets here just
00;10;29;00 - 00;10;58;17
Aaron
because they may not be as profitable as they thought they would be. But I do think, like the L2 wars, well, maybe they were great for a moment with Twitter's then algorithm to like, capture attention. I just think it fractured everything and like thinking about like token gating and all these pieces which which is interesting. But I don't think it led to the, you know, the Nirvana promised land that people were were hoping they were great experiments, but they, they also had the unintended side side effect of just excluding people.
00;10;58;19 - 00;11;16;24
Aaron
And crypto really should be for everybody, right? Like everybody has some sort of financial or cultural need. If it's if it's something like, like an NFT and it needs to be like more open, I think, as opposed to close, like even look at the NFT ecosystem. It's become more insular, right? It's not about, you know, delivering fun media for everybody.
00;11;16;26 - 00;11;32;22
Aaron
It's about like measuring out in a slightly more efficient way, like some of the legacy, you know, art markets. Right? Like that's kind of it feels like the will of that community, which I don't know if that's the right pointer either. All right. I'm done with my done with my rant. Thanks for giving me the space.
00;11;32;22 - 00;11;33;13
Chris
To do it.
00;11;33;15 - 00;11;37;17
Derek
Chris, do you have any, you have any takes? Yeah, I want to add a little flavor.
00;11;37;20 - 00;11;56;28
Chris
Well, I mean, congratulations to Aaron for channeling 2021. It's always good to see someone, believe in life after love. Keep a candle in the window and, you know, carry the torch for what we'd like to see in the world. You know, hats off to Aaron. And and congrats on getting out of the house, Aaron. That was quite the humblebrag.
00;11;57;01 - 00;11;58;11
Aaron
Thank you. Thank you.
00;11;58;14 - 00;12;23;21
Chris
I, I don't have a lot of thoughts around this. You know, these these are all great aspirations. And there's certainly things that I'm aligned with. And I think your points are all excellent Aaron. You know, it's it's one of these things where you just got, I don't know, hope and believe and, and, you know, try to be the positive force you want to see in the world, but then it's you're at the at the whims of fate.
00;12;23;23 - 00;13;04;28
Chris
I do think, you know, we're certainly swung, you know, the pendulum has swung to a very sort of, like, isolationist, everyone in it for themselves, sort of moment in time. And so, you know, the batteries, I hope the batteries recharge. And if the space does get interesting again, because, you know, I think to me, the part I miss most is the small scale experimentation, the, you know, feeling that, you know, when things are really humming, you woke up in the morning and you didn't know what you were going to get because, you know, on any given day, project could drop out of left field with ideas you'd never seen before.
00;13;04;28 - 00;13;30;10
Chris
And that that was really exciting and, you know, we do really seem to have, like, at least in the world of crypto, you know, have reached a point where no one is experimenting that people have decided, you know, what the door for that is closed. We now know what works. Let's, you know, let's really focus in on on trying to scale that out and so, you know, it's understandable.
00;13;30;13 - 00;13;54;05
Chris
Certainly. Why, some people have taken that posture. You know, a lot of people put years of their lives into this, and, you know, they what they were trying to achieve, you know, didn't pan out in a certain direction. And, you know, they get to work and they get to do things with their life. But then also, you know, on an optimistic note, leaves ground open for others to come in and, you know, try, try new stuff.
00;13;54;05 - 00;14;06;24
Chris
So I do miss the, you know, the more experimental and wacky days, you know, especially in the digital culture and a key space. And so I'd like to see a return to that.
00;14;07;01 - 00;14;25;25
Aaron
Yeah. And, well, I think it's kind of like you got to project what you want the world to be. I think crypto has done a great job bending spoons. Its task was not an easy one to go after some of the big dogs. And now that they're here, I just think it's really important to kind of keep that, you know, these like guiding principles in people's minds.
00;14;25;25 - 00;14;43;19
Aaron
So hopefully we do that. And remember, like crypto's not supposed to be exclusionary, right. It really at the, at its endpoint, it's going to be for everybody. I do think like that space is opening up. I was I was talking to somebody else this week about it. And it really does remind me of the like period before is maybe some people may argue it's worse.
00;14;43;20 - 00;15;00;03
Aaron
It kind of feels about the same. And I wonder if there's like a new project that like kicks that off. I know, Derek, you've been a big, big tensor fan. Like, I wonder if that is kind of the project, to kick off, like the next wave of crypto, even more so than some of the payment stuff.
00;15;00;05 - 00;15;02;02
Aaron
Have you been following it? I imagine you happen.
00;15;02;05 - 00;15;23;22
Derek
Yeah. So I really I like the tensor. It's have for many, many years at this point it hasn't quite had its moment yet. And I could point to a number of different reasons. I think architecturally it's very complicated. It's not I mean, I wouldn't call it like a easy ramp into kind of digesting what the heck is going on with that project.
00;15;23;24 - 00;15;45;08
Derek
I think, you know, from a targeting perspective, it hits all the right notes. It was a fair launch project. It's proof of work. It's proof of useful work, which is important for things that have more subjective outcomes. It's, you know, it's got a core group of believers that continue to grow and point the project in lots of different directions.
00;15;45;08 - 00;16;14;13
Derek
There's a way to express beliefs on where I and I commodities may go by buying a subnet and creating a subnet game, and allowing people to participate it to create commodities and so you have, you know, this very expressive layer on top of kind of the core foundation of that project that yeah, it looks like, you know, early Ethereum where people are just trying stuff and trying to create useful protocols and generate some of them are starting to generate meaningful revenue.
00;16;14;15 - 00;16;43;07
Derek
And so it's it's it's been a, a, I guess, an important project to keep track of just given it's kind of a bellwether for a lot of the themes in crypto that I think are important. You know, is, is now the moment for bit tenser to break out? I'm not I'm not sure. We've had a bunch of like, these little mini cycles in the past, over the last 3 or 4 years where a bunch of people will get nerd slapped on it.
00;16;43;07 - 00;17;06;04
Derek
More people will join the ecosystem, more subnets will be interacted with and participated in. And then, you know, there will be a new shiny object in crypto and people will kind of like lose focus or scope. But I can't say like there are more people today interacting with tensor than there were a year ago. And a year ago there was more people interacting with the tensor than there were the year before that.
00;17;06;06 - 00;17;31;14
Derek
And so I think there's signal in the fact that it continues to evolve and is plastic and grow, and more people come into that ecosystem to try things and yeah, I think eventually there will be a couple of these subnets that break out and create really durable value. And, you know, are reaching, you know, valuations and, you know, revenue multiples that rival AI labs.
00;17;31;14 - 00;17;42;25
Derek
And at that point, you know, you only need a couple of these things to really do serious numbers for people to kind of like pay really, really strong attention to it. So I think that's that's the bet you're making with the tensor.
00;17;42;28 - 00;18;01;02
Aaron
Yeah. I don't know if if it's moment is now, but it feels like a moment for something like that, like something really like a little bit different. Will will be like kind of a rallying call. I, you know, I think that like X 402 agent ecommerce and payment systems are super interesting, super important. But I don't think I will.
00;18;01;02 - 00;18;02;13
Derek
Like that category a lot too.
00;18;02;14 - 00;18;23;21
Aaron
Yeah. Me too. And it's super interesting. There's like hundreds of, projects kind of building or like supporting that already. I like kind of the posture of it against, like a more centralized, more centrally maintained, version of what it's MPC, the stripe version. It feels like you kind of need something like, like, kind of like adjacent to everything people are building to be that breakout.
00;18;23;22 - 00;18;51;29
Derek
And it's interesting that you say that there's a number of. So, you know, the I guess the core deliverable for these subnets are to kind of create some sort of useful commodity in AI that can be a pre-trained model, that can be, compute, it can be inference, an inference network, it can be storage, anything that is kind of like, well, it doesn't even have to be virtually valuable, but like virtually or digitally valuable as a resource or a commodity.
00;18;52;01 - 00;19;19;02
Derek
And, you know, with the zero to an MPP stuff, there's a new class of buyer that may want to buy useful, valuable digital or virtual, you know, commodities and that's agents. And if you think about like that flow of, okay, I have a subnet, I'm creating this subnet game. People are participating in submitting resources. Those resources get, you know, graded and validated.
00;19;19;02 - 00;19;43;28
Derek
And there's a winner in some cases, or there's a group of winners in some cases. And now that commodity can be resold. One thing that's really interesting about the zero two and AMP stuff is that can now be put into an API and gated by a micro transaction that can be surfaced to via, to an agent via an API call in exchange for, you know, a fraction of a penny.
00;19;44;00 - 00;20;04;29
Derek
And so, yeah, I totally agree with what you're seeing here and which is like it's something to definitely keep an eye on. It's like this new class of buyers who are poised to spend, you know, an immense amount of time and, and money, replacing humans on the internet over the next few years, being kind of the perfect candidate to buy some of these commodities if, if they find them valuable.
00;20;04;29 - 00;20;26;15
Derek
And so, yeah, there is like a loop closure there. That's really interesting as relates, to bit tensor commodities. But the next 1 or 2 stuff that I'm keeping an eye on, it's obviously still really early and, you know, I was chatting with, Sam from Merit Systems yesterday who's, who's building a project called Agent Cache, which is essentially like an eBay for people to put up.
00;20;26;17 - 00;20;46;28
Derek
Endpoints, for, like, valuable APIs that agents can purchase for like, a dollar. And, you know, the volume that's flowing through these systems so far, it's like it's not very much it's like a couple hundred thousand dollars maybe, but you could paint a picture where like, it's, you know, over the, over the next year, it goes from a couple hundred thousand to a couple million.
00;20;46;28 - 00;21;05;00
Derek
And then 10 to 20 million and then 200 million. And then all of a sudden, you've got more people using this form factor to purchase value, to create value, to to buy value and to create new types of value. And all of a sudden you've got a full on economy of agents that are buying and trading and selling stuff with each other.
00;21;05;03 - 00;21;18;12
Derek
And the stuff that I look for is like the past before the past. It's like, okay, if that's true, and there are like these valuable commodities that are hitting the market, like, is there a natural tie in here? And I think there could be, but it's too it's too early to tell.
00;21;18;16 - 00;21;40;03
Aaron
Yeah, completely. And that is like the next big, you know, quote unquote consumer base. Right, Derek. It's just agents. So like, thinking about what they need, like, where they're going to want to shop for like a better way to describe it or get the resources that they need. I do think it's like a super interesting framing the complexity around, like my only reservation.
00;21;40;03 - 00;21;58;06
Aaron
I just, I wonder, like, I wonder what that looks like. Like, is it too, like Rube Goldberg, like in some sort of way? Or maybe. Maybe there maybe. And I'm kind of curious if you think there's an opening for it. Maybe there's kind of like a, like a slimmed down version of that, that, a bit tenser that works and kind of catches a wave.
00;21;58;08 - 00;22;15;10
Aaron
Maybe that's next to the next step, right? Like maybe that was directionally right, but like a little too complicated. And it can get like boil down a little bit, you know, kind of it kind of did that. Right. Like there was others, there was other like, you know, blockchain smart contracts systems, but they just were like a little too complicated.
00;22;15;12 - 00;22;20;01
Aaron
Not that eath was not, but it just like simplified it down enough that people could begin to dig in.
00;22;20;03 - 00;22;58;05
Chris
Yeah. I mean, some of the burden of tensor is because it was so early, it had to do everything itself. And once you start doing that right, the the complexity scales dramatically. You know, you have validators, you've got, subnets, you've got all these, wonky pieces going on in there that I, I feel like, you know what you're pointing towards, Aaron, is more like these agent systems do so much of the, the work in-house, and it's really about, like, stream consumption or the output and the utility of what's coming out of these systems.
00;22;58;05 - 00;23;24;02
Chris
And then, you know how one system consumes, the work product, or the value of another system. And so to me, this is just this is stuff that we're just starting to catch. Glimmers of. Right? Like we're just reaching that point in time where you can have multiple agent systems, you know, kind of, kicking the tires and chatting with each other.
00;23;24;02 - 00;23;50;00
Chris
And that's, a far closer parallel to, you know, how the how the world works today itself, right? Like, we don't have to, I don't know, you buy a piece of clothing, right? Like you're buying it on a certain set of criteria, none of which involve the fact that, like, that manufacturing process sat within the same ecosystem as you as a consumer.
00;23;50;00 - 00;24;10;27
Chris
And, you know, you all have the same validation layers. You all have the same quality scores, right? Like that to me, I think is part of the challenge. Maybe around bit tenser is that it had to be its own house because there was none. But I don't think that's going to be the more popular model. You know, the more popular model is really like, what can you do for me?
00;24;10;29 - 00;24;19;04
Chris
You know, and I kind of don't care what, you know, how you go about doing it. I just care that, you know, like, you can solve a problem.
00;24;19;07 - 00;24;20;20
Derek
Yeah. I think that's very fair.
00;24;20;20 - 00;24;39;21
Aaron
I just kind of side tangent. I was talking to a technical team there, kind of thinking about like, how do we, you know, how do we keep like these agen tech world like secure. And they were telling me they set up a system. They believe it's secure there. They seem pretty good at what they do. So it's kind of like a two tier system or one they put in like an enchanting system, like an open floor.
00;24;39;23 - 00;24;59;02
Aaron
Second one has like all the controls to kind of make sure it doesn't break out of it. And they've been running it and, you know, even with like open cloud, some models like it's constantly trying to like break out of its box. Like it's just like always hitting a wall, always like wanting to, like, get out of whatever security, you know, constraints.
00;24;59;02 - 00;25;21;27
Aaron
They kind of put it in there. So I don't know. I just feel like thinking about these like a generic systems as like kind of, organic software. It's just kind of like a nice mental model, right? Where they're like, almost like kids, like trying to push the boundaries, trying to get out of, like, the parents rules, you know, really pushing the envelope, which always makes me go back to like a blockchain being a good spot for them.
00;25;21;27 - 00;25;27;03
Aaron
Right? Because it's it's really hard to break out of those rules of like a blockchain based system.
00;25;27;05 - 00;25;41;04
Derek
Yeah, totally. Totally agree. I think it's I think it is really interesting mental model for sure. I, I guess I'm curious here, Aaron, I know you spend a lot of time at this stuff. Do you I guess, like what do you what's your projection for the next year? Like what? What are you thinking?
00;25;41;04 - 00;25;56;27
Aaron
It's so funny. Yeah, I was, I was, and, you know, we host this, like, Monday night call for, other folks in the DAOs, and I was asking the same thing. I'm having a hard time seeing where the end of the year is. I mean, this has been a wild year for a number of reasons. I guess we got is mythos.
00;25;56;27 - 00;26;21;25
Aaron
What's the new anthropic model that got announced today? Which which is looks like, at least according to anthropic, it's a quote unquote step change in AI capabilities, including dramatically higher scores encoding academic research, reasoning and cybersecurity. Currently far ahead of any AI model in cyber capabilities. And they're worried about it from like a security posture. Right. It's purportedly like super expensive to run.
00;26;21;25 - 00;26;50;17
Aaron
So it's not yet ready for general release. But the fact that that's like kicking around, like internally, if it does have like significantly higher evals, is is going to be wild, right? I mean, just seeing it, seeing these systems get better and better on the software side has been nuts. But you know, if the reasoning is a step change higher, like buckle up, you know, so that's like one thing like I have a really hard time kind of picturing where the the models will be by the end of the year, except knowing that direction, they'll be better.
00;26;50;19 - 00;27;18;25
Aaron
I do think like the next big thing that people will begin to talk about is like a giant swarms. So not just like having like deterministic tools that you're calling and like, you know, agents, but more like, like non-deterministic pools of agents that just get spun up for like, specific tasks. So I've been thinking a lot about like, and I think Chris bumped into this, some of his work, like, how do you kind of coordinate all these, agenda actors together, which has like some flavor of, like, blockchain based systems in it.
00;27;19;02 - 00;27;32;09
Aaron
Right. Like, how do you maintain a common state, like, how do you make sure that people aren't running over each other? You know, these are tough questions that blockchains have dealt with that you know, get has tried to manage, etc.. So that's kind of where my head is. Derek, what do you think?
00;27;32;11 - 00;27;34;07
Derek
I like it Chris do you have a take?
00;27;34;10 - 00;27;39;04
Chris
AI is going to get better. And sorry.
00;27;39;06 - 00;27;41;17
Derek
I feel like that, doesn't it?
00;27;41;20 - 00;27;42;18
Chris
Yeah. I mean.
00;27;42;18 - 00;27;48;05
Aaron
That's that's what make it to the World Series. Those are two takes. We could probably. Yeah, we could probably sit on Chris.
00;27;48;08 - 00;28;16;17
Chris
No. I'm sorry. I'm like you really obsessing over execution time. And I'm so deep in the weeds of, a certain process right now. And so one thing I will say, right, is, as these models get more powerful, you can get closer and closer and deeper and deeper into, you know, the particulars of how certain flows run. And so what I'm like, what I'm dealing with right now is, you know, how do I shave milliseconds?
00;28;16;17 - 00;28;48;04
Chris
How do I shave milliseconds across like 30 to 40 different surface areas? To get like a cumulative improvement, you know, across the board on a very complex, interaction pattern. Right. What's interesting to me about AI is even if you can't say what a particular, thing, you know, where models will be nine months, 12 months from now, what you can say is your abilities to apply whatever is coming or only going to get better.
00;28;48;04 - 00;29;13;23
Chris
And I don't know, man, like, the more I work with this stuff, right. Like the more I realize what I actually have to have. Like I do itself in terms of like code execution or, you know, running within my system tends to be actually like pretty simplistic. It's like, I want you to run inference at this particular moment on this, this one thing.
00;29;13;26 - 00;29;35;27
Chris
But the complexity is all. And like the payoff is really around, like, what is the right package I need to be using to extract? And then what is the right payload and what is the right sequence of events. It's almost like being an infielder in baseball. The payoff right isn't. And the payoff is in like throwing the first base, right?
00;29;35;27 - 00;29;59;09
Chris
And the first baseman catches it and there's an out. And that's a pretty simplistic thing. But the footwork, the positioning, like all of those little things that are like a muscle memory, you know, to your shortstop there, right. Like that to me, like that. That's just where my head is at right now, because that's the work I happen to be doing and I like, I don't know, I, I'm just like appreciating, at the moment.
00;29;59;09 - 00;30;23;12
Chris
Right. What what better models and, you know, more reliable delivery allows me to like, really get good at in, you know, the basics, the basics, all the things I'm doing. And then to, like, just make that the standard. Right. If I can get this one process to, you know, originally have like a 14 second execution time, I can get down to six seconds.
00;30;23;12 - 00;30;47;11
Chris
Well, that's a new standard for me, right? Like that's the like level of excellence. I'm asking about everything in my system. And so like, I don't know, maybe maybe like just a little light bulb moment for me or what I'm geeking out about at this particular moment in time is not like, oh my God, I have a death ray and I can reach Saturn now and I can blow up the moons of Saturn because this shit is so powerful doing this one particular action.
00;30;47;14 - 00;31;09;19
Chris
No, it's more like I can, I can choreograph Swan Lake. You know what I'm saying? Like, it's it's this combined, level of orchestration married to, like, raising your standards of excellence and just being able to do a whole ton of little things incredibly well like that. That's sort of what, here's got me. I'll get out at this moment in time.
00;31;09;21 - 00;31;29;06
Derek
Yeah, I like that. I like that a lot, actually. Yeah. I mean, I think we will blink our eyes in this. To your point, Chris, I think this stuff is going to be way more usable, way cheaper, way more optimized for discrete and very specific functions. And and it's because the base models are getting better. They're getting larger.
00;31;29;06 - 00;31;55;04
Derek
It's because of all that inference time thinking and energetic harnessing and RL loops that are being applied to these models in very specific directions. The better environments, more multimodal. It's like, you know, being able to interact with your stuff on your hard drive and the programs that you use every day. It's like, yeah, it's all I think will blink, and all this stuff is just going to be pretty integrated into our lives.
00;31;55;07 - 00;32;15;24
Aaron
Well, I think the reasoning is here, which has been the big critique, and now it's going to be about like these types of questions you're exploring, Chris. Like the coordination, right? Like how do you coordinate it? And I think once those get like hammered out, they can and they're not deterministic. Right. Like it sounds like you're exploring like how to deterministically set them up like pretty well and they become non-deterministic.
00;32;15;24 - 00;32;38;08
Aaron
Then I think you get these swarms right. Like just massive groups of a genetic systems, like solving bigger and bigger problems. Right. And that's when I think it gets like pretty interesting because it feels like the, you know, there's like reasoning argument is just like falling away. I think there was like a New Yorker cartoon that like, made a joke that like, like it was like a corporate board.
00;32;38;08 - 00;32;52;01
Aaron
And they're like, I think our position that like, AI is not going to work, should be reconsidered, which kind of made me laugh. But I just feel like that, that, like recognition that like, oh, fuck moment is just happening with millions of people, like, across the board every day now.
00;32;52;03 - 00;33;21;16
Chris
Yeah. No, it's definitely breaking out. I walked out of book club early on Wednesday night, and by early I'm at like quarter of 11. I was like, oh man, do I want to have another Guinness and listen to the to like everyone at this table talking about how they're discovering AI and, using it in their job, or I've, I heard this a million times times before, but the fact that, like, you know, a bar in Queens, you know, a bunch of dads, you know, in their 40s and 50s, right?
00;33;21;16 - 00;33;27;29
Chris
Like that's the topic. Or like, that's the point of diffusion. You know, we're sort of getting to right now.
00;33;28;02 - 00;33;39;05
Aaron
I mean, it's worse in like my neighborhood, I heard people talking about the permanent underclass. I was like, damn, you're spending a lot of time online, like, get off the internet. Don't fall down. That rabbit hole.
00;33;39;12 - 00;33;42;08
Chris
It's not permanent underclass talk.
00;33;42;11 - 00;33;50;19
Aaron
Yeah. So it's definitely moving pretty quick. What do you think's going to happen, Derek I'll turn the question around. I'll put my professorial hat on. What do you think we go.
00;33;50;21 - 00;34;16;27
Derek
Yeah, I, I mean, I largely agree with both of you. I think I just I this the model capabilities are just getting so good. And I also think there's, you know, frontier in different directions like I, I'm, I am pilled on on world models. I'm pretty excited about that effort scaling over the next coming years and filling in some of the holes that our labs aren't able to to solve, but just in general.
00;34;16;28 - 00;34;39;11
Derek
Like, it's hard to imagine these tools are just radically, radically better in in the next 6 to 12 months. And I think where my mind goes is like, what's the direct and secondary impact of of that? And I think people are grappling with that now. I think we we're seeing the earliest signs of both public companies and private companies dealing with that.
00;34;39;11 - 00;35;07;13
Derek
I can tell you my portfolio companies are trying to figure out what that means for them. I think the laundry is being aired in public for the public companies themselves. I mean, you're seeing you're seeing just a lot of shifting around priorities and strategies and to account for kind of new realities around synthetic labor. And yeah, I, I'm, I'm optimistic that there are massive value unlocks and categories that we're not currently imagining.
00;35;07;13 - 00;35;34;05
Derek
But I also think for for a lot of people, it's going to be who aren't willing to adapt quickly or understand the power of the stuff and how seismic, its role will be on our economy. I do think that the next year probably represents a painful period. So I am I'm trying to, rap in my head about all of this stuff, but I think the thing I am confident in is these tools every day I use them are getting better, and better.
00;35;34;07 - 00;36;03;04
Derek
There's just too much. It is a little bit like modular crypto, where it's like, you know, somebody built Ethereum and then somebody built Uniswap, and then somebody built OpenSea and then launched and then you got DeFi. You can collateralize your stuff, I can borrow money and you can go leverage long. And now there's perps. And it's like once, once like a surface area becomes interesting and people just start hacking away at solving problems on it.
00;36;03;06 - 00;36;23;20
Derek
It the ramp just becomes pretty fucking incredible. And we're seeing that with AI right now. It's like, you know, obviously the the labs have kind of a chokehold on the pre-training game. But after these models get created, like I'm seeing so much experimentation, it's like, can we distill the models? Can we get them smaller? Can we run RL in this direction?
00;36;23;20 - 00;36;55;23
Derek
Well, what if we create an RL platform. Can we create a network for inference? That's private. What about a decentralized one. And it's just the energy I'm seeing around. To Chris's point, making these things more valuable, more useful, cheaper, close to zero latency. It's like these are all solved or on the track to getting solved. And and in fact, more and more things are getting unlocked now that we have more and more people and more and more resources and more and more smart brains just hacking away at stuff, and it's just it feels like this next year, it's just going to it's really going to go fast.
00;36;55;26 - 00;37;16;29
Aaron
Yeah. And I'm pumped. And one thing we didn't talk about is just the open models. Right. They're lagging what, a 1:45 quarters at the outset. And I'm just really excited to get a localized models in the open source models. Yeah. And I just think, you know like whatever we're playing around with because we're in the trenches like now will be cheap and widely available by the end of the year.
00;37;17;02 - 00;37;34;12
Aaron
And I just think that that's a huge unlock, because I do think there's a lot of cost pieces related to it. So like, why did Cloud Code have its moment? It really was because it took what Curser was already doing and just made it cheaper. And then a lot of people like, had their, you know, like, holy shit type moment, which is super cool.
00;37;34;14 - 00;37;52;15
Aaron
And I think that the cheaper and cheaper some of the models we're using now, get, you know, just see that like across the board and we'll really kind of like, test out like, what's the limit? Token pieces. Also, I like, on the composability side, it's not like, you know, some of the obvious enhancements and improvements. It's kind of like, what are these, like second and third order effects?
00;37;52;15 - 00;38;09;16
Aaron
And that's where, like that fog of AI, like, kind of jumbles up my brain a bit. But increasingly I just think like a plastic software stack is probably going to blow up, get, like, we probably won't be using it in two years, like, I remember when it came out, it was super cool. But like, at this point, like it's dated.
00;38;09;19 - 00;38;31;11
Aaron
So what does that look like? So like almost like any, you know, angle that you're kind of viewing the tech ecosystem from. It's just like all up for grabs. And I just don't remember a time that was like that. And I don't think even other evolutions, whether it was like electricity or other bits like had the same problem set like surrounding it, that it's like a fun time, hard, hard problems.
00;38;31;13 - 00;38;34;10
Chris
It's like when they discovered chemistry and.
00;38;34;12 - 00;38;35;24
Aaron
Like a little bit. Yeah.
00;38;35;26 - 00;38;37;23
Chris
Kind of. Yeah. You know.
00;38;37;26 - 00;39;10;01
Aaron
Yeah. It's like it's it's like or like writing I don't know. But yeah. Like, you know, even somebody was saying they on our team, they're like, oh man. Like a lot of politicians are going to be, you know, they're really going to regret, like, their early, like, Facebook posts in a couple years. You know, we're kind of seeing that like, like eek out around like, certain politicians or like the relatives without going into that, I just, like, instinctively like wrote back, like, why do you assume in 30 years that they'll even be politicians, you know, like, just like everything's like up for grabs.
00;39;10;01 - 00;39;11;12
Aaron
It's like a complete jump off.
00;39;11;14 - 00;39;47;16
Chris
I agree, I will also say though, like some of the biggest gains I'm getting at the moment is just from, like, picking up, like, useful natural language like packages, you know, and then trying to figure out, like, how to use those to feed into the great machines. And so, like, I, I kind of believe we live in a world of overlays and everything as usefulness and, you know, like there's so much good stuff out there that, you know, like, we'll keep drawing upon our rich technological heritage and combining them in new and interesting ways.
00;39;47;16 - 00;40;00;00
Chris
And so I, I admire your there'll be no politicians in 30 years. That's maybe the most batshit thing I've ever heard that I would love to see come, come true. But, I don't think we're going to escape them.
00;40;00;02 - 00;40;08;23
Aaron
Yeah, I either do, I, but, you know, it's kind of a fun thought exercise to kind of imagine a world that that looks like that radically different. But can you maybe.
00;40;08;26 - 00;40;27;17
Derek
I actually think this is kind of an interesting throw. The point. Can you, like, make the argument like I'm, I'm intrigued by this idea. Like what? What would have to happen in terms of our ability to coordinate our own selves and our own resources, both within this jurisdiction and internationally, where we wouldn't want to offer that to human politicians.
00;40;27;17 - 00;40;30;27
Chris
De la yeah, I think it's okay. Time I I've got.
00;40;30;29 - 00;40;34;19
Aaron
Chris is buckling up buckle and Eric.
00;40;34;22 - 00;41;00;12
Chris
All right. Yeah. I'm about to bring some own to the table. Who is one of the, most interesting. I'm sorry. Some some own. Very one of the most interesting and, like, just impossible thinkers to have come out of, like, 20th century France, right? She was, she was like, both a philosopher and a bit of, like, an idealistic, to the point of, like, mysticism.
00;41;00;12 - 00;41;21;26
Chris
Like she was like a contemporary of, like, Simone de Beauvoir. I went to school with her. Right. I just chose this really, really crazy path. But one of, her books is called On the Abolition of All Political Parties. And like, so long, they might seem like the most dangerous person I read because she is so dead right about everything she says.
00;41;21;26 - 00;41;56;14
Chris
But it's also absolutely impossible. And if you were to actually, like, follow her way of living, like she literally died of starvation in a hospital because she was refusing medical care out of solid solidarity with, freedom fighters in Spain or something like that. Right? Like amazing, and, like incredibly complicated person. But like, she wrote this whole book on the abolition of all political parties that, you know, basically just walks through why the concept of a political party leads to poor governance.
00;41;56;14 - 00;42;26;29
Chris
Right? Because your loyalty and your solidarity always has to be with your party and therefore, whatever your actual like beliefs are or what you got into this thing for becomes secondary, tertiary, and eventually obliterated by your place within an organization. And what the organization expects from you in order to perpetuate itself. Right. And so, like, I mean, we can just see that today, right?
00;42;26;29 - 00;42;47;08
Chris
Like she was she was right about it. She's also impossible. Like, how do we how do we get rid of, politicians and. Well, the way to start with that is actually to break. I mean, you'd have to do this incrementally, but you have to start by breaking the two party system, and then you would actually have to break political parties, period.
00;42;47;11 - 00;43;13;00
Chris
And only when you get to like some approximation of like New England small town governance, you know, or, or, dao consensus. Right. In a massively scaled environment. Could you then look at the actual perpetrators of why your politics are so terrible political parties and say, we don't need that shit? That, to me, is how you'd actually have to go about doing this.
00;43;13;02 - 00;43;34;27
Aaron
Yeah. I don't know how it how it happens. And I think those are great points. Thanks for layering in, some French thinkers always a good way to do it. The way I think about it, Derek is more like at some point in the next couple of years, I think it will just be like common understanding that these systems are faster, better, you know, more intelligent.
00;43;34;27 - 00;43;53;19
Aaron
And they have biases, but they're manageable biases. Right? You can like, aggregate the different models and get kind of like a convergence of like some sort of like answer like truthiness to it. And once that happens, I think people will start looking at the expert class, the remaining expert class, and say like, well, why don't we just replace them too?
00;43;53;26 - 00;44;12;04
Aaron
And maybe it starts first with like courts, the courts are super slow, very bright people there. They do in generally some states like a really good job, but it's unevenly applied, like across the planet. Not everybody has access to like, good judges, you know, fair decisions. Well, now you've got this universal system that you can call for.
00;44;12;05 - 00;44;44;12
Aaron
You know, fractions of a penny and get that high end reasoning to, like, make your life better. And so you could see, like algorithmic decision making coming to courts, right faster, hopefully more fair if architected the right way. And then that whole kind of branch of the government just gets reimagined. Same thing with like aggregating preferences. Like as the memory systems get better, I should be able to like, store and signal my preferences, publicly, in a way that like an agent can aggregate that, which would be a super hard task today.
00;44;44;12 - 00;45;02;13
Aaron
Right? But like a trivial task for an AI system, and therefore my preferences should be able to, like, cascade up into, like a democratic system or some sort. And so if that happens, then, you know, more democratic, wings of the government, but that would kind of also collapse. And then I think the last piece is just like an executive.
00;45;02;13 - 00;45;22;21
Aaron
So like that risk calculus that we're asking the executive branch to do at different, different pieces. I think it it kind of falls back to that expert piece, like, I'm not convinced that, you know, an X number of months, years, etc., that that isn't also replaced. And then you have like kind of an entirely algorithmic system. So I don't know if that's better or worse how people feel about it.
00;45;22;21 - 00;45;43;08
Aaron
It's very complicated inquiry, but I think technically it should be become increasingly feasible probably by the end of this decade. Right. And I think people will start experimenting with it. And, you know, maybe it works better, right? Maybe it works better for city. And they're allocating resources more efficiently and at lower cost. And tax rates go down and more people choose to like, opt into that system.
00;45;43;08 - 00;46;04;06
Aaron
So I think it would take a while for it to disseminate. But I think the tech is probably, getting pretty close to being ready there. And I think folks may remember like this great debate between like Reed Hoffman and, and Peter Thiel about like AI and crypto, where I like access a central planner. And I think that that's a worrying concern.
00;46;04;10 - 00;46;28;19
Aaron
So I think the questions may become like, where do we make sure, like humans are still in the loop in that process? Is it, like expressing your preferences? Is it as like a this is like a first draft and you have to it has to be approved. I don't know exactly what that looks like, but maybe at the end of the day, it is something like you just have your people make those devices with all the the new keyboard, which is just like, you know, yes, agent, you know, do better, like all these different little buttons.
00;46;28;19 - 00;46;34;15
Aaron
Maybe, maybe that's kind of what it looks like at the end. Like a final check on preferences or some escape hatch of some sort.
00;46;34;17 - 00;46;36;01
Derek
I like it, I like both of your answers.
00;46;36;01 - 00;46;52;16
Chris
A lot. Were you the guy who was, talking about a return to, like, human interaction at the start of this podcast? And now now, I suppose like human interaction models and you're just like, screw that shit, man. I'm just so algorithmically good that they can replace governance.
00;46;52;16 - 00;47;10;21
Aaron
So this is why this stuff is tricky because like, and, and I think it's tricky for the designers, maintainers, creators of this. Like, I think we have to think deeply about how to make sure that it isn't like completely that way. And like, what does that look like? And where do we like, add kind of value.
00;47;10;27 - 00;47;30;10
Aaron
But I think this is a critique of like some folks that are more like utilitarian, right. Like they they may not have like humanity at the center. And so they could build a system that's super efficient, but it's, you know, it's not socially beneficial. Right? It's not like pro human. So I don't I feel the tension, I hear it I think it's a fair, fair to call that out.
00;47;30;10 - 00;47;38;24
Aaron
But I think I think we'll find the right balance. But we have to start having the conversation kind of related to that. All right. That was my weaselly way out of that one. Chris had I.
00;47;38;24 - 00;47;41;11
Chris
Do very, lawyerly of you. Yeah.
00;47;41;11 - 00;47;59;15
Aaron
Thanks. Yeah. I don't know. Derek like to me, I think like, like I as a judge, I mean, that's what people are calling it. I just feel like courts are a good place for that to happen next. Personally. Just like a super slow, very important, you know, like un disrupted by quote unquote software part of the economy.
00;47;59;15 - 00;48;09;05
Aaron
And, I do think there's a lot of social benefit from that that could, could like could emerge, you know, people not having to fight about dumb things for as long.
00;48;09;07 - 00;48;11;04
Derek
Yeah, I can definitely see it for sure.
00;48;11;09 - 00;48;15;28
Aaron
What do you think of that vision, Derek, is that, do you have any feelings about that? If that's the. Yeah, I mean.
00;48;15;28 - 00;48;44;19
Derek
I, I, I largely agree, I think I, you know, I mean, I largely I like the, the thinking around this topic from both of you. Yeah. I'm a, I'm a believer that like a lot of these that legacy systems just like are already on the decay and that accelerates with these tools. And yeah, I don't underrate our human behavior to kind of like keep these systems in place for long past their expiration.
00;48;44;19 - 00;49;09;24
Derek
But I do think on pretty much any timeline these things will start to get phased out and less and less important and replaced with these types of tools. I think I'm curious to see the ideas around how that happens and the sequencing required to make that happen, but I think I generally agree with the trend, which is so it's it's a fun idea to think about and kind of like reason through.
00;49;10;00 - 00;49;17;20
Aaron
All right. I'm going to hard pivot to something you guys see. That, Cliff is behind bars now or got arrested. Seems like.
00;49;17;23 - 00;49;19;28
Chris
Thank God you brought some three to the show.
00;49;20;00 - 00;49;32;24
Aaron
Yeah. This is my pre, impression. Yeah. Got he's, I think, completely cooked. Or at least the beginning of his, his ascent or descent off of the the world stage is, beginning to happen. Kind of. Right.
00;49;32;24 - 00;49;38;26
Chris
I only saw the briefest of blurbs about this. He shot an alligator. Is this what.
00;49;38;26 - 00;49;44;01
Aaron
Happened? I don't know, I mean, what really is that? What happened? Oh, man.
00;49;44;03 - 00;49;58;24
Chris
Like I said, I only saw, like, the briefest of tweets around, this is your your man. He's your your trash obsession. So why don't we just do a quick, slow vehicular search.
00;49;58;28 - 00;50;00;21
Aaron
To see what happened? Yeah, let me know.
00;50;00;21 - 00;50;30;20
Chris
And see what happened to, Yes. You're a man. You've been arrested for battery in Florida. All right. To get over to TMZ here. Maybe that's why TMZ, by the way, you're like, oh, everything's going to change. And here I am on my TMZ.com. Misdemeanor battery charge. You do? Well, earlier in the day, he was seen, in the Florida Everglades on an airboat where he came across a dead alligator floating in the water and, fired some shots off of that.
00;50;30;20 - 00;50;33;15
Chris
So he's not a live animal murderer.
00;50;33;17 - 00;50;34;13
Aaron
Well, that's good.
00;50;34;15 - 00;50;39;29
Chris
Yeah, but no, TMZ Hall is telling me is, arrested for battery, so.
00;50;39;29 - 00;50;40;12
Aaron
Yeah.
00;50;40;12 - 00;50;42;12
Chris
So become the main character, right?
00;50;42;14 - 00;50;46;24
Aaron
Yeah, that's the problem. When you're maxing anything out, right? You become the main character.
00;50;46;26 - 00;50;52;17
Chris
It's, Derek. Derek dropped as soon as we started talking. Killer vehicular here. He must be a portfolio company.
00;50;52;22 - 00;51;17;22
Aaron
Probably. I do think some one other big news and it got it took a little bit of time for it to get some attention was Google Research is Turbo Con. I don't know if folks saw this, but it's, compression algorithm. It's literally making, like Pied Piper from Silicon Valley, like, come to life, but it reduces like, low key value cache by six x and delivers like eight x speed, you know, allegedly with zero accuracy loss.
00;51;17;24 - 00;51;40;05
Aaron
So it's pretty cool just to see like these core innovation just like happening. Right. When people start grumbling about it, it's kind of wild, right? Like you people are getting worried about like Ram usage or memory usage. And then here comes like something that just solves it. Like in that moment it feels like a flywheel between like big problem and solution is getting smaller and smaller, which is pretty crazy.
00;51;40;12 - 00;51;44;20
Chris
Take a vector, turn it on its side, polish it up a little bit.
00;51;44;28 - 00;52;04;14
Aaron
Yeah, but just like that speed. Like, I don't know how long that was in development. It could have been very well in development for years. I mean, I'm sure that across Google they're thinking about those like deep problems quite seriously. So I thought that was pretty interesting. And also it finally feels like, Apple's like beginning to think through what its AI strategy will be.
00;52;04;14 - 00;52;23;12
Aaron
Right? And it feels like they're just going to be the App Store for it. So they're going to allow, like any AI platform, to be queried in Siri and enable like extensions inside of their iOS, macOS and iPad OS apps, which is which is pretty, pretty smart. I think on their part, they're just going to be the aggregation layer, not like a provider.
00;52;23;14 - 00;52;28;22
Aaron
So it seems like we finally got some answers of where they were strategically pointing themselves.
00;52;28;24 - 00;52;35;26
Chris
Obviously I can see them pitching it right now. Here's our new API strategy. We're going to take 30% cut on x 402.
00;52;35;29 - 00;52;48;02
Aaron
Yeah I mean that's pretty much it, right? Yeah. Like like to your point that nothing changes, right. And maybe that's the the actual endpoint. Apple didn't didn't veer at all. They're just like we're going to keep on doing what we're doing.
00;52;48;05 - 00;52;57;13
Chris
It's like no one knows about this shit yet. They can't take us to court over it yet. Let's fucking go. 30% on a stablecoin payments.
00;52;57;15 - 00;53;19;17
Aaron
Yeah, like they'll take the skim on that, which is kind of wild. So yeah, I mean, it feels like at least we kind of know where all the chess pieces will be until the end of the year to kind of answer, Derek's question, at least, at least from the big docs in the, in the AI space. I guess the other news just kind of running through stuff, purportedly, there was a big compromise in the crypto bills.
00;53;19;19 - 00;53;44;27
Aaron
There was a big question debate, like how much of the yield that comes from, buying treasuries, if you tokenize a dollar with a stablecoin, could be passed back to customers, or holders of that stablecoin, Coinbase and some other, you know, stablecoin slingers, they wanted to pass that back to, you know, the holders, you know, kind of share the, the yield, build something that may be better than a money market account.
00;53;45;00 - 00;54;02;11
Aaron
I think the banks on the other side did not want to do that because they like to lend out that capital. That would make it difficult for them to kind of do things the way that they're doing it. They kind of reached like a compromise in the middle where it's the stablecoin is being used actively, and people don't know what that exactly means.
00;54;02;13 - 00;54;21;22
Aaron
Then you could pass back yield. But if it's just being held in a stablecoin, or if the stablecoin is being held in a wallet, you're not going to be able to do that. So maybe that's the right balance. Maybe it's a little clunky, not 100% sure. But that's been kind of like moving moving forward. And hopefully there'll be even more clarity just around the rules, the road for, for crypto.
00;54;21;25 - 00;54;41;16
Aaron
So I thought that was kind of interesting. Feels like the unsung hero. There's like tether, which just continues to. And we'll presumably just be able to continue to just print for the next, you know, a X number of years, after this stuff gets kind of solidified. So that was another big, big thing for me at least.
00;54;41;18 - 00;54;42;16
Aaron
Chris.
00;54;42;18 - 00;54;44;28
Chris
Congrats on your regulatory clarity.
00;54;45;01 - 00;54;58;16
Aaron
We finally got it. There's ten, you know, ten times less people in the space, but we got the regulatory clarity, mission accomplished, by everybody involved. I think that was the intent. Right. So everybody, everybody one and that one.
00;54;58;18 - 00;55;07;19
Chris
I watched the light yesterday. Yeah. And I like the new look mats. We have a sample size of one, but 162 and oh is on the table.
00;55;07;19 - 00;55;08;23
Aaron
I think it's going to happen.
00;55;08;26 - 00;55;28;24
Chris
How could it not? Yeah. Like every inning we're going to see 35 pitches every at bat. We're going to foul off eight eight of them. It's going to be going to have for our games. We're going to play station to station. We're not swinging for the fences. We're just moving the runners along. And you know, Squib City, here we come.
00;55;28;29 - 00;55;43;22
Aaron
Yeah. I mean, I like that strategy. I mean, it was fun to see like, Paul Skins is very good at what he does, like just get roughed up a little bit. So was happy about that. Yeah. I mean I think maybe the moves. Right. Maybe we got some hope here and, and, the Big Apple.
00;55;43;24 - 00;55;45;00
Chris
I'm ready for it.
00;55;45;02 - 00;55;56;17
Aaron
Yeah. I mean, it's it's good. It feels like, like spring has sprung a bit here. Like the flowers are coming out of the ground. March was a wild one. I guess we'll see what April kind of brings, but definitely, like, fun times abound.
00;55;56;19 - 00;56;06;19
Chris
We're, like, hitting an important benchmark in my life, which is. Can I wander around the house without socks? Is it warm enough that I don't have to put the stinking socks on, man?
00;56;06;25 - 00;56;21;18
Aaron
Yeah, for me, it's always like, can I not bring a jacket? I just feel like it's so much. It's so freeing to just, like, be able to, like, walk out at the door. Right? We're not have to worry about all that, but I'm definitely ready for some, some football ups and, and none of that other stuff, so.
00;56;21;25 - 00;56;23;20
Chris
Let's fucking go.
00;56;23;24 - 00;56;51;07
Aaron
So let's call it their welcome again to society this week it was Chris, Derek and me talking all things crypto, AI, tech culture, everything in between. And just a reminder that everything we say here is said in our personal capacity, does not represent financial advice and does not represent the views of our, employers. So let's go.