为什么大家都这么看空：Robinhood 的 AI Agent 与 SoFi 的稳定币上线
Jason YanowitzSantiago Roel Santos
- 这轮熊市基本面好于2018年，但心理层面更糟，而且两位主持人对原因严重分歧。 Santi回忆，2018年他和Mike还在讨论如果行业死掉要不要去Uber或Netflix找工作；现在的问题只是“如果代币再也涨不起来，我们该怎么办”——价值最终归于代币还是股票，而不是公链是否行得通。Santi更悲观的反驳是：“比死亡更糟的是变得无关紧要。”他表示，团队研究约30个行业后发现，稳定币的应用场景远比行业此前想象的窄。
- Santi的仓位高度集中：看多加密，但只押“少于5个名字”。 他持有Tether，认为这仍是“一门非常好的生意”，也是汇款领域的监管套利工具；另一个持仓是Hyperliquid。整场讨论中提到的稳定币用途包括博彩/市场准入、无法出金的链上资本，以及希望更快收到美元的人；Santi怀疑单纯做稳定币转账赚不到多少钱，因为价格“基本就是成本价”。
- Yanowitz的反驳本身就是一份实时案例：一家拉美应付账款/账单支付公司——“把Ramp和Bill.com合在一起”——在客户甚至不知道背后用了稳定币的情况下，年化营收跑到8000万–9000万美元，且月环比增长40%。 Rain——“我们公司有史以来最大的一笔投资”——每月新增约20%的增长；Visa结算的稳定币交易量“每隔几个月就翻倍”；Airwallex的Jack Zhang也从反稳定币转向思考做自己的Layer 1。核心判断是：稳定币不是点对点支付，而是“一套全新的结算层”。节目还提到一笔未披露交易对手方名称的全球市场平台交易。
- 谈到Robinhood的AI Agent和Liquid的“co-invest”（通过声称获得CFTC不采取行动函豁免，在ChatGPT和Claude内交易Hyperliquid永续合约）时，Yanowitz对其法律理论只给出一句：“祝你好运。” Santi把问题拆成两半：通过AI前端进行自然语言零售交易不会占主导地位——“人们还是会希望自己的金融业务存在于金融体系里”；但机构化、系统化的Agent交易迁移到交易所“不可避免”。他提到一家加密交易所CEO用历史数据训练OpenClaw Agent，结果“亏了很多钱”。Yanowitz不会把资金交给聊天应用：“政教分离”。
- Morpho的“Midnight”白皮书重新押注固定利率、固定期限的链上信贷——这是一个此前被Notional、Element Finance和Yield等项目反复尝试、几乎成了项目墓地的品类。 Santi认为，历史问题在于没有统一的基准利率，以及链上资本期限极短。Sky/Maker的治理风险最能说明问题：几个月内的基准利率调整，相当于“美联储把利率从5%加到15%”；Yanowitz向抵押贷款发起商Better询问其Sky融资是否设有利率上限时，“他没有告诉我”。至于机制——在单一到期日买入信用单位、卖出债务单位——“这就是Pendle”。
- 双方都同意，DeFi安全恐慌是真实存在的，但问题被诊断错了。 针对前OpenZeppelin创始人“DeFi里没有任何东西是安全的”以及建议链上投入资本为零的说法，Yanowitz认为几乎所有重大攻击都不是智能合约漏洞，而是OPSEC和社会工程问题；AI正在加速这一过程，因此所有软件的黑客攻击都会先上升，直到“最终黑客攻击基本会彻底消失”。Santi的实际建议是：链上交易不可逆，让过渡期格外残酷，因此“先等6个月”；两人还建议加强终端监控、设备安全和身份安全，因为Drift攻击者在出手前已经潜伏了一段时间。
- 从华盛顿回来后，Rob认为Polymarket给CLARITY Act的约55%概率被低估了——这与他此前认为50%–55%已经太高的判断完全相反。 他预计众议院全院表决更可能在7月，行业押注6月“太快”；当前两大风险是ABA的游说，以及民主党要求加入的、涉及从加密和加密业务中获利的伦理条款。监管层面，他预计CFTC会出台建设性的永续合约指引，允许类似Polymarket的境内/境外架构；中期则会出现现实世界资产永续合约和全天候现货交易。
- Saylor那句“本周我们不买Bitcoin，我们买债券了”震动了信徒——“就像耶稣说自己不能在水上行走。” Strategy的美元股息覆盖期已经降至6个月，因此进一步出售BTC“似乎很可能”；节目引用其NAV为1.2。Yanowitz的固定原则是：“所有金融工程……最终都会爆掉或自行解体——如果你想获得Bitcoin敞口，就直接买那个该死的东西，至少买IBIT。”机器人投资载体RoboStrategy的交易价格接近4倍NAV，考虑到顾问团队25%的稀释后可能更高；Santi和Yanowitz更倾向于做一个跟踪私营公司最新估值的合成产品。
1. 一场比2018年更好的熊市——除非加密最终变得无关紧要
- Santi的基准判断是，2018年是生死存亡的一年——“我记得当时和Mike一起喝啤酒……我说如果行业没了，我就去Uber找工作，Mike说他会去Netflix。”现在的问题更窄：“如果代币再也涨不起来，我们该怎么办？”所有资本市场——股票、债券、货币、商品——都会迁移到链上的命题，“过去两年得到了前所未有的验证”；剩下的悬念是，价值最终归于代币，还是重新回到股票。
- Yanowitz同意行业处于“远远好于历史任何时候”的位置，但他指出了市场对时间线悲观的来源：2018年那批人并没有暴露在真正运转的业务上。“行业里很多人觉得自己正在被甩在后面”，而支付和代币化资产仍在持续增长。
- Santi的异议构成了本期节目的主线：“比死亡更糟的是变得无关紧要。”他的团队评估了约30个预期能受益于稳定币的行业，却反复得出结论：“我们就是觉得稳定币没有发挥作用的地方。”这不是技术问题，而是利益相关方是否采用的问题。讨论中提到的稳定币用途包括卡在链上、无法出金的资本，以及希望更快收到款项或收到美元的人；对某些地区的人来说，博彩是尤其强大的加密用例，市场准入也是重要场景。“其他所有——”话说到这里，Yanowitz打断了他。
- Yanowitz用互联网时代的类比解释博彩为何最先爆发：互联网让信息流动起来，所以色情内容最先爆发；“当加密让价值流动起来时，什么会爆得最厉害？博彩。”
2. Yanowitz的稳定币证据链，对比Santi的五标的集中仓位
- Yanowitz的反驳是，怀疑者并不在真正发生变化的房间里——他曾在Twitter上交锋的一名Brookings Institution研究员，“甚至不知道Visa已经在直接结算稳定币”。他最关键的案例是一家为拉美大型企业提供应付账款和账单支付服务的公司——“想象一下Ramp和Bill.com合二为一”——月环比增长40%，年化营收跑到8000万–9000万美元，整套基础设施都建立在稳定币之上。“它的客户根本不知道。”
- 按Yanowitz的说法，增长动能还包括：Rain每月新增约20%的增长；节目提到一笔交易对手方未具名的全球市场平台交易；Visa结算的稳定币交易量“每隔几个月就翻倍”；Airwallex的Jack Zhang——“一度是Twitter上最反稳定币的人”——如今开始发帖讨论打造自己的Layer 1。重新定义的关键是：不要再把稳定币理解为点对点转账，它们正在“推动一套全新的结算层”。
- Santi承认稳定币有用，但坚持要看规模和集中度：“我绝不是说稳定币没用……细微之处在于，它到底有多大、又有多集中。”他的仓位仍然“非常看多，但集中得多——在加密领域可能少于5个名字”，Tether和Hyperliquid是其中的例子。对转账业务，他的质疑是：“稳定币转账赚不了那么多钱”——“我们谈过Tempo……单靠稳定币转账赚不了多少钱。”
- Yanowitz的回答是金融科技的切入点模式：Stripe最初只是支付网关，后来扩展到信贷、会计、虚拟账户和银行卡；那家拉美公司已经在向客户加售费用管理软件。“如果把更好的转账和更好的结算作为切入产品，就能建立非常大的业务。”Santi对终局的判断是，全球银行最终会比现有银行更好地服务中小企业和零售消费者，原因就在于稳定币。
3. 双方共同承认的一点：高度金融化的资本市场才是TAM
- Yanowitz承认2018年真正看错的地方，不是万物加密化失败，而是“全球资本市场比任何人意识到的都大得多——大了几个数量级。只要我们把全球资本市场迁移到公链上，这就是一个疯狂的机会。”
- Santi通过此前关于高度金融化的判断表示认同：“任何有互联网连接的人，都可以做多SpaceX上市前的预测市场。”即使最终赢家高度集中，TAM仍在扩大。Yanowitz补充了正在形成的产品管线：Ondo正在与OKX推出代币化RWA、合成资产和市场；而预期中的CFTC永续合约指引将覆盖在链上结算的代币化合成资产，这些市场“也需要用稳定币结算”。
- Yanowitz的总结形成闭环：市场之所以对时间线悲观，是因为真正获胜的命题“很无聊……而且不一定会把价值归于那些人们希望上涨的代币”。
4. Robinhood的Agent、Liquid的co-invest，以及Claude会不会吃掉金融科技
- 最新消息包括：Liquid的“co-invest”把Hyperliquid永续合约交易直接放进ChatGPT和Claude，据称凭借CFTC不采取行动函豁免，在全美50个州都合法；Robinhood则宣布了一套AI Agent架构，可以接收外部Agent，为其配置隔离的钱包或账户，并设置消费上限。对于不采取行动函的法律理论，Yanowitz只给出一句：“祝你好运。”随后又补了一句：“一路顺风。”
- Santi提出的核心问题是：社交媒体吞掉了没人预料会被吞掉的媒体公司，ChatGPT和Claude会不会吞掉金融科技公司——“10年后我们还会有SoFi吗？还会有Robinhood吗？”Yanowitz个人的答案是否定的：“我不想通过Twitter交易……政教分离。”不过他也承认：“我完全可能在这里错得离谱。”
- Santi把问题拆开看：通过AI前端进行自然语言交易，可能不会占主导地位——现有初创公司的产品“相当糟糕”；但由机器学习驱动的Agent交易从机构端迁移到零售端，“不可避免”。他讲了一个案例：一家大型加密交易所的CEO用历史交易数据训练OpenClaw Agent，并让它管理自己的账户——“公平地说，他告诉我，结果亏了很多钱。”
- Yanowitz认为，围绕风控边界和参数设定可以形成一门生意，逻辑类似券商开户时建立风险画像；但耐久性存疑：Anthropic和OpenAI可能会碾压任何获得市场 traction 的产品。他还提出了一个更长周期的类比：Bloomberg最终可能会意识到，“我们为什么不干脆让人们直接通过这个平台交易？”
5. Morpho向一个项目墓地品类推进固定利率
- Yanowitz的切入点是，固定利率是DeFi“缺失的基础组件”。在Compound或Morpho Blue上，利率跟随资金池利用率变化，因此借款人承担利率风险，企业也就根本不借款。传统金融的类比是30年期按揭——节目将其描述为通过掉期把浮动利率锁定——而真正的突破口是拥有实际收入的协议：假设Uniswap产生1亿美元收入，就应该能够围绕一笔固定利率贷款做规划，由此打开企业借贷、按揭以及机构/RWA借贷市场。
- Santi认为，过去许多尝试“确实运行得很差”。节目另一位发言者提到公司投资过Notional和Element Finance；Santi则回忆Yield也尝试过这一方向。复盘的核心问题是没有统一的基准利率——不同项目把ETH质押收益率、Maker或Compound当作替代央行——以及链上资本市场“期限极短”。收益耕作循环不需要30年期固定利率。治理风险进一步放大了问题：Maker/Sky在几个月内的利率变化，“相当于美联储把利率从5%加到15%”。
- Yanowitz给出一个正在发生的案例：抵押贷款发起商Better曾从Sky获得一笔规模可观的融资；当他询问这笔交易是否保证了利率上限——因为Sky治理可以“随心所欲”地把利率从3%调到10%——“他没有告诉我”。Rob的回应是：“这就说明问题了。”
- 至于机制——设定单一到期日，买入信用单位、卖出债务单位，到期按1:1赎回——双方立刻达成共识：“这就是Pendle。”但Yanowitz从Pendle的历史出发提醒，Pendle是在Ethena收益耕作敞口上增长起来的，因此真正悬而未决的问题是，“常规经济活动”会不会上链，还是最终仍停留在交易和收益耕作。
6. “DeFi里没有任何东西是安全的”——恐惧正确，诊断错误
- 事件起点是OpenZeppelin创始人的发言——Yanowitz指出，他已于2019年离开——他发帖称DeFi里没有任何东西是安全的，并表示自己建议家人、朋友和自己“不要把任何资本放到链上”——是彻底的零投入，而不是“目前先不要”，Santi特别指出了这一点。Santi还说，他私下听到信用基金及其他曾活跃于DeFi的参与者，正在集中研究Aave上的稳定币问题，以及Kelp事件是否已经解决。Kelp在法律层面尚未完全解决，但用户基本得到了补偿。
- Yanowitz的反驳是：“大多数黑客攻击，几乎全部，都不是智能合约攻击。”它们是OPSEC失误，涉及用户、密钥持有人或签名者遭到入侵。AI具有双重作用：攻击者“只需要找到一个漏洞”，一旦进入安全边界，AI就能让其“比过去任何时候都更快”地找到并抽走资金。他的判断是，所有软件的黑客攻击都会在一段时间内增加，“然后最终黑客攻击基本会彻底消失——只是重写代码需要时间。”
- Santi从实际操作层面补充说，链上交易不可逆，使这一过渡期格外危险，因此应“先等6个月”。他引用一名CISO的建议：启用终端安全和监控，因为攻击者可能已经潜伏在系统里；Drift就是延迟发动攻击的例子。另一位发言者还建议，使用Claude Code时准备独立设备、持续监控，并设置一个用于证明真人身份的安全口令；氛围式编程则会进一步扩大攻击面。
7. 华盛顿判断：CLARITY约55%的概率仍被低估，永续合约指引将至
- 与国会工作人员和监管机构会面后回来的Rob，推翻了自己此前的判断：“过去Polymarket显示CLARITY的概率是50%–55%时，我会说，这太高了。现在我认为概率高于55%。”市场概率曾低至约36%，目前约56%，他认为市场“有些低估了它”。他预计全院表决在7月，而不是行业押注的6月——“太快了”。
- 当前两场真正的争夺是：ABA仍在强力游说，尽管Tillis已经把法案移出银行委员会；民主党则要求加入有关从加密和加密业务中获利能力的伦理条款，白宫需要接受这一条款。全院辩论还会触及BRCA对开源代码、验证者、前端和智能合约开发者的保护；Rob认为这些措辞“可以解决”。
- 在监管方面，Yanowitz说CFTC主席Selig曾在3月讨论永续合约指引。他预计指引会偏建设性，允许采用“类似Polymarket今天使用的”境内/境外实体架构；短期主导市场的将是永续合约和预测市场，中长期则是“现实世界资产的永续合约”和全天候现货交易。他最后指出，很多积极变化正在发生，但价格走势并没有像预期那样让人兴奋。Santi给出的冷静剂是：Nasdaq花了15年才收复前高。
8. Saylor买债券；DAT解体规则；一个4倍NAV的机器人投资载体
- 本周最大的冲击是Saylor那句“本周我们不买Bitcoin，我们买债券了”，以及市场流传的一笔10亿美元IBIT交易。Santi说，这动摇了那些把Saylor“基本当成Bitcoin耶稣”的人：“就像耶稣说自己不能在水上行走。”关键数据是：Strategy把美元股息覆盖期缩短到6个月，因此进一步出售Bitcoin“似乎很可能”；其NAV被引用为1.2。
- Yanowitz的固定原则是：“所有金融工程，无论是DAT还是其他形式，最终都会爆掉或自行解体——而且通常不会轻松解体……如果你想获得Bitcoin敞口，就直接买那个该死的东西，至少买IBIT。”
- 同样的框架也适用于RoboStrategy，这个机器人投资载体的交易价格接近4倍NAV，计入顾问团队25%的稀释后可能更高。Santi认为没有足够有说服力的理由支撑这一溢价，并重新提出Yanowitz此前的合成产品思路：可以围绕Apptronik、Figure或其他私营公司创建合成资产，跟踪其最新一笔记录在案的交易。“加密非常擅长为资产尾部寻找流动性”，因此，做一个合成资产可能比支付进入私募股权的溢价更划算。
完整逐字稿
Nothing said on Empire is a recommendation to buy or sell any investments or products. This podcast is forformational purposes only and the views expressed by anyone on the show are solely their opinions, not financial advice or necessarily the views of Block Works. Our hosts, guests, and the Block Works team may hold positions in the companies, funds, or projects discussed.
Oh, you don’t like to see that.
What’s up, folks? Welcome back to Empire. I clicked record and my MetaMask popped up. You don’t love that.
Jeez.
You don’t love that. Oh my God, it keeps popping up. What’s going on? That’s dangerous. What’s up, guys? How are you doing?
Loving life. Short week this week.
1. Why’s The Timeline So Bearish?
We will be optimistic. We should have a barometer of how bad things are—ultra fear or euphoria. The timeline’s bleak, guys.
I’m just going to ask the question: Is this better or worse than 2018? I saw a really interesting tweet about this. Someone said this is probably worse than the most recent cycles, when you had FTX blow up, Luna blow up, Celsius, and some of TradFi. You could pinpoint it: There’s a very clear reason why prices are down or not moving.
Now it’s quite the opposite. You have TradFi continuing to be really bullish on this. I think maybe marginally less so. I’d actually be curious, Santi—I know you were in D.C.—what are you guys seeing in terms of the conversation with the people who have been the most bullish this cycle? They’ve been non-crypto natives. Is that true? Is it marginally less so? Is it just “sell in May and go away”? Are people thinking about the summer? What’s going on?
We’ve got six starting questions tied into one there.
Yeah, I like these bundled questions, man. I’m terrible at single questions.
That’s why we were so happy to have you come back, Yano, after your break. We needed a moderator.
Is this cycle worse than the last 2 cycles? What’s going on in D.C.? What are we doing with the market? Do we sell in May and go away? Hit us with the answer.
Hit us with ma'am. Listen, man. People have their attention span of 30 seconds. I just want to tell them what's on my mind.
Before we talk about D.C., let’s try to answer that question. I would say this is a much better cycle than 2018 or 2019, because I remember sitting with Mike, getting beers after work one day, and talking about what we would do if all of crypto failed. I said I would go get a job at Uber, and Mike said he would go get a job at Netflix. We were talking about what we would do if the entire industry didn’t exist.
I think the question on people’s minds today is, “What will we do if the tokens don’t go up again?” But there’s no question that you can build incredible startups and incredible companies in this industry, and the value there will accrue somewhere. Maybe it’s not to the tokens; maybe it’s to the equity. That’s the question on people’s minds today.
It’s not whether open public blockchains will work. The thesis I have—and I guess maybe you guys do, too—is that all capital markets move on-chain: stocks, bonds, currencies, and commodities. I also believe in tokens. Everything moves on-chain. That thesis has been validated more than ever in the last 2 years.
The idea now is whether value will accrue to these things called tokens, or whether we’re just going back to equity. That’s very different from what it looked like in 2018.
Yeah, totally. By the way, I was just looking it up. I think Uber was worth $28 billion in 2018. It’s worth $143 billion today, so you would have gotten a nice little return there.
That’s the trade.
Sounds like you’re doing okay. Listen, I agree with you 100%. In 2018, it was much harder, in my mind, to take the rational position that all of these ICOs were real companies building real things that were going to upend traditional capital markets and payments.
It was fun before the downturn, anyway. It was still a niche and weird industry where you could find pockets of people who were really dreaming about what the future would look like. In many ways, today it’s still a lot of dreamers saying, “Capital markets and payments are going to come on-chain, and we’re just going to get everything tokenized.” But that feels far closer than any of the things we used to talk about before.
It feels like something we’re getting a lot of institutions telling us is actually happening. We’re getting a lot of people putting their weight behind it and a lot of money going into it. I 100% agree that, from an industry perspective and in terms of adoption of blockchains themselves, we’re in a far better place than we’ve ever been.
Now, I think the problem is that a lot of the people who were here in 2018 did not care about that, and those are not the things they have exposure to. Those are not the things they invested in. A lot of people in the industry feel like they’re getting left behind right now because of the fracturing that’s happened. That’s hard for people, and I think that’s a lot of the reason for this timeline that feels very bearish at a time when payments continue to grow, tokenized assets continue to grow, and adoption continues to grow.
To Santi’s point—we’ll talk about it later—I was in D.C. yesterday, and the amount of education that has happened in D.C., along with how much attention people are paying there, is incredibly bullish in my mind.
I actually disagree. I think there’s a higher probability now that crypto just becomes irrelevant. What’s worse than being dead is being irrelevant, and there’s a very credible path to that, from my standpoint.
We’ve gone and looked at probably 30 different sectors at this point. Many of them, coming in, we would have thought, “Could stablecoins actually be pretty interesting for this business?” We come out of that, and we just don’t feel like stablecoins serve a purpose. It’s not necessarily a technology problem.
I think stablecoins tokenizing a dollar and creating a transparent financial system is useful and valuable. But convincing different stakeholders to use a stablecoin is a different conversation than I perhaps appreciated as much. There are certain business models where pushing a stablecoin to them won’t necessarily make that business more efficient.
Where I’m coming at it from, after doing this in many different sectors, is that crypto is a very narrow use case. Gambling is one that I don’t think anyone here can disagree with. It allows people in certain parts of the world to access the stock market more easily.
That’s interesting. There are always going to be stablecoins that sit on-chain because they can’t off-ramp. You can build nice businesses around that. The third category is catering to people who would rather get paid faster or get paid in dollars. Those are the 3 buckets. Everything else—
Hold on. Let me take this.
There are 2 buckets there. Gambling is the thing that ripped the hardest because what crypto does is unlock value to be moved around the world seamlessly.
What did the internet do? It let information flow seamlessly. Where was information siloed? Pornography. So what rips the hardest when the internet gets created? Porn. What rips the hardest when crypto lets value flow? Gambling.
Yeah.
I think, Santi, you’re struggling with this because—I know you put on a big trade in some of these memory names, like SanDisk and Micron. You talked about this publicly, so I think it’s fine to share. You’re struggling with the idea that this is an easier, better bear market. I think it’s harder psychologically because there’s another asset class trading in a way that crypto tokens used to—ripping your face off.
I’m not coming at it from that. My reasoning is from going in and assessing these companies. We’ve talked to a very large, representative set of private companies.
You’re talking to unsophisticated, small companies that aren’t the actual people.
Not small at all.
No, no, no. You’re totally missing the point here. This is the point that people miss in stablecoins all the time. I had this Twitter exchange yesterday with a guy from the Brookings Institution who didn’t even know that stablecoins were being settled directly with Visa. He was talking about how stablecoins aren’t used for payments.
There’s an education issue at most actual companies that do payments.
But I am talking to a company right now that is basically doing an AP bill-pay process—think Ramp plus Bill.com in one—growing 40% month over month and doing almost $80–$90 million of run-rate revenue today. They are serving some of the biggest corporates in the world, mostly in Latin America. The entirety of the infrastructure of that business is built on top of stablecoins. Nobody knows. Their customers do not know.
But the business has been able to grow its capabilities, bring its costs down, grow margins, and enter more markets because it is using stablecoins. Rain is going to have another 20% growth month. They announced a deal yesterday with [?], which is basically a global marketplace, right? The amount of stablecoins being settled directly with Visa is doubling every few months. The number of companies coming out and saying, “Hey, we’re using this,” is picking up a tremendous amount of steam. So they announced their stablecoin yesterday, right?
I think what people miss is that, because stablecoins are not—people think it has to be a peer-to-peer transfer. There’s a big education problem, right? It doesn’t have to be a peer-to-peer transfer. What it is doing is enabling a brand-new settlement layer. This is why Jack Zhang from Airwallex, who was the biggest anti-stablecoin person on Twitter for a little bit, is now tweeting about maybe building his own layer 1, right? They see, at least for cross-border payments, a huge, huge reason for this to exist and to rebuild.
Not to disagree with any of that. As I said earlier, certain payouts—if you’re doing creator payouts, for example—make sense. There are nice use cases. My point is the nuance: how big and how concentrated is this? I’m actually more bullish on Tether continuing to be a regulatory arb, particularly for remittances. Tether continues to be a really good business. Hyperliquid continues to be a good business, but it’s much narrower than I think people have historically expected.
If you’ve been in crypto for many years, you’re sort of expecting this vast transformation, and I just don’t think it’s going to be the case. I continue to be very bullish, but much more concentrated and focused on probably fewer than 5 names in crypto.
I do agree with that, and Rob might disagree, but I think in 2018 the reason that was a fun bear market for some—for me, it was very hard in some ways—was that, Santi, I think you remember this well, we thought the whole world would be cryptoified. Every game would become a crypto game, and the next Uber was going to start as a crypto company with crypto incentives. That looks like, you’re right, it’s not going to play out.
But I think what we got wrong is that global capital markets are so much bigger than anyone realizes—orders of magnitude bigger. If all we do is move global capital markets onto public blockchains, it’s an insane opportunity, even if you stop there.
Even if you stop there. I mean, there’s nothing—again, I think we shy away from saying this—this is gambling. Gambling is a massive industry. By the way, I think we talked about this many times. You remember that pod with Meltem, where I agreed with her take that everything is just going to hyperfinancialize.
The TAM just continues to grow because you’re bringing online people who historically have only been able to invest in real estate or government bonds in their particular country. Now anyone with an internet connection can go long a prediction market on SpaceX pre-IPO. That TAM is perhaps, to your point, grossly underestimated in terms of how big it’s going to be.
But it’s very concentrated. I continue to be bullish on something like Hyperliquid and Tether, less so on some of the other stuff that we expect.
This is a different point, I think, because you made one point in the beginning, which was, “I’m kind of bearish on capital markets coming on-chain and stablecoins, and I think it’s very narrow.” Then you made a point that there are only 5 companies or protocols that you care about. I think those are different points.
There’s a value-accrual debate that we can absolutely have, and who wins in these markets. I think that’s a completely fair debate. But Ondo just announced that they’re launching tokenized RWAs, synthetics, and markets with OKX, right? I guarantee you—and I know this for a fact—that at some point we’re going to get perpetuals guidance from the CFTC here in the U.S. The expectation is that this will include tokenized synthetics and things that are actually settling on-chain, right?
If that happens and we start to see an expansion of capital markets, an expansion of operating hours and trading hours, and especially the exporting of capital markets globally and the exporting of the U.S. dollar globally, those markets will also need to be settled in stablecoins. Then we’re talking about a bunch of software companies. Ramp is talking about going global with stablecoins, right?
If it’s a settlement layer that is being upended, and we have brand-new settlement layers that are all happening on-chain, I think it is shortsighted to say that, over time, that doesn’t accrue value to other things in the ecosystem. In the near term, one, that’s a huge market. Two, there will be a ton of value built doing that, and a ton of value that accrues to maybe only a handful of names.
With stablecoin transfers? What’s that? Just stablecoin transfers—you think that’s a massive market? Because that, in my mind, is pretty much at cost. You’re not making that much money with stablecoin transfers.
Well, it’s a scale game for sure.
We talked about Tempo. Tempo is not going to make that much money just on stablecoin transfers.
I think the thing to think about with stablecoin transfers is similar to the way you think about fintech more broadly, right? Stripe started out as essentially just a gateway that connected e-commerce businesses and banks, essentially. Now what is Stripe today? Their credit, their accounting, a bunch of virtual accounts and cards, and so on.
If you look at the history of fintech, people have taken these wedge products and built very big businesses around them that accrue value in a bunch of different ways. The company I was talking about earlier in Latin America started out as basically just a payments-transfer business, but now they’ve started winning customers, upselling customers, and increasing margins on the expense-management software that they have.
If you take these wedge products of just having better transfers and better settlement, you can build really big businesses on top of that as the world gets bigger and more value gets accrued. We’re seeing this in some of our companies. I think it’s a short-term-versus-long-term perspective here, and I get it—you’re more of a trader than I am.
Not really. I mean, not really. I think coming into crypto as an investor, you have to eventually be a trader and manage risk because it’s liquid venture.
Okay, honest question: When we talk a lot about tokenization and all these corporates doing stuff, I had this discussion earlier with my team. What is the best concrete, tangible application of stablecoins on a real business? And then what is it on the RWA side? There are investments and headlines and whatnot, but what’s a case study that you show a regulator or a skeptic—crypto, I mean?
When I tell people about the Rain story, their ears perk up every time. They’re very interested in what’s happening there because card economics have been a topic in D.C. for a long time, and they’ve been a topic for regulators globally. You’ve seen it in Europe.
The ability to do cross-border issuance with cards, then settle the same day and eventually in real time with Visa, without having to do it inside an issuer bank—or you can do it with a local issuer bank and a cross-border BIN—and do those 2 things together, improve the economics, and settle in more real time on the issuer side, people say, “Oh, that’s super interesting.”
Then the wheels start turning: What does that mean for the businesses I can build? How can I monetize my consumers and give less economics back to a bank? When you start to think about more people existing in this type of ecosystem, and maybe having stablecoins without knowing they have stablecoins, it looks like U.S. dollars.
On the acquiring side, the merchant side, you start to see a world where you can grow alternative networks, grow alternative businesses, and add more economics both to the merchant and to the end customer. When I talk about Rain to regulators, they’re very interested. When I talk about Rain to large corporates, whether they’re technology companies or financial services companies, their minds are blown because it’s a completely different way to think about global payments from different jurisdictions and in different currencies.
I’m obviously very bullish. I think I’ve said this before: Rain is the biggest investment we’ve ever made as a firm. I take that with a grain of salt, but the reason I’m bullish is that I have these discussions with everybody, from regulators to large corporates to global cross-border payments companies. I talk about the enablement that we’re seeing, and everybody goes, “Oh, holy shit.”
I get it. I’m a payments nerd, and that’s less exciting if you’re punting, for whatever purpose, on Hyperliquid. But I expect that there are going to be global banks that serve SMBs and retail consumers far better than the banks that have ever existed, and it’s going to happen because of stablecoins.
Yeah. I don’t know if you want to respond to that.
No, I don’t. I agree with Rob. Everything he said, I agree with. I think the problem is that maybe the thesis is a little boring if you’re not a payments and banking investor. The industry is changing. It’s changing.
Maybe that point brings it all the way full circle: I think a lot of the timeline right now and the bearishness is because it’s boring, and it doesn’t necessarily accrue value to the tokens that people want to see go up or to the things that retail is interested in.
2. Robinhood Launches AI Agents
Hyperliquid is pretty interesting. Very exciting, people.
I agree. Okay, so let’s start with Hyperliquid. There was something that happened this week. I don’t know if either of you guys are investors in Liquid Trading, but they launched this thing called Co-Invest, which is an integration that lets users trade Hyperliquid perps directly inside ChatGPT and Claude.
It’s legal in all 50 U.S. states via CFTC no-action relief, and it’s very interesting. What’s also happening right now is that I think Robinhood just announced publicly, but basically every single brokerage and exchange is working on AI agents.
Robinhood announced its AI agent this week as well. It allows you to launch an agent to trade on its platform. You set it, forget it, and watch your portfolio move. It can make you money, and it can lose you money. You tell it what you want: “I want to be risk-on. I’m very bullish on memory stocks. I want to lever up,” or, “I want to be safe.” It’ll actually put all the trades in for you.
My question to you guys is: How much do Claude and ChatGPT actually gobble up here? If you think back 15 years, I don’t think anyone was able to predict how much social media would gobble up. It gobbled up all the media companies, right? It was very tough to predict that.
There’s an idea here that ChatGPT and Claude could actually gobble up all of the fintech platforms. Will we have SoFi in 10 years? Will we have Robinhood in 10 years, or do you just bank—or do you just have all your money with Claude and ChatGPT? I don’t know if you guys have thought through that, but this AI-agent thing made me think about it.
Historically, there have been phases of bundling and unbundling with software. There’s a phase of bundling, and then unbundling. It’s a good question. I haven’t thought about it deeply, but more and more people are using Claude and ChatGPT, and there are all these wrappers built on top—autopilot, all using AI to serve the consumer better.
One thing I’ve appreciated is that people don’t like to think about money. They don’t constantly think about switching bank accounts, and I think the trust piece is something that technology doesn’t change overnight for a big part of the market.
Robinhood is still a very small fraction of the overall population in the U.S., and it has a sticky customer base. People don’t leave Schwab, Bank of America, or Merrill Lynch. Maybe the generational wealth transfer—boomers inheriting money to younger generations—is the bull case for suggesting that, but I don’t know.
I don’t know if it’s going to fully happen. There hasn’t been historical precedent to suggest that it’s going to be the super app of super apps. Maybe Kakao and WeChat are the things that give me confidence that there could be more bundling than unbundling here. But, anecdotally, would you do it?
No, I wouldn’t. I mean, I would use—
Security? I don’t know. I just feel like I don’t want the app where I look for memes. I don’t want to trade through Twitter. I don’t open Twitter and think about wanting to move money around.
Church and state.
Yeah, church and state. That’s how I think about it. But I could also see myself being completely wrong here.
I think younger generations have a very different relationship with money. They don’t understand what a bank account is. They’re fully digitally native.
The Hyperliquid guys obviously came out and said, “This is a no-action letter.” Because there’s a no-action letter for a different company, they’re able to do this as well. Good luck.
Yeah, godspeed. I was thinking the same thing.
Yeah. Having spent time with the regulators and knowing other things that are going on, good luck.
Regardless of that, I think the Robinhood thing is really interesting. Instead of building their own agent and saying, “Use our agent to do this thing,” they’re saying, “Bring your OpenClaw, bring your Hermes agent, bring your Anthropic-managed agent, plug it in, and we’ll give it a siloed wallet or a siloed account.” You can preset spending limits, work with it, and see what happens. If you lose money, you lose money, but this is your agent. You can do your own agentic trading and training on this thing.
I think it’s pretty clear to me that this will exist. I was talking to the CEO of one of the large crypto exchanges around 3 months ago, and he told me he was already doing the same exact thing in some of his personal accounts. He had taken an OpenClaw agent, trained it on a bunch of historical trading data, and given it its own accounts on his exchange. To his credit, he told me that they were losing a bunch of money, but that’s obviously something that’s going to happen.
I think that’s super interesting. I don’t actually think people are going to go through a front end and use a bunch of natural language to do a lot of trading. I think those are 2 different points. Is there retail trading through a front end using natural language? And is there a lot of agentic, AI- and ML-powered trading that happens on an exchange?
The latter is inevitable. There’s no doubt that we’ve been moving toward systematic trading more and more over the years as machine learning has grown. There’s no reason to think that will stop. I expect that it’s gone from institutional trading into retail, but I don’t expect there to be a lot of natural language, at least not through other front ends.
Maybe I’ll go to Robinhood and say, “Can you buy me this perp or this option?” and they’ll do it. But I think people will still want their finance to exist in a finance or fintech app, and they’ll want Anthropic and OpenAI to exist elsewhere.
We’ve looked at a lot of startups that have tried to attack this natural-language trading opportunity, where they do the smart routing or whatever. Maybe it’s just early, but the products are pretty terrible at the moment.
I think this goes back to the tracker, like the robo-advisor, that would follow 13F filings and copy expert portfolios. The argument is that agents will be able to do that more efficiently in DeFi, plugging directly into Hyperliquid and doing it in real time.
I go back to that example: war in Iran broke out over the weekend, your agent spun up, read it in the news, went directly to Hyperliquid, and went long oil. I think there’s an interesting business opportunity around setting the right guardrails in place to make sure that your agent is safe. People will not do this on their own.
And I think they will buy a product that they have really good confidence around: okay, this is secure, it’s a good setup. There are different parameter rules, in the same way that when you set up a brokerage account, they’ll ask you what your risk profile is and how comfortable you are with X or Y. I think we’ll see a bunch of those businesses. How enduring and durable? TBD, because I think Anthropic and OpenAI just have a nature of crushing whatever gets traction.
Um.
Yeah, I mean, they just acquired a consulting firm, basically, that was doing integration work for AI. I think Yano’s question was different, which was, okay, is fintech for retail going to get displaced by OpenAI and ChatGPT? That’s one thing, and maybe OpenAI, ChatGPT, and Anthropic will launch their own fintech or whatever.
I don’t think the opportunity set disappears. They still continue to have trade-offs in terms of what the best ROI is on their time and their compute. I think the enterprise application side right now is so big that, for a while, we’ll see them focus on that opportunity.
But I don’t disagree with you that they’re obviously going to get into more infrastructure for trading, and you’re going to see all of the systematic trading shops integrate new models and do some of their own training.
Think back 30 years ago. If everyone’s using a Bloomberg Terminal to find information so they can make better trades, one day Bloomberg wakes up and goes, “Why don’t we just let people trade through the platform?” I think eventually you could see that.
The interesting thing will be whether they acquire a banking license and do it in a traditional way, or whether they integrate wallets and stablecoins and Hyperliquid builder codes, or whatever it looks like.
3. Morpho Midnight
Speaking of other cool DeFi stuff, Morpho launched their “Midnight” white paper [?]. It was a non-custodial protocol for fixed-rate, fixed-term credit markets. Santi, I’m sure you’ve invested in about 17 of these things already.
You guys did an entire season on fixed rates.
We did a whole season on interest rates in crypto. We did interest-rate swaps. The market was about $700 million for this at the time.
That podcast no longer exists because that podcast bored the hell out of people.
We had to pivot the direction a little bit.
What are they doing now, and what’s different?
Okay, let me tee it up. Fixed rates are this missing primitive. Fixed rates in capital markets exist, and they allow for a bunch of things. Fixed rates in DeFi do not really exist yet.
If you look at Compound or even Morpho Blue today, the rates are tied directly to pool utilization. What happens is, if you have inflows or outflows that spike, that changes the borrowing cost. If you’re a borrower, you’re faced with what’s called interest-rate risk. You never really know your true all-in cost over time.
The result is that there’s a lot of whale borrowing, retail borrowing, and trader borrowing. But if you’re anyone else—if you’re a business—it’s actually quite hard to plan anything long term. You end up either not taking out the loan, or your cost is too high because you have to cover that interest-rate risk.
Fixed rates remove that risk entirely. They allow borrowers to lock in an exact cost for a defined term, and lenders get guaranteed terms without having to babysit their utilization.
If this doesn’t make sense, or you’re wondering why this matters, many people listening to the podcast have probably dealt with this. It’s called a 30-year mortgage. Thirty-year mortgages are a floating rate, and then they use interest-rate swaps to basically fix your mortgage rate. That allows you to have a fixed rate for businesses borrowing.
Blockworks, if we were borrowing, would not take a floating loan because it would be too risky for Blockworks. But if we could take an on-chain fixed loan, that would allow our CFO and our finance team to actually plan this out.
Where this gets really interesting in DeFi is that there are finally protocols making real revenue. Imagine Uniswap—I don’t know Uniswap’s revenue off the top of my head, so I’ll make it up—making $100 million. If Uniswap wants to take out a loan, it’s now a small business or startup making $100 million. It should be able to take out a loan, but if it’s a floating loan, it’s very hard to actually do that and plan for it. If it’s a fixed loan, you unlock real business and corporate borrowing.
You unlock fixed-rate mortgages, business and corporate borrowing, institutional and RWA lending. This has been tried many, many times, and many times it’s failed. The demand side probably wasn’t there, but this could be the time when that actually changes and works.
I’d love to hear—I was kind of joking, but I would guess you’ve invested in many of these.
A lot of them, and I think none of them really worked. To tell you the truth, I don’t have a full postmortem. I’ll have to refresh my notes, but you’re right: it’s very hard to borrow.
If you were to take a loan on Maker, for instance, or Sky, you’re subject to governance risk. If you look at the rates, Maker or Sky governance sets the benchmark. That has moved dramatically in the span of a couple of months, which is the equivalent of the Fed hiking rates from 5% to 15%. That just becomes really difficult.
A lot of institutions have shied away from borrowing on-chain for this reason. Mortgages in the U.S. are kind of the best product. A 30-year fixed-rate mortgage is an amazing product that doesn’t exist anywhere else in the world. Anyone who took advantage of that when rates were zero is sitting pretty.
But if you go to Europe, you don’t have that. You then have to think about other ways to hedge your risk. There are swaps for this reason. If you’re a business, you can use swaps to manage your exposure, but it’s been tough to do on-chain.
There were also interest-rate swaps that were tried. I think Yield tried to do this, but none of them really got off the ground. I don’t have a good answer as to why they didn’t gain traction.
Rob, did you guys invest in any of this stuff? This is like the 2018–2020 era.
I have to think back. Off the top of my head, I don’t think so. But we definitely did a lot of smaller checks in that 2019–2020 era, and I’m probably not remembering all of them. Definitely nothing in the last 4 or 5 years that has been notable.
We’re also—not VC investors, but at times we’ve owned a—well, we’re not investors in Morpho. I do think this is important. I think this is positive. I don’t have a ton more to say about it other than—
I quickly refreshed my memory, courtesy of my robot here—my quant. The reason why it had been tough historically is that there was no universal benchmark rate. People would try to do this with the ETH staking yield as the central bank, and then Maker, and then Compound, but there’s no universal benchmark like the U.S. Treasury.
The other thing is, I think the capital markets on-chain have been extremely short—extremely short-dated, like short duration. These are like short-duration Treasuries. There hasn’t been—and I think this is something that I’m excited about—which is bringing quality credit on-chain.
If you bring quality credit on-chain, I think you fix this issue, because if you’re trying to do it with yield farming and looping strategies, all of that is incredibly short. There’s no real need for fixed rates. You’re not buying a house for 30 years; you want to live there for the rest of your life.
Now, obviously, I’m biased because that’s an opportunity that I’m aggressively going towards. The interesting thing, talking about the Framework guys, is that I think they should come on and talk about what’s going on with Better. Better is a mortgage originator, and I think they tapped Sky for a pretty sizable facility.
I don’t know the particulars of how that agreement was structured, but when I pinged them, I was like, “Hey, was this a bespoke, brokered, structured deal where Maker guarantees a rate?” Because I said, “What happens if Sky governance just decides on a whim to change rates from 3% to 10%? Better is, at that point, pretty exposed.” He didn’t tell me, but I would love to know how those agreements were structured.
I think if you’re the CFO of Better, you probably would have wanted a strong guarantee that there’s a cap to the rate.
I’ll tell you, that tells you.
So, there you go. In summary, I think short duration has been the name of the game on-chain. There’s no reference rate and there’s no quality, but if we fix that, then you do create a really nice business out of it.
I think the way Morpho is doing it is—I have to say, I haven’t actually read the white paper; I was just reading some tweets—every market has a single fixed maturity date. Let’s call it 3 months from now. Instead of lending USDC and earning interest, you buy what are called credit units, and instead of borrowing USDC and paying variable interest, you sell debt units.
At maturity—let’s call it 3 months—1 credit unit is redeemable for exactly 1 token of USDC. The fixed rate is locked at the moment of trade.
So this is Pendle.
Yeah, it’s Pendle.
Pendle. Exactly.
Yeah, but they go out to a maximum of 1 year. It’s not like a 5- or 10-year maturity.
Yeah. To be fair to Pendle, that’s another one where that business, I think, started, if I remember right, as this grander vision around what could happen around all these different markets for off-chain hedging, et cetera. It ended up growing a lot because of what we were talking about earlier: the yield farming in Ethena, basically.
Ethena was by far and away the biggest exposure there for a while. Then we had all these interest-bearing stablecoins and tokens, et cetera. I think some of this topic, or this conversation, goes back to what we think people will be doing on-chain.
Will it continue to be this, as Santi keeps saying, gambling, trading, and yield farming, et cetera? Or will there be more regular economic activity coming on-chain, with people wanting to use these on-chain primitives to manage risk?
Yeah. What’s different this time around is that there’s very little yield farming, which is super short, block-centric—to the minute. The other interesting thing I’m curious about, Rob or Yano, in your conversations is: Is the reason there hasn’t been fixed rate beyond yield-farming, short-duration stuff—no reference rate, whatever we’re investing in? We’re investors in Notional, the other protocol Element Finance. There’s just a literally graveyard out there.
Is smart-contract risk the reason why everything’s short-duration? Is there nervousness in the market around smart-contract risk?
I mean, I don’t think that’s the reason initially. There’s some conversation around what the curve looks like in terms of demand and pricing. This actually maybe goes into a different topic: how concerned everybody is right now about DeFi and smart-contract risk.
That wasn’t nearly on top of everybody’s mind in the way it is today, at least until the last few months. If you go back a year from now, none of these things had worked, but a bunch of people had still tried to make them work, and it was still a topic of conversation. I mean, Yano had a full podcast about it. Crazy.
4. Is DeFi Uninvestable?
You’ve got to pull up the tweet that I retweeted from the founder of OpenZeppelin. Do you guys see this?
Yeah, but he isn’t the founder. He was the founder of OpenZeppelin. He left in 2019. These are smart guys. I mean, they’re on top of it.
He said, “Nothing in DeFi is safe.”
“Nothing in DeFi is safe.” He said, “I’ve advised family, friends, and myself included to put zero capital on-chain.” Actually, he didn’t say “at the moment”; he just said, “Full stop.”
Interesting. I’ve privately continued to talk to credit funds and other folks who have historically been very active in DeFi, and there’s a very noticeable focus on questions like, “What’s the state of stablecoins on Aave right now? Is the Kelp situation resolved?”
Kelp is not fully resolved, but it’s mostly resolved. They’re making progress.
It’s definitely not resolved. There’s all this legal stuff happening in the background.
But everyone on Aave was made whole, basically.
From the user’s perspective, it is solved.
Yeah, I was just going to say I think there is this point around everything in DeFi being unsafe. We kind of had this discussion right after Kelp on this pod. A lot of people snapped back at him and were like, “Oh, we’re more battle-tested than ever.”
Again, most of the hacks—almost all of them—are not smart-contract hacks. They actually have something to do with a vulnerability through a specific user, a key holder, a signer, et cetera. I would say there’s this thing happening where it’s like, okay, it’s not smart-contract hacks; it’s not the software itself, and you can improve the software with AI in the same way that AI can find vulnerabilities in the software.
There’s probably this point in time where, again, the attackers have a foothold because while everyone is rewriting code or making code more robust, they’re able to go and attack you. They only have to find 1 vulnerability. But I still continue to think that the biggest problem is OPSEC. It’s people specifically having their security perimeter infiltrated, not through smart contracts.
I think that is primarily the issue. But what is happening now is that once someone is in your security perimeter, the ability to do what they need to do to find the vulnerability and then find the way to extract capital is happening faster than ever. That is specifically AI-related.
I do think that we’re going to continue to see more hacks across not just crypto but all software globally for a period of time, and then eventually hacks will basically go away altogether. It just takes time to rewrite code.
Yeah. I agree that you’re going to have a tumultuous period where you catch up, but the problem is that the lack of reversibility on-chain makes it much harder. My perspective was: wait it out. Wait it out for 6 months. You don’t want to be there when this happens.
You’re better off being in a position where—hey, look, one thing I will say is that not every protocol and not every smart contract is created equal. But the social-engineering stuff is tough, because most of these protocols are upgradable and there are admin keys. Rightfully so—you want to have that—but it’s quite tough.
I think it was a CISO that commented, like, “Have endpoint security. Make sure that you’re monitoring your stuff if this happens, because you might be infiltrated right now and not know.” They’re just lurking there in the background, and they’ll hit you in 6 months. No, I’m serious. This is what happened with Drift 6 months ago.
And if you’re using Claude Code and don’t have separate devices, at least have something constantly monitoring your device to see if there’s anything running in the background that shouldn’t be running. It is table stakes. It literally costs $100 a month in the enterprise solution, like CrowdStrike or whatever.
Get on it and have peace of mind. Seriously, there’s really no excuse if you’re in this industry, because you’re also putting a lot of other people at risk. Have a safe word with people to prove humanity. The tricky part is, when you think about OPSEC, if you’re vibe-coding, there’s a higher surface area of just being hit. That’s what makes it all the more problematic.
5. Takeaways From DC
Yeah. Maybe one last topic here before we wrap. Rob, you were just in D.C. Santi touched on this at the beginning, but I’d love to get any takeaways from the D.C. trip.
Yeah. Usually when I’m in D.C., I see some combination of members of the House or members of Congress generally, or their staffs, and then regulators, which is again what I did yesterday.
I think there are a couple of topics. One is CLARITY, and the bullishness around CLARITY is pretty positive. When I was last in D.C., you could definitely feel the bullishness among the Republican side of the Hill, and you could definitely feel it among maybe a few of the very big crypto supporters on the Democratic side. But now you're starting to see a little bit more bullishness among people who maybe hadn't put a bunch of their weight behind CLARITY or GENIUS before.
There's an expectation that we'll get a vote, maybe in July. The industry keeps saying, “Oh, June.” I don't think that's realistic; I think that's too quick. At the floor vote, there'll be a couple of things that are top of mind. One of them is the ABA, the American Bankers Association. The bankers are continuing to fight against this bill. Even though Tillis took it out of the Banking Committee, we'll continue to see them lobby hard against the bill.
People have talked about the ethics language that the Democrats are going to require to put into this bill around the ability to profit off crypto and crypto businesses, which the White House is going to have to get comfortable with. Those are probably the major 2 things. There is still going to be a lot of debate on the floor.
Typically, the Banking Committee and the Agriculture Committee would come together and reconcile before it went to the floor, but there have been some holdups in Agriculture. So you're basically just going to get the Banking version before Agriculture is able to reconcile as well.
There's also a lot of conversation right now around this thing called BRCA in the CLARITY Act, which basically gives developers protections for developing open-source code that maybe ends up getting used for nefarious reasons. But this is just open-source code that allows you to swap, send, or lend, etc. I think there are carve-outs for, okay, you're not a money transmitter if you're a validator, or if you're just a front end, or if you're just a smart contract. If you create any of those things, you have developer protections because of it.
There's going to continue to be debate around that as well and the exact language, but that all seems solvable to me. In the past, when Polymarket showed that CLARITY had a 50–55% chance, I was saying, “Oh, that's too high.” I now think the likelihood is higher than 55%. I think this is now underpricing it a little bit, and that's exciting. I could talk about some stuff on the regulatory side, but I'll stop there on CLARITY.
Nice. That's great to hear: 55% underpriced.
Yeah, the lowest was around 36%. Now it's up to 56%, so that's encouraging.
What about on the regulatory side, bro?
On the regulatory side, listen, I think there's a lot coming. Chair Selig from the CFTC talked specifically in March about guidance related to perps. Everyone in D.C. still expects there to be perp guidance. I still expect that to be constructive around how certain types of businesses can theoretically have offshore and onshore entities, the way Polymarket does today, and how they can fragment liquidity and do business in a regulated manner and a nonregulated manner.
I think that's going to be super interesting, and I expect both perps and prediction markets to continue to dominate a lot of the conversation on the regulatory side. Today, I think it's a lot of perps for 24/7 markets, but I expect that—not in the near term, but in the medium to long term—to also become: How should we be thinking about perps for real-world assets, and eventually those markets also becoming 24/7 on the spot side?
It's all very positive in my mind. Exactly the way we started this podcast, a lot of positive things are happening, and yet the price action hasn't made people as excited as you'd think they'd be.
6. Microstrategy & Robostrategy NAV Premiums
15 years, man. 15 years. That's what the Nasdaq took to recover from all-time highs. We can't end this podcast, Jason, without—we're going to end it on a high note with content—but Saylor, we've got to talk about Saylor this week before we drop. He probably had one of the most epic tweets that I saw circling around, which was, “This week, we're not buying Bitcoin; we're buying bonds.”
And then you saw a massive, billion-dollar trade of IBIT 1 or 2 days ago. What's going on in Bitcoin land? We don't typically talk too much about Bitcoin, but Bitcoin carries a lot of weight. What do you read there?
I haven't been following Saylor land, actually. He wants to do a podcast, so we'll have—I think we'll go straight to the man. We'll get Saylor on the show.
Yeah, I mean, a lot of the conversation you heard, or at least I heard, for a long time was, “Is Saylor just going to keep buying forever? Is he going to continue to use Strategy to be able to buy and accumulate more Bitcoin? Saylor's taking us to 200 grand.”
He was almost Bitcoin Jesus to a lot of people. When he turns around and says, “Actually, I'm going to sell Bitcoin to go and buy bonds,” I think that shook the market a little bit.
Yeah, a little bit.
It's like Jesus saying you can't walk on water, man. [laughter]
It's a pretty big detraction from the gospel.
If you go to their website right now, they have all their data on their dashboard. They took their U.S. dollar dividend coverage down to 6 months. Are they going to have to sell more Bitcoin to do that? It seems likely.
I keep coming back to the fact that all financial engineering, whether it's a DAT or anything, eventually blows up or unwinds, and it usually doesn't unwind easily. I will credit what MicroStrategy has done, but I've always said it: if you want exposure to Bitcoin, just buy the goddamn thing. Buy IBIT, at least.
Clever financial engineering, whether it's this or that or the NAV, always compresses, and financial engineering usually just does not persist forever.
Strategy's NAV is at 1.2 now.
Yeah.
Yeah. I mean, speaking of NAV, should we talk about RoboStrategy? What an interesting one that one is.
What's the NAV? What's the premium to NAV there?
It's, I think, close to 4× at the moment.
4×. Yeah.
Yeah.
I think it's maybe even higher than that when you account for the 25% dilution that the advisory team made there.
Yeah, again, this is a bit different in the sense that it doesn't justify the premium to NAV. But these are all private marks, and it goes back to the fact that there's been just so much wealth creation in the private markets that people want to have exposure to venture deals.
This vehicle attempts to do that for robotics, which is a hot trade, but I haven't seen a reasonable, compelling argument as to why NAV should be there.
Again, I think this was something you flagged in the last podcast, Jason, which is that you're better off buying a synthetic. Someone should just create a synthetic on Apptronik or Figure, or pick whatever robotics private company—or any private company—that you like, and someone creates a synthetic on that and just tracks the latest recorded trade or something like that.
You'd have to trust whoever is doing that and pricing that, but that to me feels like it's just a better way to get exposure to things in the private markets. The good thing is crypto is really good at finding liquidity for a tail end of assets. So a synthetic is, in my opinion, just better for this type of stuff.
7. Content of The Week
Yeah. Content of the week. What do we got? Santi, kick us off.
No, man. I can't go first. I've been talking too much. What do you think about that, Jason?
All right. For content of the week, last week I talked about Obsession, which is the movie that I think I flagged. The movie was really good, and it's this trend that's happening in horror right now where people are going from YouTube creators to getting their own feature films. It was made for $1 million, and it's going to do over $100 million, which is amazing.
Maybe we need an Empire pod entirely about movie financing, so let's think about that one next.
Since you mentioned that, I see it everywhere now. It's all over the place.
Yeah, it's everywhere.
Yeah. Yeah.
So the movie was really good, and it's this trend that's happening in horror right now where people are going from YouTube creators to getting their own feature films. It was made for $1 million, and it's going to do over $100 million, which is amazing. Maybe we need an Empire pod entirely about movie financing, so let's think about that one next.
Bring it on-chain, folks. There is some stuff through Republic, but another one—literally the exact same YouTube creator-to-feature-film story—called The Backrooms came out last week and is also awesome. I'm just telling people: horror. We're having a horror renaissance right now. It's a great time. See the movies and make money.
Horror? Oh, I thought we were talking about crypto. There's plenty of horror there, too. [laughter] Wait, Rob. Okay, I don't have a content of the week.
I just want to ask: did you guys see The Devil Wears Prada?
The second one? Yeah, I did see it. I talked about it on this pod.
Oh yeah, you went and did your whole routine and got a Scotch—or is it good?
You know, it was like a lot of these nostalgic remakes that happen a decade-plus later. They hit a bunch of the main themes so that you feel good about what happened before, and you're seeing the same things, but there's no real new character development, et cetera. So it's pretty surface-level, but I enjoyed myself.
I saw it on a Friday night with my wife and had 2 Old Fashioneds. What's it on Rotten Tomatoes? Is it like 48?
Oh, let's see. I'm not a big Rotten Tomatoes guy because the problem with Rotten Tomatoes is that people just say yay or nay. Everyone can say, "Oh, that was mediocre," and I have 100%.
You really have to get on—I'm blanking on the name of the app right now—but there's this really good app where people actually review all of their movies. I'll get you the name of it.
Yeah. Yo, what's your content of the week, man? Will, our producer: Letterboxd.
Letterboxd shout-out.
I was like, "I know the app." Yeah, let's go.
Best contribution to this pod. Will, you take—[laughter]
Out. I have a Letterboxd too. I should have known, but I'm not good at it. If anyone finds me, you can see all my reviews, so you can see what I really like.
Letterboxd, baby. Yeah, mine is I'm watching the show about Rupert Murdoch and the Murdoch family on Netflix. Very interesting. A media mogul, a titan. It's good.
All right, folks. Appreciate you guys listening. We got a good episode coming out on Monday. Sorry we missed this week. We've been going back and forth debating this new update to Empire: guests mostly in person, high quality over quantity. But we don't want to be skipping episodes either. So, sorry for missing this week. We got a good one coming out on Monday. We'll see you then.
What's the episode on?
Tease it, baby. You have to force people to open the app.
Listen, no, it's with the Electric Capital guys, Avichal. I think Avichal is bringing a special guest on too, one of his partners.
Cool.
Awesome, guys. All right. Well, have a good weekend. Thanks for coming.
All right, folks. Have a good weekend.