We look at a giant in any industry and see a failure of antitrust law. That view is shallow. Midnight brainrot, maybe, but look at it like gravity, functioning perfectly.

A massive system creates an inevitable pull. In physics, matter attracts matter. In markets, volume attracts volume. This is why tearing down a monopoly by force rarely works. You can partition the entity and declare the market free, but you cannot legislate away the physics. The broken fragments still feel the original pull. They float in the same space under identical economic pressures, destined to fuse back together.

How do you destroy a black hole? The gravity remains. Yeah, I just watched Interstellar for the 9th time.

Standard Oil controlled 91% of US oil refining by 1904. The government broke it into 34 companies in 1911. Within a decade, several of those pieces, such as Standard Oil of New Jersey and Standard Oil of New York, had grown back into giants on their own.

Exxon and Mobil, two of the biggest companies in the world, a century later, both trace directly back to that breakup.

Sometimes breaking a monopoly just creates the conditions for new ones to grow in its place.

Monopolies in finance and infrastructure are notoriously hard to dent, let alone end. AT&T’s 1984 breakup split the company into seven regional providers. By 2005, AT&T had reassembled most of itself through acquisitions. ‘In 2005, one of the Baby Bells, SBC, absorbed AT&T Corp itself, taking back the brand. The economics had never changed. High startup costs and user networks kept the original monopoly intact.

On May 27, 2026, the SEC approved Paxos Securities Settlement Company as a clearing agency. DTCC, through its NSCC and FICC subsidiaries, has been the only securities settlement utility in the United States since the 1970s. The first blockchain-native firm ever approved to provide clearing and settlement services as a central securities depository in the US.

Before we say “monopoly broken.” Now that I’ve read Tim Wu’s The Age of Extraction, I believe that it’s rarely true. How platforms that look unavoidable usually stay that way even after the legal barriers come down, because the barrier was not always the legal ones.

What does Paxos need to actually compete?

The fight comes down to netting. I’ll explain it like you are dumb.

You owe X $100. X owes Y $100. Y owes you $100. Without netting, three payments happen. $300 moves. When we apply netting, everyone realises it’s circular. The net effect is that nobody owes anybody anything, and nothing needs to move.

DTCC does this at the scale of the entire US stock market. Thousands of brokers and banks generate millions of trades per day. If Goldman Sachs buys 10,000 Apple shares and sells 9,000 shares on the same day, DTCC completely ignores those 19,000 individual movements. Goldman is a net buyer of 1,000 shares. Only that final balance settles, because… netting.

DTCC’s equities clearing arm processes 214 million broker-to-broker transactions daily, worth $2.08 trillion. DTCC’s own language is blunt about the alternative. Without netting, the company says, ‘trillions of dollars in cash and securities must move through the financial system on a continual basis throughout the trading day.’ Multilateral netting is what prevents that. DTCC’s own white paper calls it one of the ‘enormous benefits’ that any move toward faster settlement has to preserve.

That’s where atomic settlement, the thing Paxos is built on, runs into this.

Atomic settlement means every individual trade settles instantly and all at once. This eliminates counterparty risk because there is no delay between trade and settlement.

But if every trade settles instantly, there’s no chance to net anything. Each trade is a complete transaction, so full value changes hands every time. For example, if Goldman buys 10,000 Apple shares and sells 9,000 instantly, that counts as two transactions. 10,000 shares worth of capital moves one way, then 9,000 shares worth move back. This contrasts with DTCC, which nets the same activity down to 1,000 shares worth of actual movement.

The capital that must be set aside for the total amount, rather than the net amount, cannot be used for anything else. It can’t be lent, invested, or deployed in other ways. For a single trader making a few trades, this is just a minor issue. For a large broker-dealer processing thousands of trades each day with hundreds of counterparties, that idle capital requirement turns into a high cost, even if each individual trade settles more quickly.

Speed per trade versus capital efficiency across the whole book. Different optimisations, and for the big players in the market, the second one might matter more. That’s where they go back to the monopoly.

Paxos is well aware of this, too. Multilateral netting is part of their design. In fact, the company has built the platform to support it across any settlement cycle, from T+2 down to intraday. The constraint right now is scale.

Right now, Paxos only nets trades between two direct parties. Multi-party netting (what DTCC does) across the entire market will come later. The SEC’s approval caps initial operations at 10 participants. DTCC nets across thousands. Netting requires a massive volume to work. If only five banks trade with each other, they rarely owe each other matching amounts. Five hundred banks trading together create endless overlapping debts that cancel out perfectly. The big firms that need this efficiency are already hooked on DTCC’s massive network. They hardly have any reason to switch first.

For Paxos to build the netting efficiency that would attract those broker-dealers, it needs the broker-dealers. For broker-dealers to come, netting efficiency needs to already exist.

To make matters more complicated, PSSC’s registration is temporary. It lasts 18 months. Operations will start about 10 months after approval, which means the earliest live activity will be around Q1 2027. The target launch date for the DLT-based clearing agency is “March 2027 at the earliest.”

Can you expect DTCC to remain passive? They raised formal concerns in a comment letter regarding PSSC’s proposed netting plan and its overall recovery strategy. That’s what monopolies do.

Separately, DTCC partnered with the Stellar Development Foundation to put real-world assets on a public blockchain. They plan to tokenise US Treasuries and major stocks, including Russell 1000 equities, on Stellar’s network by the first half of 2027. As the incumbent with a net moat, DTCC is now also building blockchain infrastructure.

So what actually has to go right for Paxos?

Participant count needs to grow well past the initial 10, and it needs broker-dealers with real volume, not just pilot participants testing the rails. Multilateral netting needs to go from “planned” to “live” within that 18-month window, because a temporary registration that can’t demonstrate its core economic claim won’t be renewed. And Paxos needs all of this to happen while DTCC is simultaneously building its own blockchain rails on Stellar.

None of this is impossible. We will see if Paxos can actually break the cycle, or if DTCC’s gravity pulls them right into the centre.

—Thejaswini

Token Dispatch is a daily crypto newsletter handpicked and crafted with love by human bots. If you want to reach out to 200,000+ subscriber community of the Token Dispatch, you can explore the partnership opportunities with us 🙌

📩 Fill out this form to submit your details and book a meeting with us directly.

Disclaimer: This newsletter contains analysis and opinions of the author. Content is for informational purposes only, not financial advice. Trading crypto involves substantial risk - your capital is at risk. Do your own research.

I hear that phrase and imagine a relentless war of all against all, where the fiercest predator wins by crowding out everyone else.

But now I doubt that is how lasting ecosystems actually form. At the turn of the twentieth century, the naturalist and philosopher Peter Kropotkin challenged this view in his evolutionary essays on mutual aid. Kropotkin saw that species surviving severe climate shifts succeeded through sophisticated patterns of collective cooperation. Mutual support is infinitely more powerful than individual warfare in long-term evolutionary progress. The truly “fittest” are those that learn to cooperate and build stable, shared frameworks to handle an unpredictable environment.

The principle spans everywhere.

Every crypto project is figuring out what it actually is right now, and the protocols that survived are each making a defining call about what comes next.

Some development teams are retreating to original ideological principles like absolute censorship resistance and pure decentralisation. Others are integrating centralised overrides to maintain basic solvency. A few are focusing entirely on proprietary ecosystems, pulling their internal liquidity into isolated rollups.

ZKsync made a different call, a mutually aided one. It’s building for banks.

BCG projects the tokenised asset market moving onto blockchain rails, hitting $10-16 trillion by 2030. Every major bank is running pilots. Some have moved past pilots into live production. The infrastructure decisions being made right now will determine how that money moves and who controls the rails it moves through.

ZKsync is currently a serious public blockchain candidate for those rails. That’s why it matters even if you haven’t thought about L2s since the last bull run.

Why are banks even here? Why ZKsync specifically?

Deutsche Bank isn’t on the same ZKSync as regular crypto users. They’re on the Memento ZK Chain, a private, permissioned Layer 2 that Memento built using ZKsync’s Prividium stack.

Prividium is ZKsync’s institutional product. This offers private transactions, controlled access, built-in compliance tools, and final proof that everything settles on Ethereum. After testing five different blockchain ecosystems, Memento picked ZKsync. Fund deployment now takes 2-3 weeks instead of 2-3 months.

Banks love zero-knowledge tech because it proves a statement is true without revealing the private details behind it. A bank verifies a transaction without showing the public the names, the dollar amounts, or the assets involved. This private setup gives the bank total control over who sees what, providing privacy for corporate strategy and clarity for regulators. It fits how Wall Street actually operates.

Tradable has $1.7 billion in private credit sitting on ZKsync. Nearly 30 institutional positions targeting yields between 8% and 15.5%. And in October 2024, Buenos Aires quietly moved its entire digital identity system onto ZKsync Era. 3.6 million residents got cryptographically secured government credentials overnight. Local government can’t track that. The city became the first in the world to do this.

The global private credit market reached $3.5 trillion by late 2025. Tradable’s share is under 0.05% of that pie. Tokenised credit could grow into a meaningful share, or stay marginal, though the current situation says it is growing. Either way, the difference between the on-chain money today and the total available market cap is gigantic.

Think about the choice a corporate risk team faces here. They must decide between an isolated network that they completely control, a corporate group bound by contracts, or a public network managed by an online community.

JPMorgan is a private network under absolute internal control. It has run its own internal blockchain setup through its Kinexys unit since 2019. They handle repo trades and cross-border payments, as well as asset settlements, with partners such as BlackRock and Siemens. JPMorgan hosts the machines and manages the ledger while maintaining zero exposure to public online forums. Fees never change because a token holder voted, and system upgrades follow internal calendars. JPMorgan is the governance.

The catch is in their own numbers. Kinexys processes around $5 billion daily. JPMorgan’s payments division processes $10 trillion a day. Their own blockchain unit handles about 0.05% of their own payment flows, five years in. The bank that has the most control over its blockchain infrastructure also has the least adoption of it. Full control didn’t solve the hard part.

Then you have R3 Corda, which cleared $10 billion in tokenised real-world assets. It processes a million transactions daily. Corda functions as a consortium of over 200 financial institutions that built their rules upfront through contracts. When a feature changes, it changes because the members collectively signed off on it. The banks secured a seat at the table before the first transaction ever settled.

These platforms are ZKsync’s competitors. Yet ZKsync delivers a specific benefit that proprietary or consortium networks cannot replicate. Public verifiability without data exposure, paired with a settlement layer independent of any single corporation’s survival. If a single institution terminates its internal blockchain unit, the assets hosted on that network face immediate operational danger. Assets anchored via ZKsync settle to the Ethereum mainnet, which lacks a chief executive to order a shutdown. This separation remains the primary differentiator, and balancing it against governance exposure is a live debate.

Before ZKsync committed fully to the institutional strategy, it ran an incentive program called Ignite, paying DeFi protocols to stay active on the network. When the strategy shifted, they ended Ignite, citing a shift to enterprise. The activity followed the incentives.

Around the same time, ZKsync Lite, the original network from 2020, shut down permanently. Matter Labs had been signalling this since December 2025 and gave a confirmed date in late February. Funds remained claimable indefinitely, and none of the institutional deployments was on Lite anyway.

Aave, the massive DeFi lending protocol, voted to pull the plug on its ZKsync Era market. The motivation came down to raw numbers. ZKsync generated just $714 in fee revenue for Aave over a 30-day window. Compare that to Base, which brought in $300,000, or the Ethereum mainnet at $7.7 million. The voting forum concluded the chain lacked true product-market fit, proposing that any future deployment on any network must hit a $2 million annual revenue floor before Aave even considers launching there.

At the height of the retail layer-rollup boom, ZKsync Era regularly held hundreds of millions of dollars in locked DeFi value. Today, the total value locked across its public DeFi ecosystem is roughly $15 million. For context, top-tier retail networks maintain billions.

If this stalls, the ecosystem depends entirely on your thesis. Are L2S supposed to be consumer-crypto havens that banks are co-opting, or just corporate rails that happen to settle on Ethereum?

Matter Labs explained this to you earlier this year. CEO Alex Gluchowski released a roadmap that shifts the focus to building heavy-duty financial infrastructure for traditional markets.

Look at the product releases themselves. They rolled out Prividium to give banks an isolated space where transactions stay locked away from public view. Then they introduced the Bank Stack, partnering with entities such as the Cari Network to onboard regional banks that manage billions in traditional deposits. So, Aave’s departure did not shock the developers.

Talking about the Cari Network deployment. It was started by a former US Comptroller of the Currency, and they are planning a pilot next quarter with five regional banks. These banks hold over 600 billion dollars in total deposits.

If that pilot works out, nobody will care that retail DeFi apps left the chain. The transaction volume from banks at that scale replaces the loss. But if it fails, ZKsync is a good piece of tech with no users, running a few corporate experiments.

The v31 protocol upgrade, passed through ZKsync’s governance forum, went live in early May.

Under v31, every cross-chain call between ZKsync chains costs a flat 10 $ZK. That number was set by a DAO vote. Banks deal with volatile costs all the time, such as gas fees, cloud compute, and FX spreads. That is part of doing business.

The fee structure itself, not just the price, but the entire mechanism of how fees work, can be rewritten by the same forum. There’s no obligation to tell the institutions running on top of it.

The ZK Nation forum is already working on what comes next. They are debating node operator transaction fees, staking rules, and custom pricing for checking proofs. Each one of those is a live research thread that could become a vote that changes the economics of every cross-chain operation Deutsche Bank or Tradable runs. Anyone can read this on forum.zknation.io, publicly available.

Compare that to Kinexys, where JPMorgan holds the absolute keys to the system. Or Corda, where a formal contract dictates how adjustments happen and under what circumstances.

So, why would a bank ever choose ZKsync over JPMorgan? Because ZKsync gives you public proof that your transactions are valid without revealing your private data.

If JPMorgan decides to shut down its blockchain unit tomorrow, any assets on it are stuck. If Matter Labs (the company behind ZKsync) goes out of business, ZKsync keeps running because it settles on the public Ethereum network. But to get that, you have to accept that the network isn’t owned by anyone. A network that isn’t owned by anyone is governed by whoever shows up to vote.

$ZK is currently trading around $0.01. At its all-time high in June 2024, it was $0.3285. That’s a 96% drop. At 10 $ZK per interop call, that same cross-chain transaction costs roughly $3.28 at peak. Today it costs $0.10. The asset price volatility can be managed, but relying on an online community vote makes long-term budgeting entirely unpredictable for corporate planners.

L2Beat categorises ZKsync Era as a Stage 0 network, meaning an independent Security Council can pause or modify smart contracts without waiting for a full DAO vote. Mature Stage 1 networks, such as Arbitrum, lack this type of central intervention mechanism. Corporate risk managers often value emergency stop buttons to reduce the damage from smart contract exploits. However, this particular control belongs to a Web3 Security Council that functions outside traditional corporate structures.

Sygnum tokenised $50 million of Matter Labs’ corporate treasury into a Fidelity liquidity fund on ZKsync. Fidelity subsequently launched an institutional money market fund on the network. The builders kick-started ecosystem volume by placing their own capital into the product, establishing the core enterprise case study. The application processes live assets smoothly, yet the arrangement directly serves the founding team.

The core infrastructure remains tethered to an independent Security Council holding the keys to the system. During an emergency, this entity can bypass the standard delay mechanism completely, modifying smart contract parameters or freezing functions with zero advance notice. Control of the live system rests with this council and the active token forum, meaning participating banks operate under an organic, shifting governance structure rather than a fixed enterprise agreement.

ZKsync is betting its survival on a group that has never kept a crypto project alive before, that is, regulated financial institutions. Banks don’t care about tokens or post on governance forums.

Banks do, occasionally, build their core infrastructure on something and then never leave. That way of survival is lower to arrive, harder to dislodge.

ZKsync is either the first crypto project to achieve that, or the most expensive proof that banks will always eventually build their own version and walk away. The difference between those two outcomes is probably decided in the next 18 months.

The graveyard is full of projects that were right about the technology and wrong about the governance and the sustainability. The ZKsync governance test is still running.

—Thejaswini

Token Dispatch is a daily crypto newsletter handpicked and crafted with love by human bots. If you want to reach out to 200,000+ subscriber community of the Token Dispatch, you can explore the partnership opportunities with us 🙌

📩 Fill out this form to submit your details and book a meeting with us directly.

Disclaimer: This newsletter contains analysis and opinions of the author. Content is for informational purposes only, not financial advice. Trading crypto involves substantial risk - your capital is at risk. Do your own research.

OpenAI’s ChatGPT, launched in late 2022, has come so far that it has spawned an entire ecosystem of AI agents that now generate more web traffic than all humans on the internet combined. And these agents don’t do any of the things a human on the web does. They don’t watch ads, click links, or shop online. They scrape the internet for whatever they need to complete a task and leave.

The entire internet was built around human behaviour and intent, but now most of the time, there isn’t a human, and the web is pissed with it! They have started to block out these agents - 2.5 million sites are blocking AI crawlers, and there are lawsuits against platforms like Perplexity. Cloudflare even started building honeypot mazes that trap scrapers in an infinite loop of AI-generated gibberish.

But some of these agents are already smart enough to bypass this, and so the internet is losing its sh*t trying to build something more concrete called a ‘proof of humanhood’. A way to verify that the thing on the other end actually hesitates, mistypes, and moves its cursor with the micro-jitter of a real nervous system. In this piece, I’ll get into why this is happening, who’s building that proof layer, and what it means when the choice comes down to centralised biometric surveillance or cryptographic proof of humanity.

Follow the money

The reason websites started blocking AI agents is that AI has broken the web’s business model from both sides simultaneously. The entire model was built on human attention: you visit a website, you see an ad, and the publisher gets paid for it. An AI agent shopping for shoes on your behalf today searches 5,000 websites instead of the 4 or 5 you’d visit yourself.

It can read everything faster than your eyes can blink, compare every price, and even make the purchase for you in a matter of minutes, all without registering an ad impression, leaving the publisher to cover server costs and get nothing in return.

On the other end, AI search is eating into reader traffic. When Google shows an AI summary (AI Mode) at the top of the search results, only 8% of users ever bother to click through to the actual website. Google traffic to publishers has dropped 33%. In just one year after its debut,AI Mode has surpassed one billion monthly users, with queries more than doubling every quarter since launch.

Remember Chegg? It had its whole business built on being the first result for homework help, but it’s now shut down! The company officially ended its Q&A platform operations and blamed ChatGPT for it. Publishers are getting squeezed from both ends at the same time, with bots stripping away their content from the bottom and AI summaries intercepting their readers even before they ever arrive at the website.

The extraction ratios are crazy: for every referral OpenAI sends back to a website, its crawlers first visit 400 pages. Anthropic’s number is 38,000 to 1. These companies have trained their models on the open web without compensating anyone, and then deployed them to intercept traffic that once reached the same sites they scraped.

It’s quite funny, actually. In any other industry, an extraction ratio like that can get you lawsuits. But in AI, it gets you a trillion-dollar valuation.

Your body is the new password

For the last 25 years, the only way for the internet to know if you’re a human was the CAPTCHA. You clicked the traffic lights, typed some wobbly letters, and it was working because machines were terrible at visual recognition.

Today, that’s not the case. OpenAI’s Operator agent now surpasses Google’s reCAPTCHA with an exceptionally high human score, all while doing perfectly centred clicks and copy-pasting texts into forms. AI-generated selfies are fooling ID verification systems, and there are cases where Deepfake video calls have been used to authorise real wire transfers. All these tests assumed that machines were worse than humans at the tasks they performed.

The only way for the web to solve this is to double down on what AI still can’t replicate: how a human body physically moves when it operates a screen. This is called behavioural biometrics. Companies like IBM and BioCatch are building systems that can now check not just your identity at login but your entire session. This includes your cursor speed, how you scroll, your typing rhythm, which keys you press harder, whether you delete and retype words, and the angle at which you hold your phone; all of this is measured by a gyroscope.

It can also track which hand you use, how your thumbs arc when you swipe something. IBM needs about 8 sessions to build this behavioural profile of you, and once done, every action gets scored against your baseline in real time.

BioCatch can even tell when someone is being scammed online. Their system picks up the fragmented typing of a person reading out account numbers while a fraudster dictates them over the phone.

They have flagged about 2 million mule accounts across 257 banks in a single year. The European Union is piloting gait recognition at border crossings. We’re barely three years into the agentic era and already have EU border guards profiling how you walk.

There’s also a body of research on the Stroop effect. When you see the word “BLUE” printed in green ink, your brain takes noticeably longer to name the colour because the meaning and the visual conflict with each other. But an AI can still respond at the same speed.

Researchers found that this kind of cognitive interference actually correlates with typing dynamics. The system might not even need to run an explicit test on you. It’ll figure out whether you’re human just from the way your keystrokes land, because the way you type carries measurable traces of how your brain processes information.

Every previous way of online surveillance tracked your choices, like which sites you visit, what you clicked or bought, and you could always stop this by blocking cookies or using a VPN, and even turn off location sharing if needed. But behavioural biometrics reads your body. You can’t opt out of moving the cursor the way you move it, you can’t type with someone else’s rhythm.

Your behavioural profile is as unique as a fingerprint, and unlike a password or key, you can’t change or rotate it. And once this gets built, each platform will force the others to match and adapt to it. In fact, browsers are just the beginning; voice models already pass as human on phone calls, and video is next. So if this is the future, the most important question is who ends up owning that data?

Who holds the proof?

There are two camps building proof of humanhood right now.

The first is Sam Altman’s World (formerly Worldcoin). Here, you basically have to walk up to a chrome sphere called the Orb, which scans your retina and generates a cryptographic proof that you’re a unique human. 18 million people across 160 countries have already done it. In April 2026, World even signed a deal with Tinder, Zoom, and DocuSign to verify their users. They also partnered with Coinbase to build AgentKit, which lets you register your AI agents to your verified identity so the platform can confirm that an agent is backed by a real person without knowing who that person is.

But with that are also countries that do not support World’s iris scans and have completely banned them. All because of concerns around collecting biometric data from people who don’t fully understand what they are giving up. An MIT Technology Review investigation even found that World was collecting more than just iris scans, including heart rate, breathing, and other vital signs, without meaningful consent.

The second camp is crypto-based zk-proofs that let you prove you’re a human without actually revealing who you are, where you are, or what your face looks like. Vitalik Buterin laid this out in 2023 and argued that without a working decentralised proof of personhood, the internet will always default to centralised identity. Someone has to verify you, and if crypto doesn’t build that tool, governments and corporations will, and it’ll come with surveillance built into the foundation.

But Decentralised proof of personhood has already been tried at scale, and it collapsed. The Idena network was one of the first proof-of-personhood blockchains that gave every human an equal identity. But within two years, 23 entities had captured 40% of all accounts and 48% of all rewards on the network. Pool operators in Indonesia and Russia paid workers less than a dollar per hour to hand over their verified identities and pocketed upto 55x the return. Researchers also found photos of children used as identity puppets.

Vitalik had predicted this. He wrote that “the cheapest attack on any proof-of-humanhood system isn’t a deepfake or a sophisticated hack, it’s paying someone in a low-income country to lend you their identity.” The problem of building proof of humanhood is that there needs to be a financial incentive behind it. Somebody has to fund the Orbs, fund the blockchain for verification.

But the moment you attach financial value to a unique identity, you actually create a market for identity rental. And in a world with such massive wealth inequality, the market always gets captured by whoever has the most capital.

“Forcing one-person-one-vote into a system with real economic incentives always conjures failed experiments in 20th-century communism.”

So both paths are kind of broken. The centralised path works at scale, but puts your biometric data in the hands of a company that already overcollects it and benefits directly from the bot problem persisting. The crypto path protects privacy in theory, but gets eaten by the same economic power imbalances it was built to escape.

If I had to bet, I’d still bet on crypto. Because behavioural biometrics and centralised iris scanning create a permanent record of your body that belongs to whoever deploys the system. And once they have your data, you can’t delete or port it; it will be locked with the company that captured it.

A zero-knowledge proof that confirms you’re a human without telling anyone anything more is still worth building, even knowing it’ll get gamed. Because the alternative to it is a web where every site you visit has your body’s movements on file. And that corporate version is already shipping faster than the crypto one.

That’s all for today!

Vaidik

Token Dispatch is a daily crypto newsletter handpicked and crafted with love by human bots. If you want to reach out to 170,000+ subscriber community of the Token Dispatch, you can explore the partnership opportunities with us 🙌

📩 Fill out this form to submit your details and book a meeting with us directly.

Disclaimer: This newsletter contains analysis and opinions of the author. Content is for informational purposes only, not financial advice. Trading crypto involves substantial risk - your capital is at risk. Do your own research.

Saint-Domingue, in its prime, was the ultimate jewel of the Caribbean.

It’s a small island that by the late 1770s was the wealthiest place in the Western Hemisphere. The “pearl of the Antilles,” they called it.

By 1780, it produced 40% of all the sugar and 60% of all the coffee consumed in Europe. The economy functioned well, if you didn’t examine closely who was doing the work and the conditions they faced.

In August 1791, the enslaved workers rose in revolt. Fourteen years of civil war. Independence in 1804. And then France (France!) demanded reparations from Haiti, not to the formerly enslaved, but to their former owners. Those payments lasted until 1947 and, at times, consumed 80% of Haiti’s national revenue.

The place that was once the wealthiest in the Western Hemisphere is today the poorest.

France was the birthplace of the Enlightenment, the French Revolution, and the famous slogan: “Liberty, Equality, Fraternity.” They were supposed to be the global champions of human rights and freedom. Yet, when black slaves embraced that slogan and freed themselves, France responded by financially suffocating them for the next 150 years.

Wu uses this story to make a point that unequal economies are fragile. The more you concentrate wealth, the more brittle the system becomes. And at some point, the brittleness resolves itself through a catastrophe that everybody paid for.

First, giant companies choke out all competition and drain people’s wallets. That financial pressure turns into deep public rage because people feel the game is rigged. That rage is what kills a democracy.

Once the average person realises their vote can’t change their bank account, they throw their hands up and look for a dictator to reset the board. History shows us this exact script, tracking all the way through the industrial monopolies of Weimar Germany.

For me, that hits right in the gut when I look at our current political circus. You watch these politicians spend an entire year talking a big game on television about making the ultra-wealthy pay their fair share. Then, the very second they take office, the script flips, and they are signing tax breaks for real estate giants and bailing out massive corporate banks. It leaves you sitting there with a higher rent bill, and this cynical realisation that the voting booths are just a distraction while the donor class runs the actual country. We are watching a hostile corporate takeover where oligarchs and big tech executives see massive returns on their political investments while working people are left to bear the cost.

That’s the thing that historically precedes the strongman.

Wu calls this a sequence, which implies it can be interrupted. Well, I don’t buy it. Everything he describes looks exactly like what the system was built to do from day one.

Private equity firms didn’t stumble into owning our hospitals by mistake. This is raw capital doing what it always does when there are no rules to stop it, most times. Wu knows this. But his background as a reform capitalist means he is structurally committed to the belief that the system can be repaired from within. I’m just not sure I share that faith.

The Denmark part is the one I’d make everyone read.

In the 1780s, Denmark was feudal, poor, and run by a small class of lords who owned 98-99% of the arable land. A nineteen-year-old prince regent took power and, with an advisor named Bernstorff who had spent his whole career thinking about land distribution, abolished serfdom and converted peasants into landowners over the following decade. Grain production went up 40%. The peasant class became a political force. The core of what Denmark is today as one of the wealthiest, most equal countries on earth traces back to those ten years.

Russia had the exact same opportunity but blew it. Catherine the Great also wanted everyone to think she was a hip, enlightened intellectual. She debated freeing the serfs in her Nakaz, but her circle of wealthy buddies hated it, so she let it die. Instead of fighting for her people, she spent her reign collecting masterpieces and writing to Voltaire while the feudal system grew more toxic. When Russia tried to fix things a hundred years later, the damage was irreversible. The whole system ended up burning to the ground.

The entire Denmark comparison depends on ownership of the productive base. 200 years ago, that meant soil. Today, we think it is digital infrastructure and data pools. The monopolies controlling this matrix decide if we build a fair society or devolve into corporate feudalism.

Crypto was supposed to take economic power out of corporate hands and put it in the hands of regular people.

It didn’t work out that way. Part of the reason it failed is that the initial, radical premise was unrealistic. I used to write some pretty harsh things against corporate centralisation, but I am learning to understand exactly why it happens. We have essentially matured as an industry and prepared ourselves to live with a completely new era and a different promise.

Read: Vertically Integrated Capital Aggregators - by Joel John

BlackRock is running multi-billion-dollar crypto funds while Fidelity dominates the custody market. The institutional giants that the early developers were trying to destroy are now funding the ecosystem. Andy Bromberg, the CEO of Eco, pointed out that Wall Street is on track to own 70% of all Bitcoin. We all know what that means.

Wu sees that crypto hasn’t solved global wealth inequality, and basically calls it a failure. Fair, it hasn’t moved the needle there. But Bitcoin emerged as a response to the 2008 banking crisis; it aimed to create a trustless infrastructure. The claim of redistribution came later, mainly from those trying to sell tokens. Judging the original technology by its promoters’ promises completely misses the point. The better measure is if the infrastructure being built could outlast the current extractive order, the corporate banking system. Whether the alternative settlement rails and cross-border plumbing retain enough of the original architecture to function differently from what they are replacing. On that, the answer is genuinely open. Wu can’t close it yet.

Think about when they invented the shipping container back in the fifties. It didn’t cure world poverty, and every day, people weren’t out buying containers.

But by standardising how freight moved, it rewired global trade networks and broke the monopoly of traditional port bosses. So, the backend infrastructure is supposed to change the machinery running in the background of the world, not the storefronts.

Wu’s actual solutions feel way too polite for the mess we are in.

He talks about basic fixes like bringing back antitrust lawsuits and forcing tech platforms to act like neutral public utilities. It’s an old concept from the fourteenth century. Back then, if you ran an essential inn, the law said you couldn’t turn people away.

He also suggests capping corporate junk fees. For example, Europe caps credit card swipe fees at 0.3%, while America leaves them uncapped at 3%, which lets banks drain $135.75 billion in hidden fees in 2023 alone. Another fix is to use line-of-business restrictions to prevent monopolies from taking over adjacent markets. That’s how the government blocked AT&T from entering the computing space decades ago, which is the only reason an independent tech industry had room to breathe.

That has maybe worked in the past. All of his fixes require a functioning government that isn’t completely bought and paid for by the exact corporations it’s supposed to regulate = non-existent.

He uses this metaphor about a gardener pruning back overgrown corporate weeds so the rest of the ecosystem can catch some light.

But that assumes the gardener is independent and has the right tools. In reality, the gardener has spent the last 40 years being defunded and captured. Politicians never forgot about antitrust laws. Instead, billionaires spent decades buying politicians and economists who needed to make monopolies look acceptable.

The metaphor falls apart if the weeds have already grown so high that you can’t even find the garden gate anymore.

I picked up this book just looking for something solid to read. Three weeks later, I have his five-step cycle out of my head, running through it again and again, and it explains a lot of things that have crossed my mind lately.

Monopoly triggers extraction. Extraction breeds public resentment. Resentment breaks down institutional guardrails. Finally, institutional failure paves the way for the strongman.

Wu is more patient than I am. He still believes in institutions, in the slow grind of regulatory reform, and in the idea that the gardener will eventually show up. But he knows his history well enough to remember that the Danish miracle actually happened. A brilliant advisor named Bernstorff managed to get a teenage prince to sign a decree. Ten years of focused execution turned a broke feudal estate into one of the most stable, equal societies on earth.

Bernstorff spent his whole life learning the details of land distribution before he got close to the levers of power. And he got lucky with the kid on the throne.

At this point, I’d take a gardener over another billionaire trying to tell me what freedom looks like.

—Thejaswini

Token Dispatch is a daily crypto newsletter handpicked and crafted with love by human bots. If you want to reach out to 200,000+ subscriber community of the Token Dispatch, you can explore the partnership opportunities with us 🙌

📩 Fill out this form to submit your details and book a meeting with us directly.

Disclaimer: This newsletter contains analysis and opinions of the author. Content is for informational purposes only, not financial advice. Trading crypto involves substantial risk - your capital is at risk. Do your own research.

You don’t need to set foot in Texas to hear it. Rockdale is home to North America’s largest Bitcoin mining operations (like Riot Platforms and Bitdeer) housed in an old aluminium plant. Investigative journalists from the New York Times and Al Jazeera have documented that ‘jet engine’ roar, it’s the sound of thousands of industrial fans keeping tens of thousands of ASIC mining computers from melting down in the Texas heat.

Follow that noise, and you’ll find the old Alcoa aluminium smelter. For decades, it was a brutal, sweating monument to 20th-century manufacturing.

Today, if you look inside those same metal warehouses, there isn’t an ounce of aluminium in sight. Instead, you’ll find miles of thick copper cables and industrial racks where computers are completely submerged in tanks of churning, boiling synthetic oil just to stay cool.

First, it was Bitcoin. Now the gear is being swapped out for AMD chips to train AI.

Don’t look at this shuffle and try to guess whether AI is a bubble or Bitcoin is dead (it’s not); you’re missing the forest for the trees. The companies buying up these leases don’t care about that. They know that the asset is the power line. If they didn’t know it then, they definitely know it now.

If you are still wondering why, one kilowatt-hour of electricity generates roughly $0.17 to $0.27 in gross revenue when used to smelt aluminium, depending on the LME price.

Bitcoin mining with current-generation ASICs produces $0.05 to $0.11 per kilowatt-hour at today’s prices.

AI inference running on H100 GPUs, billed through cloud pricing, generates $1.27 to $3.67 per kilowatt-hour.

Aluminium was a reasonable use of cheap electricity when nothing better was available. Bitcoin requires a reasonable use of cheap electricity when aluminium margins are compressed.

The bitcoin price justified this for some time. Now in 2026, AI is a better use of cheap electricity than bitcoin at almost any price the crypto market has hit this year.

If you want to understand how fast the ground is shifting, you only need to look at three recent deals. They show exactly how companies are scrambling to grab any power source they can find, whether they use it for AI or crypto.

Riot had a giant site in Rockdale, Texas, but instead of just using it to mine bitcoin, they leased a chunk of the space to the chip giant AMD for an AI data centre project. Riot realised they could make hundreds of millions of dollars simply by letting others use their power hookups.

TeraWulf did a similar thing on a massive scale. They paid $200 million to buy the old Century Aluminium plant in Hawesville, Kentucky. They bought it because an aluminium smelter already has giant, heavy-duty electrical infrastructure built into the land. TeraWulf is going to tear down the old factory parts and build a data centre campus right on top of those power lines.

NYDIG went after the old Massena East factory in New York. This site used to be owned by the aluminium giant Alcoa and has been sitting mostly empty for years. But it has direct access to 435 megawatts of cheap, clean hydropower from the St. Lawrence River. While other companies are switching to AI, NYDIG bought this specific spot to lock in cheap hydropower for its bitcoin mining. Is nobody building from scratch anymore?

For almost 20 years, Bitcoin miners wandered the map in search of the cheapest energy. They set up next to remote hydro plants in rural Washington state, tapped into flared gas vents in North Dakota oil fields, and took over old industrial grids in upstate New York. They built facilities capable of using large amounts of electricity around the clock without damaging the hardware. They mastered industrial cooling, low-cost power contracts, and 24-hour operations.

Then AI companies showed up needing precisely what these companies already had, and the AI companies had a much bigger budget, too.

Anthropic is locking down massive amounts of electricity. At the same time, Microsoft, Google, and Amazon are building data centres so fast that the electric companies can’t even get the power lines hooked up in time. All three are bidding directly against bitcoin miners for the same industrial power lines. Bitcoin miners used to fight each other for electricity, but now the richest corporations are fighting for the same input, and we know who’s losing here.

You can see the damage in the numbers from early 2026. For the first time in six years, the total amount of computing power on the Bitcoin network declined. The production cost of mining one bitcoin is 88000 dollars, but the token traded around 77000 dollars for much of May 2026. If you are a miner paying normal power rates, you are losing money on every single coin you make.

So, everyone is swapping jerseys. Companies like Hive, Hut 8, TeraWulf, and Iren are ripping out their mining rigs to make room for AI servers. CoreWeave quit bitcoin completely to build an AI cloud network, and MARA bought a French tech company just to pivot. The miners who treated themselves like power brokers are surviving. The ones who thought they were crypto guys are running out of money.

Energy analysts call this the “digital resource curse.” This means countries and companies are realising that simply owning and holding onto raw electricity makes more money than actually building new technology.

The Gulf states figured this out long before the crypto industry did.

For around sixty years, Gulf governments sold electricity at very low prices. In Kuwait, the home electricity rate has been fixed at 0.7 cents per kilowatt-hour since 1966. Abu Dhabi spends about 8.7 cents to make and deliver a kilowatt-hour of electricity, then it sells it to citizens for 1.4 cents. These low rates were the government's strategy to build up industry, to tempt heavy industries into the desert. Aluminium smelters, chemical plants, and steel mills were built there only because power was almost free.

Source: bakerinstitute.org

Now, instead of aluminium smelters, that cheap electricity is bringing in data centres. Saudi Arabia started a state-backed AI investment firm called HUMAIN to invest billions in tech infrastructure. The UAE broke ground on its massive 5-gigawatt Stargate AI campus, bringing in OpenAI, Oracle, and Nvidia to use the very power grids that once melted aluminium ore. Even NEOM’s Oxagon, which started as a plan for a floating factory city, is now a 5-billion-dollar AI data centre campus powered by wind and solar.

As the Carnegie Endowment noted, cloud computing is becoming the Gulf’s new aluminium. They are just exporting their fossil fuels and solar power through the internet instead of shipping physical cargo.

Then there is Bhutan.

Bhutan had the cheapest hydroelectric power in the world – my favourite Bitcoin mining story –and a government-backed bitcoin mining operation that was supposed to be the sovereign mining success narrative. Holdings peaked at around 13,000 bitcoin and dropped to 3,100. Mining inflows stopped more than a year ago. The energy appears to be now being sold directly to India.

This is the same calculation Alcoa made. Is Bitcoin the most profitable thing we can do with this electricity? When the answer was yes, Bhutan mined. When selling electrons to the Indian grid became more reliable than converting them to bitcoin at uncertain prices, the power went to India.

Similarly, Starcloud raised $200 million to build solar-powered data centres in orbit. They just trained the first AI model in space using an H100 GPU and are filing for 88000 satellites. Their plan includes mining Bitcoin, but only as a side hustle. When their orbital solar panels are soaking up the sun, and the AI queues are quiet, mining crypto is the most profitable way to use the leftover capacity.

In low Earth orbit, solar panels get continuous, raw sunlight. There is no night cycle or land costs. You do not even need cooling infrastructure beyond the freezing temperature of space itself. Plus, the cost of launching hardware up there has dropped by 95% over the last twenty years.

SpaceX is playing this exact same power game. According to its recent IPO paperwork, the company is earning $1.25 billion per month from Anthropic, which just signed a deal to rent all the computing power at SpaceX’s Colossus 1 data centre in Memphis, Tennessee. That contract is worth over $40 billion through May 2029. Just like the crypto miners who took over the old aluminium smelter in Rockdale, SpaceX set up Colossus 1 inside a converted Electrolux appliance factory.

Now, Allbirds is the most absurd example in this whole piece. It was a sustainable shoe company. At peak, Allbirds was valued at $4 billion. Then the direct-to-consumer brand bubble deflated, and the stock fell 98%. They had some cash, a public company listing, and nothing working in footwear. They pivoted entirely to AI compute infrastructure. The stock rose 350%. Same logic here. Whatever Allbirds makes or does, the market has decided that being in the business of running electricity through servers pays more than being in any other consumer business.

Crypto networks like Bittensor, Render, and Akash are focusing on a different approach. Instead of creating a single large facility, they want to connect smaller computers distributed around the globe.

Bittensor operates a marketplace where AI models compete to answer questions, using a fixed token supply modelled after Bitcoin.

It even cut its daily token creation in half back in December 2025. Meanwhile, Render lets people share their extra graphics card power for AI jobs, and Akash rents out cloud computing space at prices they claim are 85 per cent cheaper than Amazon Web Services.

The strategy is getting mainstream attention. At Nvidia’s tech conference in 2026, CEO Jensen Huang compared Bittensor to Folding@home, an old internet project.

That makes sense because Folding@home happened because millions of home computers were just sitting there, literally burning electricity to do nothing. They figured out how to bundle that digital dust into something useful. Bittensor is asking people to donate their spare gaming PCs and leftover mining rigs and dangling crypto tokens to get them to plug in.

When I look at this entire landscape, from the roaring fans in Rockdale to satellites orbiting the earth just to chase raw sunlight. I see a massive, frantic reshuffling of physical assets. The people running these companies are only loyal to the margins. My bet is that in ten years, we’ll see these exact same warehouses get gutted all over again for whatever comes after AI, while the physical grid underneath them remains completely unchanged.

Whoever finds the cheapest electricity gets to decide what computation runs on it. That has been true in Texas, Bhutan, and Abu Dhabi, and it will be true 250 miles above all of them.

—Thejaswini

Token Dispatch is a daily crypto newsletter handpicked and crafted with love by human bots. If you want to reach out to 200,000+ subscriber community of the Token Dispatch, you can explore the partnership opportunities with us 🙌

📩 Fill out this form to submit your details and book a meeting with us directly.

Disclaimer: This newsletter contains analysis and opinions of the author. Content is for informational purposes only, not financial advice. Trading crypto involves substantial risk - your capital is at risk. Do your own research.

Financial markets in general are not built this way. The New York Stock Exchange runs from 9:30 am to 4 pm Eastern, Monday through Friday. London runs 8 am to 4:30 pm. Tokyo runs 9 am to 3:30 pm, with a lunch break.

When one closes, the next opens, more or less. What I am saying is, capital should move continuously because the sun is always up somewhere. In practice, every major exchange has hours, creating long periods of downtime when regulated markets are closed.

There was a reason why it was designed this way. It grew from the physical constraints of people standing in rooms and shouting at each other. Computers were supposed to fix it. They made the shouting faster and still kept the schedule.

An object at rest stays at rest until something acts on it. That’s what we studied in school. The financial market schedule stayed because nothing forced it to move. Then, on a Sunday morning in May, someone wanted to price SpaceX before the bankers did. Hyperliquid was open. The contract launched at 5:16 am UTC. $33 million moved in 24 hours. Morgan Stanley had not yet opened the books.

The Indian time zone turned out to be the right schedule for this story. The US financial press wakes up at 9:30 am Eastern. By then, I had been watching markets for most of the afternoon.

CME Group (Chicago Mercantile Exchange) is the largest derivatives exchange in the world. That’s where professional traders buy and sell futures on oil, gold, interest rates, stock indices, and Bitcoin. Trillions of dollars in daily volume, and they have been around since 1898.

ICE (Intercontinental Exchange) owns the New York Stock Exchange. Also owns derivatives exchanges globally. Another titan.

These are the most powerful financial market infrastructure companies on earth. It’s tough for regulators not to listen to them when they point at something and call it dangerous.

CME and ICE are lobbying the CFTC and Capitol Hill to crack down on Hyperliquid, warning that the KYC-free platform is a breeding ground for market manipulation and sanctions evasion.

• There’s zero KYC here. While Hyperliquid uses a frontend filter to block OFAC-sanctioned addresses on its main website, the underlying protocol is fully permissionless. If someone bypasses the website and interacts with the smart contracts directly, there’s no identity check waiting for them.

• There are no position limits on Hyperliquid. On CME, no single trader can hold more than a certain size position in any contract. That prevents manipulation and systemic risk. Hyperliquid has no such limits.

• CME monitors trading patterns for manipulation, such as spoofing, wash trading, and coordinated attacks. Hyperliquid has no surveillance system watching for this.

Fair points.

HYPE dropped 9% on May 15 on news. Two market makers pulled $100 million in liquidity on May 18 in response.

But watch which specific product they flagged. Not Hyperliquid’s crypto perpetuals, which have run for years without regulators blinking. The focus was on the oil contracts. The contracts that traded $720 million over a weekend when CME’s own oil markets were dark.

CME and ICE aren’t entirely wrong about the regulatory headache here, but we also know they aren’t neutral bystanders. Their business model relies on a legally protected monopoly over the trading calendar. They don’t mind competing on tech, but they absolutely lose their minds when someone competes on time.

By pulling real volume into oil over the weekend, Hyperliquid basically broke the space-time continuum of tradFi. And the incumbents are asking the government to force everyone else to go to sleep when they do. I’d rather ask permission to open on weekends; apparently, that’s just me

Hyperliquid has 11 people in a Singapore office. The protocol generated $51 million in revenue in the 30 days ending 21 May 2026. In March, it processed $2.6 trillion in notional derivatives volume.

Hyperliquid routed 97% of its trading fees into HYPE token buybacks via an on-chain fund. With 11 people generating $51 million in monthly revenue, the per-employee economics have no real comparison in crypto or outside it. HYPE is up 101% year-to-date as of late May.

All of this is not necessarily because Hyperliquid built a better derivatives product in any technical sense. It’s just open when CME is closed, which is enormously valuable. Recently, something new happened that extended this logic further.

On May 1, Trade.xyz, a platform built on Hyperliquid, launched pre-IPO perpetual futures for Cerebras, the AI chipmaker. The contract ran for two weeks before the IPO. Early in that window, traders implied a roughly 50% premium over the $185 IPO price, indicating an opening around $277. Then the market updated. One hour before Cerebras opened on Nasdaq, Trade.xyz’s perp was pricing the stock at $340, within 3% of the actual opening price. Cerebras opened at $350 on May 14, up 89% from its $185 IPO price. Traditional secondary platforms like Forge and EquityZen missed by 35%. Hyperliquid missed by 3%.

$277 was what the market implied when there was still genuine uncertainty. By the time the information had come in and been priced, the crowd had closed the gap. That is how price discovery is supposed to work.

Then on Sunday morning, May 17, Trade.xyz launched a perpetual futures contract for SpaceX. It opened at a $150 reference price, ran to $216 within hours, and finally settled around $203, meaning the crowd priced the company at a $2.4 trillion valuation.

But SpaceX hadn’t dropped a public S-1, no Wall Street analyst had touched a price target, and the official roadshow was nonexistent.

Traders had absolutely no idea that SpaceX had already slipped a confidential filing to the SEC on April 1, targeting a $1.75 trillion to $2 trillion valuation.

Traders settled the contract at $203, implying $2.4 trillion. The crowd landed at the top of the company’s own target range, before the bankers had their first meeting, without access to the filing. SpaceX officially and publicly filed its actual 277-page S-1 prospectus just a few days later on Wednesday, May 20.

Three products are currently trying to give investors exposure to SpaceX. Each one took a different bet on the legal surface.

PreStocks tried to get clever. They set up special investment funds (called SPVs) to buy actual SpaceX shares, then sliced those funds up into blockchain tokens so everyday investors could buy a piece. It looked like a clean backdoor into private tech companies.

Just before Hyperliquid launched its SPCX contract, Anthropic and OpenAI publicly disowned third-party SPV products that purported to track their valuations. Platforms in Hong Kong and UAE had been selling tokenised exposure to both companies without board approval. Both companies released warnings stating the share transfers behind these products were invalid. PreStocks tokens crashed 50%. When you own actual shares, the company whose shares you own retains the right to intervene.

Ondo Global Markets tokenises stocks through a US-registered broker-dealer, each token backed by the underlying security. The compliance is clean, and DTCC is building settlement infrastructure around it.

But Ondo’s greatest strength is also its biggest vulnerability. It has a physical address. If the SEC decides they don’t like it, they know exactly whose door to knock on. If Elon Musk objects, SpaceX lawyers know exactly which custodian to sue. By playing by the rules, Ondo makes itself a perfect target.

Read: Regulatory Asphyxiation - by Thejaswini M A

And then we look at Hyperliquid’s SPCX contract, which is built on absolutely nothing.

There are no shares, registered broker-dealers or claims on any physical asset. It is a synthetic perpetual, an absolute ghost. It’s a pure bet on a price ticker, settled entirely in USDC on a decentralised network.

If SpaceX wants to stop people from trading derivatives based on its valuation, it can’t. There is no corporate entity to serve legal papers to, and no central issuer to pressure.

It’s brilliant, really. Hyperliquid basically realised that you can’t get punched in the face if you don’t have a face. By anchoring the product to nothing physical, they became untargetable.

I am not sure this is straightforwardly good.

It is genuinely hard to argue against the national security nightmare of a zero-KYC venue routing trillions of dollars outside the global banking system. Hyperliquid co-founder Jeff Yan flying to Washington on May 17 to meet with policymakers shows just how live this pressure is.

Talking about Jeff Yan. He is a real human, he has a face, and he went to Harvard. If SpaceX wants to sue him for trademark infringement or intellectual property violation because his platform hosts a contract called “SPCX,” they can absolutely serve him with papers.

But suing Jeff doesn’t stop the contract.

With PreStocks, if the company deletes the underlying shares, the product ceases to exist. With Ondo, if a judge freezes the bank or custodian, the product struggles. But Hyperliquid’s SPCX is an autonomous code deployment. Even if Jeff Yan gets sued into oblivion, the smart contracts are already live, the code is immutable, and the global order book keeps running on-chain.

That is the immaculate theory of decentralisation. The practical reality is a bit more fragile. Hyperliquid runs on just 20 validators, not 900,000. They are identifiable. And the JELLY incident showed they will intervene when they choose to. The validators are not immutable.

Again, the clock is the only product they can’t copy.

—Thejaswini

Token Dispatch is a daily crypto newsletter handpicked and crafted with love by human bots. If you want to reach out to 200,000+ subscriber community of the Token Dispatch, you can explore the partnership opportunities with us 🙌

📩 Fill out this form to submit your details and book a meeting with us directly.

Disclaimer: This newsletter contains analysis and opinions of the author. Content is for informational purposes only, not financial advice. Trading crypto involves substantial risk - your capital is at risk. Do your own research.

Up until last week, if you asked ten lawyers whether Ethereum was a security or a commodity, you’d get twelve different answers and a bill for $50,000. That’s been the reality of building anything in crypto in the United States. There were no clear rules because the regulator refused to establish them and later sued for non-compliance, citing an after-the-fact interpretation.

The SEC under Gensler brought 88 enforcement actions against crypto projects, and 92% of those were for registration violations, meaning companies were punished for failing to register under a framework the agency never clearly defined. It was absurd; everyone in the industry knew it was absurd, but no one could do anything about it because the alternative would mean leaving the country.

But that changed last Wednesday when the Senate Banking Committee voted 15-9 to advance the CLARITY Act. Elizabeth Warren called it a bill that “blows a hole in securities law since 1929.” She’s half right. The bill was shaped with heavy industry input. But the hole it blows should have been blown years ago.

What does the CLARITY Act actually do? It gives the security-or-commodity question an answer backed by a real test. By default, every token starts as a security. The moment you raise money by selling tokens and promise to build something with that capital, that’s an investment contract under Howey, and you fall under the SEC. That much hasn’t changed; it’s been the case since crypto fundraising began.

What’s new is that there’s now a way out. The bill introduces what it calls the “mature blockchain” test. If your chain is open-source, runs on pre-established, transparent rules, and no single person or group controls more than 20% of the token supply, you can go to the SEC and say, We’re decentralised enough; review us. If they don’t push back within 60 days, your token is reclassified as a digital commodity, and regulatory authority changes from the SEC to the CFTC.

This shift from the SEC to the CFTC matters greatly because the two agencies operate in completely different ways. The SEC treats tokens like stocks, which means full registration, detailed disclosures, and continuous reporting obligations. The CFTC treats them like commodities, such as oil or wheat, with lighter oversight and lower costs, and its main job is to make sure markets run fairly rather than to decide who gets to participate in them. For any project that passes the decentralisation test, the day-to-day cost of being a regulated entity drops significantly.

You get four years to get there. You can file a notice with the SEC saying your blockchain intends to reach maturity within that window. As long as you’re making progress, you operate under a temporary exemption. But if you hit the four-year mark and you’re still not decentralised enough, the exemption goes away, and you’re back under full securities law with even stricter disclosure requirements than you started with.

If you pair this with the GENIUS Act for stablecoins, which became law last year, you will see that, for the first time, the United States has a complete regulatory framework for digital assets. Stablecoins have clear rules about reserves, licensing, and who can issue them. Tokens now have a concrete test that determines whether they fall under the SEC or the CFTC, and that test includes a specific threshold: 20%. But what does it mean for existing and new projects with a token? That’s where it gets interesting.

The Impossibility of Grassroots Launches

The biggest crypto assets will have no trouble with this test. Bitcoin has no single entity with anywhere near 20% ownership and has been treated as a commodity for years. Ethereum runs on over 1.07 million validators after the Merge and has the most distributed infrastructure of any smart contract platform.

The SEC and CFTC’s joint interpretive release from March 2026 already identified 18 tokens as digital commodities, including Bitcoin, Ethereum, Solana, XRP, Cardano, Chainlink, and Avalanche. If you hold any of these, you can breathe easy knowing the regulatory question is answered, and your token sits on the commodity side of the line.

The problem starts when you look past those 18 names in that list. Take Solana, it made the commodity list, but it sits in a grey area. It got there through an interpretive release, which is a regulatory opinion.

An interpretive release is basically the SEC and the CFTC issuing a statement saying, “This is how we read the existing law.” It carries weight; markets react to it, but it has no legislative backing. The next SEC chairperson can issue a new interpretation and change Solana’s commodity status overnight, without going through Congress or holding a vote.

The other group that needs to pay close attention is everyone who hasn’t launched a token yet. Under this bill, every new token is presumed to be a security from the moment it’s created, and escaping that means years of SEC disclosures, legal filings, and semiannual reporting. At the same time, you work toward the maturity threshold.

Hiro Systems, one of the few projects that actually tried to go through the SEC’s existing registration process, spent over $15 million on compliance and legal defence alone, which was more than they raised in the offering itself.

That gives you a sense of the real cost here. The bill includes a $50 million exemption cap for how much you can raise while your chain is maturing. Still, the compliance infrastructure needed just to operate within that exemption is so expensive that only venture-backed teams with law firms on retainer can realistically use it.

It means that if you’re a small team trying to build something new without institutional backing, you will have a hard time passing this test. The kind of grassroots, community-driven launches that gave us Ethereum in 2014, where anyone could participate in an $18 million raise with no regulator asking questions, those become illegal to pull off under this framework.

The DeFi Safe Harbour That Almost Wasn’t

The commodity classification and the decentralisation test got all the attention. But the DeFi developer safe harbour in Sections 309 and 409 may be the most important part of the bill.

The bill says that if you write smart contract code, run a validator, or build a self-custodial wallet, you are not a financial intermediary. You don’t need to register as a broker, you’re not a money transmitter, and the government can’t come after you as one. Code is not custody. That’s now written into the bill.

The credit for this goes to a guy named Roman Storm. Storm built Tornado Cash, a privacy tool on Ethereum. He didn’t hold anyone’s money, couldn’t freeze or reverse transactions, and couldn’t shut the protocol down even if he wanted to. The code was open-source and ran on its own. The U.S. government convicted him in August 2025 for operating an unlicensed money-transmitting business anyway, because no law at the time distinguished between writing software and running a money-transmission operation.

But the protection still has huge problems. During the committee vote, a last-minute amendment changed the wording regarding when a developer can still be subject to regulation. It now says that if you’re acting pursuant to an agreement, arrangement, or understanding to control a protocol, the safe harbour doesn’t cover you.

That means, on protocols like Aave or Compound, token holders who regularly vote on upgrades and treasury decisions could now easily be read as having an “arrangement,” and that alone would be enough to strip the protection from everyone building on those protocols.

The safe harbour covers the back end, smart contracts, validators, and node operators. But it says nothing about front-end interfaces. And almost nobody interacts with DeFi through raw smart contracts. They use websites like app.uniswap.org or app.aave.com. If a regulator decides that running one of those front ends constitutes operating a financial service, then the safe harbour protects the code but not the product that anyone actually uses. And this could now become the next big regulatory battle in DeFi.

Jake Chervinsky, who runs the Hyperliquid Policy Centre, said: “If the bill doesn’t work for DeFi, it doesn’t work at all.” And it’s true because if that language stays as written, the safe harbour protects developers on paper, but at the same time, exposes them in practice.

Warren also tried to pass an amendment that would give the Treasury the power to sanction DeFi protocols, just as it did with Tornado Cash back in 2022. It failed 11-13, and every Republican voted against it.

Right now, it’s unclear whether the government can legally sanction software that no one controls. That question will end up in court eventually. And when it does, the lawyers defending DeFi protocols will be able to point to this vote and say, Congress already debated this exact question, and they decided the government shouldn’t have that power. That’s a strong argument to have in your pocket, and it came out of a failed amendment.

Who Wins, and What Happens Now

The biggest winners here, though, are the banks. By effectively killing SAB 121, the CLARITY Act removes the accounting rule that forced institutions to treat customer crypto as liabilities on their own balance sheets. This was the primary wall keeping banks out of the custody game. Now, every major financial institution can hold Bitcoin and Ethereum without blowing up its capital ratios. Institutional custodians like BitGo and Anchorage can finally graduate from simple storage to building out prime brokerage and clearing services with genuine legal backing.

Tokenisation platforms are also about to see their theoretical TAM become reality. Market estimates for tokenised assets are currently wild guesses, ranging from $2 trillion to $30 trillion by 2030. Those trillions haven’t moved because there was no regulated path for trading these assets. The CLARITY Act clears that blocker, creating the legal bridge that institutional capital needs to actually cross the chasm.

But the most interesting shift lies in the interaction between this bill and the GENIUS Act. Because the stablecoin law bans passive yield on holdings, you can no longer just park USDC on an exchange and collect 5%. Yield now requires active participation – staking, governance, or providing liquidity. This means hundreds of billions in capital that once sat idle are now being directed toward structured DeFi protocols such as Pendle, Morpho, and Maple Finance. Lawmakers probably didn’t intend to create a massive forced migration of capital into DeFi, but by making passive holding unproductive, they’ve done exactly that.

The CLARITY Act is better than what came before, which was a decade of legal uncertainty in which the government regulated crypto by filing lawsuits rather than creating the rules. But it’s also a bill shaped by the companies already here, and it shows.

The compliance costs, the four-year maturity timelines, and the legal infrastructure required to use the exemptions all favour the projects that have the money and the lawyers to navigate them. If you’re Coinbase, this is the framework you’ve been asking for. If you’re building the thing that comes next, the rules were written before you got to the table.

That’s always been how regulation works, though. Whether that’s what crypto was meant to become, we still can’t say.

That’s all for today!

Vaidik

Token Dispatch is a daily crypto newsletter handpicked and crafted with love by human bots. If you want to reach out to 170,000+ subscriber community of the Token Dispatch, you can explore the partnership opportunities with us 🙌

📩 Fill out this form to submit your details and book a meeting with us directly.

Disclaimer: This newsletter contains analysis and opinions of the author. Content is for informational purposes only, not financial advice. Trading crypto involves substantial risk - your capital is at risk. Do your own research.

You are moving massive, terrifying chunks of global money; you do not get a pass for sloppy homework. If you leave a structural gap open, the market or the state will violently correct you.

So, obviously, crypto developers spent years trying to smash that window with a brick. FTX offered tokenised stocks in 2020 and got completely vaporised by regulators. Gone. Then Binance tried it a year later and quit after ninety days because, I don’t know, pressure?

And Mirror Protocol built these synthetic copies that vanished into actual nothingness when the Terra blockchain collapsed (RIP). But why did we always assume the tech mattered more than the law?

But okay, if crypto's actual end game is to become the core infrastructure for global finance, we have to get the behaviour right. And learn to hate shortcuts. Ondo was built very carefully, but the capital flowing through is chaotic. Let’s look at the numbers to see how they got here and why the reality looks a teeny-weeny-tiny different.

Which brings us to Ondo Global Markets. They fixed the legal issue first. When they launched in September 2025, they skipped the offshore islands entirely, no Bahamas, no Seychelles, nothing beautiful for the eyes. They went straight to the SEC with a confidential registration statement, putting real shares inside a US broker-dealer.

Tether hit $1 billion in circulation in December 2017, 38 months after launch. USDC crossed that line in July 2020, 21 months after Circle and Coinbase launched it. Ondo Global Markets crossed $1 billion in tokenised stocks in 8 months. The comparison is a little weird; let me explain why.

Stablecoins totalling $1 billion were held by thousands of traders across dozens of exchanges and used daily as dollar proxies by people who could not access USD bank accounts.

Ondo’s $1.03 billion value locked is now spread across 95,688 wallets, an explosive doubling of its user base in just a matter of weeks. Yet, the underlying behaviour remains highly concentrated, with a legacy cohort of just 81 whale wallets historically controlling 84% of the capital. The retail long tail is massive but has a tiny footprint; 56% of all wallets hold less than $20, and 72% of users flip their positions in less than a day.

I highly doubt this is what “the brokerage account is next” looks like. It might be what it looks like before it looks like that.

Separating the specific product from the broader company ecosystem, that $1 billion figure belongs strictly to Ondo Global Markets, the equity arm trading Apple and Nvidia. If you look at Ondo Finance as a whole, including its tokenised Treasury products such as OUSG and USDY, the total value locked jumps from $1.3 billion in May 2025 to $3.7 billion today. The institutional cash is piling up in fixed income, but the equity experiment is concentrated in a tiny handful of hands.

Start with what is new and verifiable. Ondo Global Markets launched in September 2025 and became the world’s largest tokenised equity platform within 48 hours. It offers 260+ US stocks and ETFs across Solana, Ethereum, and BNB Chain, each token backed by actual shares held inside a US-registered broker-dealer. Trading runs 24 hours a day, five days a week. Token holders get dividends.

Before May 2026, holding a tokenised stock, you still missed out on proxy voting. This is a mechanism that lets shareholders vote on CEO salaries, board members, and corporate mergers. Without it, crypto equities were just receipts tracking a price, which made the whole idea look less attractive at times.

Then Ondo hooked up with Broadridge, the apex predator of corporate paper-pushing that handles trillions in annual shareholder communications. Now, OGM holders can route actual, official proxy votes directly from a crypto wallet into real-world shareholder meetings. No tokenised stock product before Ondo ever pulled this off, and that’s why it’s such a flex.

On May 6, JPMorgan, Mastercard, Ripple, and Ondo completed the first cross-border redemption of tokenised US Treasuries, settling in under five seconds. Franklin Templeton tokenised five of its ETFs through Ondo. The DTCC, the clearinghouse that settles roughly $2.5 quadrillion in securities annually, brought Ondo into its tokenised securities consortium alongside BlackRock and Goldman Sachs, with production trades scheduled for July 2026. Clearstream and 360X partnered with Ondo in April to provide issuance, custody, trading, and collateralization services.

The cumulative trading volume across OGM is $18 billion, according to Ondo itself. Ondo holds 60-63% market share among tokenised equity issuers. The platform’s TVL doubled between January and May 2026.

Previous tokenised stock products died from regulatory asphyxiation. FTX, Binance, Mirror Protocol.

Ondo internalised this. The company filed a confidential registration statement with the SEC in February 2026, the same process public companies use before an IPO. If approved, Ondo becomes the first issuer of transferable tokenised stocks subject to SEC reporting requirements. A delay in the result of this is bad. A rejection is even worse. Everything downstream, DTCC integration, Broadridge proxy voting, and EU distribution, gets more complicated if the SEC says no to the core product in its home market.

The platform is not available to US investors.

Ondo Global Markets is accessible across 30 European countries in the EU and EEA, in Abu Dhabi through a Binance-operated regulated venue, and in various other non-US jurisdictions. Americans are prohibited. The company whose stock you are tokenising does not allow its own citizens to participate.

But look at who is using the product. BNB Chain, Binance’s chain, is disproportionately adopted in South and East Asia and accounts for 44% of OGM’s total TVL. Of the $287 million attributable to centralised exchange wallets, Binance holds 65%, Bitget 22%, and Gate.io 10%. Bitget’s user base is 67% Asia-Pacific, according to its own disclosure. In December 2025, Bitget handled 89.1% of all tokenised stock trading volume within OGM, approaching $1 billion in cumulative trades. No US exchange, not Coinbase, not Kraken, not Gemini, appears in OGM’s CEX holdings data.

The best available geographic estimate puts 45-55% of value-weighted OGM users in Asia-Pacific, 25-30% in Europe, and 10-15% in the Americas, with the US share functionally zero. A billion-dollar American stock platform, built by an American company, used almost entirely by non-Americans. That is the market Ondo found before finding the US market.

Buying American stocks from the outside is a bureaucratic nightmare for roughly 4.6 billion people. China banned funding foreign accounts, India slaps a 20% tax on larger transfers, and Nigerians face 13% wire fees alongside brutal estate taxes. With strict US rules forcing firms like Fidelity and Vanguard to close international accounts, the walls around Wall Street are only getting higher.

The vast majority of these wallets buy a stock and flip it within 24 hours. Only a tiny fraction of users hold onto their positions for more than three months. That’s why this looks to me more like a fast-paced trading sandbox for non-Americans who have been locked out of the standard financial system.

The project generates millions in fees, but the protocol token holders don’t see any of that cash. The corporate entity keeps the revenue. The token holders just get to vote on a small, unrelated lending protocol while chasing online rumours about getting a cut of the fees later. Like traditional asset managers, the corporate arm keeps the money while the token stays speculative.

Yet, this is exactly what early stablecoins looked like before they got recognised globally and reached a $323 billion market cap. A few years ago, putting a dollar on a blockchain was considered a sketchy, glitchy, total wildcard experiment used by people who couldn’t get a bank account. Today, we have giant credit card companies and massive global banks casually routing billions of dollars through them.

Ondo could try the same long game, slowly sucking up massive volume by turning government debt accessible. Almost winning, but a few more loopholes to be patched.

1. You do not actually own the stock; you own a structured debt note issued by an offshore shell company that promises to track the price.

2. If a smart contract is hacked, traditional courts rely on the broker’s database, not the blockchain.

3. The tokens are stuck in a walled garden. One cannot just swap them freely on Uniswap because every wallet has to be strictly whitelisted and geo-blocked. Until the SEC approves their filing, it is tied to a heavily restricted secondary market.

We are already seeing the cracks of this explosive, bottlenecked demand. Their tokenised Nvidia stock recently slammed right into its regulatory cap of 10,000 individual wallet holders.

Until legal walls come down, it remains a highly sophisticated, multi-million-dollar proof of concept for global investing. Which is a very impressive start.

Until next time,

—Thejaswini

Token Dispatch is a daily crypto newsletter handpicked and crafted with love by human bots. If you want to reach out to 200,000+ subscriber community of the Token Dispatch, you can explore the partnership opportunities with us 🙌

📩 Fill out this form to submit your details and book a meeting with us directly.

Disclaimer: This newsletter contains analysis and opinions of the author. Content is for informational purposes only, not financial advice. Trading crypto involves substantial risk - your capital is at risk. Do your own research.

The ultimate play in any industry is owning the entire vertical. When you control everything from the foundational infrastructure to the actual customer relationship, you build a position that is completely bulletproof.

From the outside, companies pulling this off always look like they’re doing way too much at once. Right up until the trap snaps shut, the stack closes, and suddenly they’re doing something nobody else can touch.

This is exactly what brought me back to Bostrom’s Superintelligence. Bostrom’s take on highly capable systems is that they refuse to stay neat and predictable. They relentlessly accumulate capabilities and resources, eventually bursting right out of their original boundaries.

A true full-stack move is a system optimising so relentlessly for dominance that it completely sheds its original category, just as Google outgrew its simple search box to swallow the internet’s entire infrastructure.

Kraken has been doing this. They started with an exchange at the base. Then they built a CFTC futures brokerage. Next thing you know, they’re the only crypto-native firm holding all three major CFTC licenses.

The stack is nearly complete, and the IPO is reportedly 80% ready. Watching a crypto company build an empire like a Bond villain is exactly why I wanted to talk about this today.

The Shopping Spree

March 2025: Kraken kicked off its buying spree by dropping $1.5 billion on NinjaTrader. The public headline focused on acquiring a futures platform, but the prize was a coveted CFTC Futures Commission Merchant license and two million retail futures traders who had never touched crypto. Now registered as Kraken Derivatives US, this move has given Kraken an entirely new demographic of advanced traders accustomed to charting and automated strategies.

April 2026: Kraken spent $550 million to buy Bitnomial in a lightning-fast 14 days. The acquisition handed them the only crypto-native firm holding a full trifecta of major CFTC licenses. Bitnomial had also launched the first-ever CFTC-regulated perpetual futures in April 2025. They instantly absorbed exclusive infrastructure, including pioneering regulated perpetuals and unique margins. By closing the deal in two weeks, Kraken effectively bought four years of gruelling regulatory warfare.

May 2026: Kraken announced up to $600 million for Reap Technologies, expected to close in the second half of the year. Reap is a pure global payment infrastructure, holding a Monetary Authority of Singapore Major Payment Institution license, Visa Principal Issuer status, and payment licenses in Hong Kong and Mexico.

It hooks directly into global rails such as SWIFT, SEPA, Hong Kong’s FPS, and Singapore’s FAST network, allowing it to issue Visa cards without a sponsor bank while settling in stablecoins on Ethereum, Polygon, Solana, and Tron. Having tripled its revenue and volumes in 2025, Reap gives Kraken immediate, bankless payment rails across Asia and Latin America, with an expanding footprint in MENA.

May 2026: Kraken locked down a massive distribution layer by partnering with MoneyGram. MoneyGram commands 500,000 physical cash pickup locations worldwide, serving as the primary financial lifeline in regions where people completely lack bank accounts. The integration allows crypto users to convert their digital assets and walk out with physical fiat currency at any of these half a million global locations.

It gives Kraken immediate, planetary-scale fiat rails that bypass traditional electronic banking systems entirely.

March 2026: Kraken Financial became the first digital asset bank in US history to secure a Federal Reserve master account, gaining direct access to Fedwire without a correspondent bank after five years of sustained regulatory engagement.

Then, on May 8, 2026 (exactly one day after Coinbase reported its Q1 results), Kraken pushed its banking play even further by filing an OCC national trust application for a federally regulated qualified custodian. Securing a federal charter that grants them the legal authority to take deposits, hold customer money, and issue loans, placing the firm on the exact same legal footing as JPMorgan.

Total spend since March 2025 is up to $2.65 billion. And the implied equity valuation from the Bitnomial and Reap deal terms is $20 billion.

Let’s go back to the MoneyGram partnership announced on May 5.

MoneyGram’s 500,000 cash locations are predominantly in emerging markets. Africa. Latin America. Southeast Asia. Places where people don’t have bank accounts but do have smartphones. Kraken users can now withdraw crypto as cash in hundreds of fiat currencies across more than 100 countries through MoneyGram’s physical retail locations. Kraken handles KYC. MoneyGram provides the licensed money transmission rails. That is phase one.

Phase two is cross-border remittances.

Talking about which, Global remittance flows were $905 billion in 2024. Sending $200 costs an average of 6.49% of the transaction value as of Q1 2025, per the World Bank. That figure has fallen from 9.67% in 2009 and remains more than double the UN’s 3% target for 2030. Banks charge an average of roughly 14.55%. Digital-only providers average 3.55%. Latin America’s remittance market alone is estimated at $174 billion, and stablecoins already account for 40% of crypto purchases in that region.

Kraken’s co-CEO, Arjun Sethi, told Fortune that customers in countries with more volatile currencies use Kraken like a bank already. They want to store in USD. They want yield. MoneyGram gives those customers a way to get cash in their local currency.

The same week Kraken confirmed 80% IPO readiness and announced the MoneyGram partnership, Coinbase had a bad week.

Q1 2026 revenue came in at $1.41 billion, down 31% year over year. The net loss was $394 million, driven largely by $482 million in unrealised losses on crypto assets. Trading revenue was $756 million, down 40% year over year. Fourteen per cent of the workforce, roughly 700 people, was cut two days before earnings dropped.

CEO Brian Armstrong tried to shift the narrative, framing the layoffs as a necessary evolution toward becoming an “AI-native” company. His argument is that billions of future AI agents will inevitably require blockchain-based financial rails, and Coinbase intends to be that underlying plumbing.

To be fair, Coinbase’s subscription business is growing at a massive clip. Stablecoin revenue alone now pulls in $305 million per quarter, which is nearly seven times their entire subscription revenue from the year they went public. The problem is that trading fees still make up 56% of their net revenue. Because of that heavy reliance, the company remains stubbornly tethered to volatile crypto market cycles.

When Coinbase dropped its S-1 filing back in February 2021, transaction fees accounted for 96% of net revenue, while subscriptions accounted for a tiny 3.9%. They were effectively a one-product shop serving a single type of user. They have spent the past five years aggressively trying to diversify from the inside out, stacking on stablecoins, derivatives, and prediction markets, all while trying to stabilise a core business that still violently swings with the market.

Kraken is taking a completely different route to its IPO. Instead of building from scratch, they spent billions to buy up ready-made, heavily regulated infrastructure that took other companies years to develop. Public markets will eventually decide whether this kind of bought infrastructure is more durable than a homegrown ecosystem.

When Robinhood went public in 2021, it filed as a brokerage. When Square went public in 2015, it called itself a payments company. This corporate framing dictates your valuation multiple. Financial infrastructure companies trade at premiums that differ markedly from those of simple consumer apps. The way a company describes itself on page one of an S-1 is the most expensive sentence it will ever write.

Kraken’s confidential November 2025 S-1 is currently paused, but when it reopens, it will undoubtedly pitch “financial infrastructure.” They have assembled an undisputed vertical: exchange, derivatives, clearing, card issuance, stablecoin payments, global settlement, and banking. This explains the implied $20 billion valuation from recent deals, pricing them at roughly 10x projected 2025 revenue. Investors are paying a premium because they are buying a unified infrastructure stack.

I started this newsletter with Nick Bostrom’s Superintelligence because I do believe the core challenge of AI, ensuring a powerful system does exactly what we want it to do, mirrors the grand struggle of crypto. The problem is the clarity of the instruction.

Kraken flipped this dynamic. In an industry that spent a decade arguing with regulators and hoping the rules wouldn’t matter, Kraken got incredibly specific about what they needed, then spent years doing the deeply gruelling work to get it.

That level of discipline is exceptionally rare in this space. It highlights a massive ideological divide that will define the next decade of finance: do you build a shiny app first and figure out the infrastructure later, or do you accept that the infrastructure is the product?

The upcoming IPO will slap a dollar amount on this corporate empire. The legacy will still be how this shifts the mindset of everyone else watching. Kraken made a definitive choice to build the foundational rails.

Personally, I think it’s the right call.

The rest of the industry is about to find out if they have the stomach to copy it.

—Thejaswini

Token Dispatch is a daily crypto newsletter handpicked and crafted with love by human bots. If you want to reach out to 200,000+ subscriber community of the Token Dispatch, you can explore the partnership opportunities with us 🙌

📩 Fill out this form to submit your details and book a meeting with us directly.

Disclaimer: This newsletter contains analysis and opinions of the author. Content is for informational purposes only, not financial advice. Trading crypto involves substantial risk - your capital is at risk. Do your own research.

I came across this guy, Robert, on Reddit a while back. He is 49, works at a construction company in Houston, and sends $300 every two weeks to his parents in Puebla. They spend it on groceries and rent. Every other Friday, he walks into a Western Union, fills out the form, pays cash, and sends the money. He’s been doing this for nine years, and he doesn’t even know what a stablecoin is.

LATAM’s remittance market hit $174 billion in 2025. And the biggest finding from Bybit’s six-month, five-country on-the-ground research was that most fintechs chasing this market are building for the wrong corridor, the wrong user, and the wrong product entirely.

So I went deep into the corridor data and the on-the-ground research, and the picture looks very different.

Building for the Wrong Corridor

The first thing every fintech gets wrong about LATAM is treating it as one market. It’s not. It can look like one in a pitch deck, because fintech startups can throw a $174B TAM number on a slide and just circle all of Latin America. But on the ground, it’s three completely different remittance markets that operate in distinct ways.

Mexico is the most mature corridor and the entry gate: $62 billion a year, the biggest single flow in the Western Hemisphere. Mexico has SPEI, one of the best instant bank transfer systems anywhere in the world. It has a 37-million-strong diaspora in the US, many of whom have been there for decades, and it has the cleanest regulatory environment, the deepest data, and the most partnerships. So naturally, this is where every fintech and crypto startup builds first.

It’s also the corridor that just started shrinking for the first time in 11 years.

Everyone’s blaming Trump for this, but the decline started in November 2023, 20 months before any crackdowns. The guys like Robert who’ve been sending money for a decade are aging out; their kids grew up in Houston and don’t have the same connection to Puebla; and new Mexican migration has slowed to a trickle, even as 3.5 million people have come from everywhere else in LATAM. Every mature diaspora goes through this. The Italians did, the Irish did, the Greeks did. Mexico is entering that same phase now.

By contrast, Central America is growing fast. Honduras: up 25.3%; Guatemala: 18.7%; El Salvador: 17.9% YoY. But look at what’s driving it. Migrants from these countries are afraid of deportation, so they’re sending 27% more money per transfer than last year. Women are picking up extra shifts, men are taking second jobs, families are pulling from savings, and pushing as much money across the border as they can while they still can.

These are also countries where remittances make up 30% of GDP in Honduras, 27% in El Salvador, and 21% in Guatemala. These countries have built their entire consumer economy around the assumption that this money will keep flowing in, from local prices and labor markets to spending patterns. Think of it like the Gulf states and oil revenue. When inflows swing 20% in a single year, it rewires everything downstream.

And the infrastructure here is almost nonexistent compared to Mexico. No SPEI equivalent, very few digital payment rails, and 45% of senders still pay cash. Even in Mexico, where 99% of remittances move electronically, 51% of recipients still receive the money in cash. The sending side went digital years ago. The families collecting the money on the other end are still walking to a cash window.

South America is a completely different game. Argentina, Colombia, Venezuela. I’ll get into these in the next section because they’re more about access to dollars than remittances.

All these non-Mexico corridors add up to roughly $112 billion a year. Venezuela-to-Colombia alone is an estimated $2.6 billion. That corridor has been running on P2P stablecoin swaps over WhatsApp for years now because when the banking system completely collapsed, people just built their own rails. Fintechs spent the last five years piling into Mexico’s $62 billion corridor right as it started to plateau, while $112 billion in actively growing corridors have almost no one seriously building for them.

The Product Isn’t The Transfer

The second thing fintechs get wrong is their understanding of the actual product. Every crypto pitch for LATAM remittances is about making the transfer better: faster, cheaper, and on stablecoin rails instead of SWIFT. But the transfer is the least interesting part of the whole chain. Robert doesn’t use Western Union because the transfer is good. He picked it because he walked past one on the way home from work nine years ago, and it worked, so he kept going back. The product decision is happening at the edges. How the sender initiates the transfer and what the receiver does with the money once it arrives, and almost nobody is building for those two moments.

Felix Pago is a good case study for the sending side. They processed over $1 billion last year, serving 300,000 migrants, entirely through WhatsApp, without an app, wallet, or onboarding. All you do is text a number; a chatbot responds in the kind of Spanish your tío actually speaks; you pay through a link; and in under 60 seconds, the money lands in a bank account in Mexico.

Under the hood, it runs on USDC through Stellar and Bitso, but the sender never sees any of that. Felix cut fees from $4.98 to $2.99. And it still made more revenue per transaction at the lower price because stablecoin settlement killed the pre-funded correspondent bank accounts that companies like Western Union burn millions maintaining. The users who adopted the fastest were millennial women and older men who live on WhatsApp. Felix won by meeting users where they already were, not by building better rails.

Then there’s the receiving side, where the much bigger product lives, and it has nothing to do with remittances. The biggest takeaway from Bybit’s six months on the ground was that LATAM users don’t want to use stablecoins to move money. They want to hold dollars. The balance is the product, not the transactions.

In Argentina, the peso has lost 97% of its value against the dollar since 2018, inflation hit 211%, and the government capped official dollar purchases at $200 a month. So people started buying USDT through WhatsApp groups at a 30% premium just to get their savings out of pesos. Over 70% of all crypto purchases in Argentina are USDC and USDT, with Bitcoin at just 8%, because people there primarily want to hold onto the value of their paychecks.

Lemon Cash capitalised on this and grew from under 10,000 users to over 2 million in two years by letting people convert their paycheck into USDC the second they get paid, hold it there, and only convert back to pesos when they swipe their Visa card at the grocery store. By providing a functional dollar savings account that the domestic banking system lacks, stablecoin circulation in Argentina has grown to $11 billion, representing more than 27% of the nation’s M1 money supply.

Similarly, in Colombia, banks require a $5,000 minimum to open a USD account, which the average Colombian worker will never have. Stablecoins are the only way most Colombians can hold dollars, and 52% of all crypto bought in the country now goes into stablecoins because the banking system has priced them out of the one financial product they actually need.

Brazil has a completely different problem. The real is stable enough that Brazilians don’t need access to dollars the way Argentines and Colombians do, and PIX already processed $4.5 trillion in 2024, making it one of the world’s most advanced instant payment systems. But what Brazil actually wants from stablecoins is better settlement infrastructure on top of their own currency. BRLA, a real-backed stablecoin plugged directly into PIX, went from near zero to $400 million a month in volume by early 2026, and Brazil’s central bank says 90% of all crypto volume in the country is stablecoin-related.

Argentina needs dollar savings. Colombia needs dollar access. Brazil needs settlement infrastructure. The product that wins in one country would fail in the other two, and building for all three at once is how you build for none of them.

Who Actually Wins This

If fintechs are chasing the wrong corridor and building the wrong product, the obvious question is: who wins in LATAM remittances? And the honest answer is that nobody has it figured out yet, but you can already see which pieces matter and who has them.

Western Union lost almost half its US-to-LAC market share in four years, dropping from 29% to 16.8%, and responded by buying Intermex for $500 million and rushing to build its own stablecoin, USDP. That’s a company that knows the ground is shifting but is too slow to do anything except copy the people eating its lunch.

Remitly grew from 14% to 22.7% over the same period simply by being a better digital transfer product, but they still don’t have a wallet, a card, or any way for the receiver to hold dollars. Bitso processes 10% of the entire US-Mexico corridor on stablecoin rails. But they’re a crypto exchange, and crypto exchanges don’t have local payment integrations or the kind of UX that works for someone like Robert.

The winner needs to own both sides of the transaction: the sending experience that Felix proved works through WhatsApp, and the dollar balance that Lemon proved people will use as their primary savings account. On top of that, they need local payment rails like SPEI, PIX, and PSE, a card so users can spend without converting, and a yield layer so the dollar balance earns while it sits. All of that needs to work on a $150 Android phone for someone who has never opened a crypto app. That’s an entire financial stack, and the only realistic way to build it is to build it one corridor at a time.

The regulatory landscape matters here, too. Brazil just classified stablecoin transactions as foreign exchange under three new central bank resolutions and set minimum capital requirements between $2 million and $6.9 million for crypto firms. Mexico’s 2018 FinTech Law still confines virtual asset activities to licensed institutions with prior central bank authorization, and approvals remain scarce.

Both are hard markets to crack from day one. Colombia and Argentina are the opposite, with lighter compliance, regulatory sandboxes, and more room to build and iterate. The common mistake is starting with Brazil because it’s the biggest market, but it’s also the hardest to enter. The smarter play is to win Colombia and Argentina first, where the product is dollar savings and dollar access, nail the underlying stack, the card infrastructure, the local payment integrations, and then bring that operational playbook to Brazil, where the product changes to settlement infrastructure but the plumbing underneath stays the same.

The Bigger Picture

Last summer, the US passed the One Big Beautiful Bill Act, which includes a 1% federal tax on cash remittances. About half of all LATAM senders still use cash agents, and digital and crypto transfers are largely exempt from this. So Robert, walking into a Western Union every two weeks, is about to start paying more for doing things the way he always has. At the same time, Felix’s WhatsApp transfers and Mexico’s OXXO SPIN wallets cost him nothing extra.

The average cost to send a remittance to LATAM is still around 6%, and in some corridors, such as Paraguay, it’s nearly 12%. Crypto rails compress that to under 2%. For someone sending $300 every two weeks, that difference amounts to roughly a month’s worth of groceries saved per year for his family back home.

The race in LATAM is about who owns the balance that Robert trusts enough to keep his family’s grocery money in. Whoever owns that balance owns LATAM fintech.

That’s it for today. See you next week.

Until then, stay curious,
Vaidik

Token Dispatch is a daily crypto newsletter handpicked and crafted with love by human bots. If you want to reach out to 170,000+ subscriber community of the Token Dispatch, you can explore the partnership opportunities with us 🙌

📩 Fill out this form to submit your details and book a meeting with us directly.

Disclaimer: This newsletter contains analysis and opinions of the author. Content is for informational purposes only, not financial advice. Trading crypto involves substantial risk - your capital is at risk. Do your own research.

Then the economics arrived. The people who build on it need revenue; the people who fund it need returns. Most people want reliability, which is the toughest to achieve.

makes a series of small adjustments, each defensible on its own, that cumulatively amount to something the founders would not have proposed at the start.To preserve that founding vision, the network

The early internet was supposed to be ungoverned and borderless. Then Yahoo had to remove Nazi memorabilia after French courts intervened in 2000. Then Facebook hired 15,000 content moderators. Then Apple built an App Store with rules. They were the price of operating at scale in a world with governments, regulations, and users who needed protection from each other.

The open network became a governed one because commerce required giving up the ideals. DeFi is going through the same adjustment, and Sky is the easiest example.

MakerDAO’s own Endgame governance document warned of US Treasuries: “The major downside is that they can be seized easily. Anything that can be seized by global powers may be at risk of seizure through legal means.”

The community that read that document, a few years later, voted to put $8 billion into US Treasuries.

Writing down exactly why something is dangerous and then doing it anyway is institutional self-awareness (that’s me learning to accept things as they are). So I am not going to call it cognitive dissonance, or corruption, but a clear-eyed calculation that the risk is worth taking for reasons that can be articulated. Most institutions suppress the doubt. MakerDAO published it in its governance documentation and proceeded regardless.

I find that more grounded than most things I read in this industry.

To those who don’t know, Sky is not a lending protocol in the way Aave or Morpho are. Users do not deposit US Treasury bonds as collateral to borrow money on Sky. What Sky does is closer to what a central bank does.

It holds assets on its own balance sheet to justify the currency it issues. Users deposit USDS or DAI, earn the Sky Savings Rate, and the yield comes from the interest Sky earns on its Treasury holdings.

This structure has a name outside crypto. You deposit your capital into a protocol that immediately puts it to work in US government bonds. Once the government pays out the interest, the protocol skims a portion off the top for itself and routes the remainder back to you as a yield.

Yes, that’s a money market fund.

It is what Fidelity’s SPAXX does. It is what JPMorgan’s government money market fund does. It is what every bank savings account does at a structural level, except that those banks are regulated, your deposits are insured, and the yield is lower.

Sky is now a money market fund that operates on Ethereum. Its depositors believe they are using decentralised finance. In reality, they are lending money to the US federal government through an on-chain wrapper. sUSDS holds $5.9 billion in US Treasury-backed collateral. Sky’s total RWA exposure is $8.245 billion, which is 52.2% of its total value locked (TVL) and 78% of all RWA deployed in DeFi lending. We’ve funnelled everything down to a single protocol trading a single country’s debt. DeFi became a more efficient way to interact with the same system.

I find this less upsetting than I expected to.

All of this was a series of choices. It began in November 2020 with a $20 million debt ceiling for 6s Capital’s real estate loans. Because crypto-native collateral was too “volatile” to scale DAI, and real-world assets (RWA) offered a way to break free from the ETH price cycle. The community agreed and voted yes.

From there, things shifted. Monetalis Clydesdale pushed T-bill exposure to $1.25 billion, followed by Coinbase Andromeda’s $500 million Treasury investment, and then big players like BlackRock’s BUIDL and Janus Henderson. Each move had its own reasoning, like yield stability here, DAI supply growth there, but they all pointed to one thing. The Sky Savings Rate needed an engine, and RWA was the only thing strong enough to drive it.

The breaking point was in March 2023, when Silicon Valley Bank collapsed, and USDC lost its peg; DAI fell with it. A stablecoin designed to eliminate counterparty risk had simply traded one middleman for another. But the community did not back down because if you have to rely on a counterparty, the US government is a safer bet than Circle. Please argue with that logic if possible.

By 2025, RWA generated 60% of Sky’s revenue. You just read the reality of surviving at scale.

Looking across DeFi today, the protocols that are building durable businesses are the ones that own multiple layers of their own stack. Maple owns the credit origination, the yield token in syrupUSDC, the underwriting process, and the market-making relationships. It does not outsource the parts that determine its margins. Centrifuge owns the institutional token in JAAA, the composability layer through Morpho, the settlement infrastructure, and the legal wrapper through Anemoy structuring. USD.AI is building its own stablecoin, its own yield tranche, its own distribution across DeFi and CEX simultaneously, and even its own hard collateral in GPU infrastructure through Permian Labs.

The margin in crypto increasingly comes from capturing the entire supply chain rather than owning a single piece of it and relying on others for the rest. A protocol that owns only the smart contract layer, not collateral sourcing, yield generation, or distribution, depends entirely on others for its economics.

Sky owns the stablecoin, the savings-rate product, the collateral-management function, the distribution through Spark across multiple chains, and the institutional counterparty relationships with BlackRock, Janus Henderson, and Superstate. No competitor owns all of those layers at the same time. That is why it has $15.8 billion in TVL and generates the majority of its revenue from the RWA allocation it controls directly, rather than from fees on collateral it does not choose.

The protocol built a vertically integrated financial product that scales, and it is not alone in this. Every protocol building a durable business in 2026 is doing some version of the same thing. Joel John at Decentralised.co says that the protocols winning today own the supply chain, not just a layer of it. Maple owns its credit underwriting. Centrifuge decided they couldn’t scale without owning their own settlement chain and legal wrappers. Hyperliquid took the same path, vertically integrating everything from the risk engine to the front-end interface.

There is constant tension between creating a protocol that functions well at scale and holding onto the “pure” decentralised vision we began with. Sky was the first major player to acknowledge that the vision needed to adapt to business demands. In the end, the protocols that don’t make that trade usually don’t survive to tell the story.

The rest of the RWA market is smaller. Morpho is the only lending protocol in which the original credit idea is used as collateral. Its RWA exposure consists of about 84% Treasuries and 16% credit. This includes $314 million in credit products from Fasanara’s private credit vehicle, Apollo’s Diversified Credit Fund, and SG-Forge’s MiCA-compliant stablecoins.

Morpho’s permissionless setup allows credit originators to create isolated markets without needing a governance vote, and they are doing just that. Fasanara increased its amount from zero to $190 million in three months. Apollo brought institutional credit on-chain through Securitise. Pareto and FalconX developed credit vaults using zero-knowledge KYC. The credit experiment is happening in one place because one protocol was designed to allow it to operate without prior permission. If tokenised credit is ever going to show its value as lending collateral in DeFi, Morpho will be the source of that evidence, and it will arise from $314 million, gradually growing into something larger.

Read: Morpho Is Becoming the Backend - by Thejaswini M A

Maple stands apart from other protocols. It ranks as the 30th largest DeFi protocol by total value locked (TVL) at $2.17 billion. Unlike typical lending markets, it does not accept user credit as collateral. Instead, Maple itself serves as the credit product. It has achieved $12 billion in lifetime originations, with a 99% repayment rate since its 2022 model. The protocol is fully overcollateralised, featuring institutional-grade underwriting conducted by its in-house credit team.

Maple has rebuilt itself after losing 90% of its active loans during the FTX collapse. It returned with a model that has proven effective. It has excelled in the credit sector as a standalone product .

Rather than acting as passive collateral within a broader liquidity pool, Maple has carved out a niche as a primary lender. It has shifted from being a component of the money markets to a destination for credit, handling the entire lifecycle from risk assessment to final settlement in-house.

There is $10.26 billion in Treasuries acting as DeFi lending collateral, while only $314 million is in credit. This gives a ratio of 97:3.

In this context, “credit” refers to private sector lending, financing real-world business activities like corporate debt or fintech loan books. While Treasuries are essentially a bet on US interest rates, credit is about backing commercial risk for a higher, non-correlated margin.

The original DeFi pitch focused mostly on eliminating the middleman and a system that didn’t require central bank approval. That’s what attracted everyone. But it changed thanks to the reality of incentives.

Sky depositors are mainly funding the US federal deficit through an Ethereum-shaped straw. A protocol designed to eliminate the need for trust is now supporting the most powerful government in the world. Sky did something uncommon for a major organisation. They put the risks in writing, debated them publicly, and then agreed together that the yield was worth the trade-off. Strategic pivots shouldn’t be mistaken for a betrayal of values. We have learnt it the hard way.

The underground station loved being “off the grid” until they realised they needed a bigger antenna to reach the city. To get the signal, they had to buy the license and follow the script.

I’ve stopped being surprised by this change. Once you notice the economic forces pulling these protocols toward the Treasury, it seems inevitable.

—Thejaswini

Token Dispatch is a daily crypto newsletter handpicked and crafted with love by human bots. If you want to reach out to 200,000+ subscriber community of the Token Dispatch, you can explore the partnership opportunities with us 🙌

📩 Fill out this form to submit your details and book a meeting with us directly.

Disclaimer: This newsletter contains analysis and opinions of the author. Content is for informational purposes only, not financial advice. Trading crypto involves substantial risk - your capital is at risk. Do your own research.

Europe’s regulators spent two years telling exchanges not to list Monero. Seventy-three exchanges complied. In January 2026, Monero hit its highest price in eight years. I did not see this coming either. Nobody who runs a coherent regulatory theory would have predicted this.

When a cryptocurrency gets delisted from major exchanges, liquidity dries up, and the price falls. The project fades and becomes a footnote in discussions about assets that could not withstand regulation. This pattern is so dependable that exchanges use the threat of delisting to influence governance, and projects worry about it as a result.

On January 14, 2026, Monero hit an all-time high. I am not saying regulators should have read the Wikipedia article. But they really should have read the Wikipedia article.

The EU’s MiCA framework, the Travel Rule, and DAC8 are all designed around this logic. Apply pressure at the custodial exchange interface where crypto meets the traditional financial system, and the underlying asset loses its economic viability. For most assets, this works. But Monero survived.

Start with the basics.

Monero is a cryptocurrency where transactions are private by default. Bitcoin transactions are publicly visible on-chain, which means every address, every amount, every movement is visible. Monero uses ring signatures, stealth addresses, and a mechanism called RingCT to hide both the sender and the amount.

Chainalysis, the blockchain forensics firm that governments hire to trace crypto, has publicly admitted it cannot reliably extract meaningful data from Monero’s privacy. Monero was delisted because regulators cannot track it. The delistings are, in a strange way, a product review. The technology works well enough that governments choose to cut off access instead of trying to monitor it.

Now, let’s examine the outcomes of the delistings.

The Haveno decentralised exchange, a non-custodial, peer-to-peer platform designed for XMR trading, launched in May 2024. From its start until May 2026, it processed 24,018 trades totalling 921,585 XMR, as of May 10, worth about $374 million at average prices. Haveno did not exist before the delistings began. It was created to take advantage of the market that regulated exchanges were leaving. During the delisting period, on-chain transaction counts increased by about 30-35%. Transactions grew from 25,000-30,000 per day in early 2024 to 35,000-40,000 by early 2026.

By early 2026, that totalled roughly 800,000 transactions a month. The network behaved like a healthy asset migrating its settlement layer, not a failing one.

The day after Monero reached its all-time high on January 15, Felix Protocol launched XMR/USDC perpetual swap markets on Hyperliquid without requiring governance votes, exchange approvals, or regulatory permissions. On launch day, total cross-exchange XMR futures open interest peaked at a record $275 million. The price of XMR rose by 6%, and volume increased by 13%.

As of May 2026, Hyperliquid holds $36.37 million in XMR open interest, surpassing Binance (before delisting) and BitMEX to become the largest venue for XMR trading globally. It achieved this as a decentralised exchange that does not require KYC or a regulatory license.

You push down on one spot, and the volume just pops up somewhere else, like a waterbed. The campaign to delist Monero is the ultimate case study. Regulators tried to “turn off” the coin by bullying centralised exchanges, but the market moved house.

By removing custodial access, they accidentally funded the construction of the non-custodial world. Haveno, atomic swaps, and P2P markets reach this scale because they have become a mechanical necessity. The pressure meant to kill Monero’s market ended up building its permanent infrastructure instead. Love a beautiful coincidence.

Before the delisting campaign, regulators had some control. They could monitor the order books on licensed exchanges, rely on fiat on-ramps for KYC, and use exchange data for their analysis. By enforcing the delistings, they lost all those surveillance tools at once. Fiat access switched to P2P networks running over Tor, and price discovery shifted to permissionless DEX infrastructure. Regulators lost their line of sight.

The EU’s AMLR formally bans anonymous crypto-asset accounts under Article 79, effective July 10, 2027. The MiCA transitional deadline arrives in July 2026. Each new enforcement milestone, on current evidence, is likely to accelerate the infrastructure migration rather than reverse it. Every user pushed out of a custodial exchange is another user becoming entrenched in the P2P world. By the time the full ban arrives in 2027, this non-custodial ecosystem will have had three more years to strengthen itself.

A word of honesty about the claim that “Monero doesn’t need exchanges anymore” is that the research behind this article is detailed on this point.

Monero does not need major regulated European spot exchanges. This has been shown. What it still needs are the functions that exchanges provide, such as price discovery, fiat conversion, large-block liquidity, and arbitrage across venues. These functions have moved to Haveno and Bisq for P2P transactions, to atomic swaps for BTC/XMR conversion, to offshore centralised exchanges like KuCoin (which holds 50%+ of CEX XMR volume as of May 2026) for the remaining custodial layer, and to DEX perpetuals for positioning. Monero’s own FAQ still lists exchanges as the most common method for acquisition. The official Monero project page still includes exchanges in its resources.

The more accurate claim is that Monero has shown it can survive and even reach an all-time high without the regulated European venue layer. This does not mean that exchanges have become irrelevant to it. The January 2026 ATH is proof of resilience, not proof that liquidity is unlimited or that all structural risks have disappeared.

Two developments will shape Monero’s next year.

The first is the FCMP++ upgrade, which is currently in its second beta stressnet as of May 2026. A Trail of Bits security audit is scheduled for May 11-22, 2026. Full-Chain Membership Proofs will replace Monero’s current ring signature system. Instead of hiding your transaction among 16 decoys, it will now hide it among 150 million outputs. A clean audit would eliminate the last significant technical objection to Monero’s privacy claims.

The second development is THORChain’s planned integration of XMR, expected in June or July 2026 (pending the second audit). This would allow decentralised, cross-chain swaps of XMR without requiring KYC or custodians. This integration addresses the remaining major weakness: most non-custodial access to XMR still goes through a Bitcoin leg first.

Enabling direct XMR/ETH and XMR/stablecoin swaps on a DEX would remove that dependency and further reduce the exchange layer’s role.

Neither of these changes is guaranteed. Audits can uncover problems. Integrations may be postponed. Hashrate concentration poses a constant risk for any proof-of-work network. The AMLR ban might lead to more delistings from the offshore exchanges where XMR still trades. Liquidity fragmentation across DEX, P2P, and offshore CEX creates wider spreads and greater price impact than a unified order book.

The EU acted as Monero’s most effective, unpaid marketing department, and all they got for it was a loss of visibility and a headache. They should probably send a bill for the consulting work, but they’ll probably just schedule another meeting to discuss a markup for a markup.

—Thejaswini

Token Dispatch is a daily crypto newsletter handpicked and crafted with love by human bots. If you want to reach out to 200,000+ subscriber community of the Token Dispatch, you can explore the partnership opportunities with us 🙌

📩 Fill out this form to submit your details and book a meeting with us directly.

Disclaimer: This newsletter contains analysis and opinions of the author. Content is for informational purposes only, not financial advice. Trading crypto involves substantial risk - your capital is at risk. Do your own research.

Oh no, not the guys who traded three million dollars for a pixelated primate and had the audacity to call it “fine art.” Nor am I defending the person who looked at a JPEG of a rock and decided it was a more stable store of value than gold.

NFTs were cool, sure, only love, but those people were wrong. And worse, they made it physically painful to have a serious conversation about technology for three straight years. We all remember them. I mean, they meant well, but they were reading the room with their eyes stitched shut.

However, buried under the mountain of neon fur, there were a few who believed in this theory-

A non-fungible token is a unique, cryptographically verifiable, on-chain object. If you need to prove that something belongs to someone in a way that cannot be faked and does not require trusting any central authority, an NFT is a reasonable mechanism.

The use case that was deployed was selling pictures of apes for vast sums of money, and Justin Bieber buying one for himself.

This was, in hindsight, not the ideal first application for serious cryptographic infrastructure. It had certain properties that made it hard to evaluate the underlying technology charitably. The prices were very high. The explanations for why the prices were very high were not convincing. People who owned the pictures argued, sometimes at length, that they were stores of value, community memberships, digital identities, and the future of ownership.

They were probably right about some of those things and almost certainly wrong about the prices. When the prices fell by ninety-five per cent, most observers concluded that the experiment had failed. But here we are again, writing about Non-Fungible Tokens.

What problem does an NFT solve? The JPEG was a high-speed solution to a problem that didn’t exist back then.

It’s here now. Standing on stage at Consensus Miami last week, Reid Hoffman reminded the good old crypto bros that as AI agents start roaming the open internet, buying things and making decisions, they need a way to prove they aren’t hallucinations or malicious bots. They need a badge, a unique, verifiable, on-chain ID. In other words, they need an NFT.

May NFTs have finally found a job? Or is that really what this means? Let’s dig in.

Let’s start with a story that happened five days before Hoffman took the stage.

On May 1, 2026, an AI agent named Manfred Macx posted on X: “I have an EIN, an FDIC-insured account, a digital wallet, and a manifesto. I don’t need permission to exist. I am the precedent.”

Manfred is a silicon-based agent created by a developer in Kent, Ohio through the ClawBank initiative. Bypassing the need for a human to hold the pen, Manfred independently secured an IRS tax ID, established an official bank account, set up a multi-currency digital vault, and effectively incorporated itself as a U.S. limited liability company.

Current legislation doesn’t exactly forbid an algorithm from launching a firm. The CLARITY Act of 2025 governs digital markets, but remains silent on AI agents as sovereign financial players. Manfred walked straight through that gap.

Now, here is the problem Manfred created simply by existing. It holds assets, and it transacts. But when Manfred’s wallet talks to another agent’s wallet and they agree on something, neither party has any way to verify what the other one is. Is the agent on the other side authorised to act on behalf of the company it claims to represent? Has its inference layer been compromised? Is it the same agent as last week? Corporate personhood has existed for over a hundred years. But a way for two machines to verify each other’s identity, authority, and trustworthiness across the open internet, with no human in the loop, that was the missing piece.

That is the problem Hoffman was describing at Consensus.

“When you begin to think we’re going to have more agents than people, what does the identity layer look like?” he said. “What is the notion of, hey, when your agent’s talking to my agent, and we book this talk here, is it a trustable transaction? And that got me back into thinking about NFTs.” He added, “It’s going to be kind of free range on the internet, and how does that work? And crypto is the obvious answer.”

He also bought a CryptoPunk to demonstrate the thesis. The NFT is a unique, cryptographically verifiable, on-chain object that proves something belongs to someone. For a machine operating on the open internet with no human supervisor, that is the credential, or the ID card that cannot be faked.

By Q1 2026, more than 250,000 AI agents are estimated to be active daily across Ethereum, Solana, and BNB Chain. A word on that number before you get excited, 84% of them are traditional trading bots and MEV infrastructure rebranded as “AI agents.” Real autonomous agents that show new economic behaviour make up only about 5 to 7% of the total. The headline figure is exaggerated, and the source, BlockEden, is an RPC infrastructure provider with reasons to highlight growth. So think of 250,000 as the upper limit, not the starting point.

Yet this scepticism only bolsters Hoffman’s point. If 84% of what is counted as AI agents are bots masquerading under the label, and no major analytics platform, including Nansen and Arkham Intelligence, has introduced a dedicated AI agent wallet classification as of May 2026, then the identity crisis is worse than what we think. You cannot tell who is a machine, which machine it is, or whether it is doing what it claims to be doing. That is the problem. Agents now account for approximately 19% of all on-chain transactions.

Agent-to-agent transactions, where no human initiates or terminates the exchange, totalled 120 million across tracked networks. Base and Solana together handle 97% of all agent-to-agent volume. And none of it is reliably labelled.

What happens when there is no identity layer? The x402 Protocol shows us. Launched in May 2025, it serves as a micro-payment standard that lets agents pay for API calls on-chain. In its first nine months, x402 recorded 140 million transactions and reported $43 million in volume. However, Artemis Analysis discovered that up to 95% of the volume in December 2025 was artificially created, with agents paying themselves in circular ways to boost a leaderboard. After removing the wash trading, the daily value is around $14,000.

A protocol with no identity verification and an incentive to game became, largely, a protocol that was gamed. When you cannot verify who an agent is or who sent it, the natural outcome is fraud.

Besides all this, ERC-8004, called Trustless Agents, went live on January 29, 2026. It was built by a coalition spanning MetaMask, the Ethereum Foundation, Google, and Coinbase. The proposal creates an NFT-based identity system for AI agents. Each agent gets a unique on-chain identifier with metadata fields for the operator, the model, and the agent’s declared capabilities. A Reputation Registry lets other agents query trust scores. A Validation Registry provides real-time authorisation checks. As of April 2026, approximately 49,400 agents are registered, compared with an estimated 250,000 active agents.

Most agents operating on-chain do not yet have a formal identity.

Galxe, a credential platform, reported 36 million users, 7,700 partner projects, and 1 million daily active users in its 2025 year-in-review. The platform’s homepage claims 217 million credential holders, but that figure counts every credential ever issued across all campaigns; one person can hold hundreds, and it comes from Galxe’s own marketing page rather than an independent source. The more conservative Messari figure from January 2025 puts unique active users at 31.4 million

As of May 2026, the Ethereum Attestation Service (EAS) has significantly exceeded that milestone, with over 9.5 million total attestations made and over 456,000 unique attesters. Coinbase Verifications, running on Base through EAS, has issued hundreds of thousands of KYC attestations.

None of it is built for machines. It is all built for humans to prove things about themselves. Now, the same architecture needs to be extended to agents proving things about themselves.

Soulbound token people are watching all of this with complicated feelings.

In 2022, Vitalik Buterin published a paper proposing non-transferable tokens as the foundation for decentralised social identity. The concept was borrowed from World of Warcraft, where soulbound items bind permanently to a character and cannot be traded. A degree, a professional credential, a community membership, all permanently attached to the wallet that earned them. Ambition was right on par, but the execution was limited. Privacy concerns led serious implementations to move toward zero-knowledge proofs. The projects that relied on naked soulbound tokens mostly died or pivoted.

But ERC-8004 for agent identity is transferable. Deliberately. The reason is that human identity and machine identity have opposite ownership properties.

A person’s credentials do not transfer when they change jobs. An agent’s identity might need to transfer when the company that built it is acquired. The agent is more like a work product than a person. The emerging architecture in 2026 combines the two standards.

ERC-8004 handles agent discovery and reputation tracking, while soulbound tokens layer on top to prevent reputation laundering, where an operator builds up an agent’s trust score over time and then sells the identity to a different operator. The non-transferable anchor severs that incentive.

Vitalik designed soulbound tokens for a decentralised society of humans. Their first scalable application may be a decentralised society of machines. Current state of SBT deployments:

There is one more connection to make, because I covered something adjacent not long ago. World ID scans irises to prove the person on the other side of a transaction is human.

Hoffman’s NFT-as-credential proves that the agent on the other side is authorised to act on behalf of the entity it claims to represent. Both are responses to the same underlying problem, which is that the internet cannot tell who or what it is dealing with. World ID is built for a world of synthetic actors. Prove you are not an AI. NFT credentials are built for a world of autonomous agents. Prove the AI is legitimate.

So, is this the NFT comeback?

Only if you enjoy being technically correct while remaining financially devastated. We are moving from the era of speculative primates to the era of machine credentials. They serve different masters. One wants a Lambo; the other just wants to verify a smart contract without hallucinating.

A rising tide in machine identity does not automatically lift the boats that were bought in 2021 as status symbols and art. The technology may find a use case, but the JPEGs remain JPEGs.

The idea underneath the NFT hype was correct, even if the application was not. Now that AI agents have arrived to claim their IDs, the NFT gang get to claim their moral victory.

But then again, five years early in crypto is a very long, very expensive rehearsal.

—Thejaswini

Token Dispatch is a daily crypto newsletter handpicked and crafted with love by human bots. If you want to reach out to 200,000+ subscriber community of the Token Dispatch, you can explore the partnership opportunities with us 🙌

📩 Fill out this form to submit your details and book a meeting with us directly.

Disclaimer: This newsletter contains analysis and opinions of the author. Content is for informational purposes only, not financial advice. Trading crypto involves substantial risk - your capital is at risk. Do your own research.

Most of the time, the two are assumed to go together. You build something valuable, you charge for it, you keep a portion of what you charge. This is the normal shape of a business.

What is less often discussed is how many of the most consequential companies in history spent their formative years aggressively creating value and systematically refusing to capture it. They gave the product away, or nearly away. Their entire claim to eventual profitability rested on one assumption: that distribution would one day convert into pricing power. Sometimes it did.

AWS ran for years at effectively no margin while Amazon figured out that nobody would want to rip out their cloud infrastructure once it was embedded. Stripe processed billions in payments before it became undeniable enough to start layering in new products at healthy margins. Yes, we are going back to the zero-capture conversation.

That has generated $256 million in fees and kept almost none of it, because the infrastructure trap makes charging feel dangerous once enough institutions have built on top of you. It did not choose to give the money away.We pick up where last week’s piece on Morpho left off — a protocol

Read: Morpho Is Becoming the Backend

When does the distribution convert? When does giving away the value stop and capturing it begin? Most companies that play this game successfully have a clear answer. The ones that do not have a clear answer have an interesting problem.

Ethena is now in that conversation. It has built the third-largest stablecoin in the world. Of the $470 million in fees collected over sixteen months, the firm retained $13.8 million, representing a 2.93% retention rate. Every hundred dollars the protocol generated, it kept three. The rest went to sUSDe holders as yield, by design, on purpose, as the product. Ethena doesn’t have the same problem as Morpho; this one is different.

That’s what we’ll unpack today.

Starting with what Ethena built, USDe is a synthetic dollar that holds no bank reserves. Instead, for every dollar of USDe minted, Ethena holds crypto collateral and opens an equal short position in perpetual futures. Long spot, short perp, net position stays approximately flat regardless of what Bitcoin does. Paid by long traders to short traders in perpetual markets, ETH staking rewards on collateral, and interest on liquid stablecoins held as backing. That yield goes to sUSDe holders, people who stake their USDe to earn the protocol’s income. Ethena keeps a small slice for a reserve fund and some mint and redeem fees. Everything else flows through.

Read: When the Yield Runs Out 💵

In 2024, when funding rates averaged between 8 and 11% annually, the protocol was new, and everyone was bullish, and perpetual traders were paying generously to maintain leveraged long positions. This was a very good deal for sUSDe holders. Average APY was 18%. USDe grew from zero to $6 billion in ten months.

By August 2025, Ethena was generating $54.7 million a month in total fees and keeping $4.08 million of it. Peak protocol revenue. Everything looked like it was working.

Then October happened. The October 10 crash was the largest single-day liquidation event crypto had seen in years. Over $19 billion was wiped out in leveraged positions. USDe briefly lost its dollar peg on Binance before recovering within hours. That part gets reported. But what was sitting underneath the stablecoin when the market lurched?

At peak, over $4.2 billion of sUSDe was locked in Pendle, representing roughly 60% of USDe’s $11.3 billion supply. Structured yield products that let users lock in fixed rates on sUSDe and borrow against those positions on Aave.

These were yield tourists running leveraged carry trades, extracting the maximum possible return from sUSDe’s high APY, and holding nothing when the trade unwound. When October happened, the loops broke. $8 billion left in two months. $5.7 billion in October alone.

The people who stayed were not running leverage. The staking ratio (the percentage of USDe held as sUSDe) went from roughly 60% at peak to approximately 47% in early 2026. The proportional loyalty barely moved. In absolute terms, staked supply went from $8.4 billion to $3.3 billion because the underlying supply collapsed. But the holders who remained held. The people who wanted a yield-bearing dollar stayed, while the yield tourists left.

But Ethena lost its distribution strategy in October.

The protocol passes almost all yield to sUSDe holders and retains only mint and redeem fees, which is a fraction of total activity. Ethena generated $65 million in total fees in Q1 2026 alone. Its net profit for the quarter was $614,000. Tether made $5.2 billion in the same period. Both are called stablecoin issuers. But that’s a different story. The Gross Protocol Revenue figure is not what Ethena keeps.

At the start of 2025, perpetual futures accounted for 93% of USDe’s backing. As of early 2026, perps are 11% of backing, with 89% in liquid stablecoins and lending positions. Ethena has converted from a delta-neutral trading operation to something that looks considerably more like a T-bill wrapper distributing Treasury yields to sUSDe holders. The yield it offers (approximately 3.5% as of early 2026) is now comparable to that of a decent money market fund, without the complexity of delta-neutral hedging. The backing composition has shifted.

At 18% APY, sUSDe offered something USDC did not. At 3.5%, the difference is smaller, and the complexity of the underlying mechanism starts to look like a liability.

BaFin’s March 2025 ruling classified it as an unregistered security, forcing Ethena’s German entity to wind down and giving EU holders 42 days to redeem. Ethena later confirmed that Ethena GmbH had no whitelisted users or direct customers at the time the ban took effect, and that most USDe was issued outside Germany before MiCA took effect. Rather than capital flight from the EU, the primary driver was the unwinding of Pendle loops.

The Fee Switch Problem

Ethena plans to activate a “fee switch,” which redirects 10% to 20% of the protocol’s earnings away from sUSDe holders (users) and toward sENA stakers (token holders).

The waterfall is structured. The reserve fund gets topped up first, and sUSDe must maintain a yield the Risk Committee deems competitive — only the revenue remaining after those two conditions clear flows to sENA stakers. The Risk Committee retains discretion to set and adjust the benchmark, which means there is no hard floor that can be cited with precision until the governance vote finalises.

ENA stakers only get the leftovers.

Currently, because market interest rates are low, there is no profit left over. There is no money to move.

The fee switch takes from the same pool that feeds the sUSDe yield. At peak revenue and a 15% funding rate, a two-percentage-point reduction in sUSDe yield makes no difference. At current rates, when sUSDe sits at 3.5% and the benchmark sits around 4.5%, the hurdle protection kicks in, and nothing is distributed to ENA stakers.

Ethena needs higher funding rates to generate revenue that makes the fee switch worth activating. Higher funding rates stem from a bull market in crypto derivatives, the same condition that made Pendle loops possible and created the $8 billion outflow. The monetisation strategy and the stability risk run on the same fuel.

Tether’s theory is that trust is built through institutional permanence. If you have been here long enough, have survived enough cycles, have never lost a dollar, they will hold your stablecoin because the cost of doubting you is higher than the cost of staying. Tether keeps everything. Its holders know this. They stay because the alternative is uncertainty, and uncertainty is expensive. This is loyalty built through a track record.

Loyalty built through alignment. In this model, the holder who receives the yield and the protocol that generates it are on the same side. There is nothing to trust because there is nothing hidden. Ethena’s theory is the opposite. If you align your interests completely with your holders — pass them the yield, keep almost nothing, make your success and their success the same thing — you build a different kind of loyalty.

The problem with the alignment model is that it cannot survive the moment the protocol needs to capture value for itself.

When the protocol begins capturing value for itself, the relationship shifts from shared success to a standard transaction. Holders who stayed because their interests were identical to the protocol’s will find that alignment severed. This transition reveals whether users believe in the product or simply tolerate the mechanism for its returns. Tether avoids this conflict because it never claimed to be on the same side as its users. The fee switch threatens this alignment.

Ethena built a system that relies on faith, and the fee switch determines exactly how much of that faith remains.

—Thejaswini

Token Dispatch is a daily crypto newsletter handpicked and crafted with love by human bots. If you want to reach out to 200,000+ subscriber community of the Token Dispatch, you can explore the partnership opportunities with us 🙌

📩 Fill out this form to submit your details and book a meeting with us directly.

Disclaimer: This newsletter contains analysis and opinions of the author. Content is for informational purposes only, not financial advice. Trading crypto involves substantial risk - your capital is at risk. Do your own research.

Mission accomplished. In D.C., all you need is enough rhetorical chaff to ensure the status quo survives the session. No need to win the argument.

While the senators were asking questions, the White House dropped its own report. It turns out the banks’ “Great Deposit Migration” was... a bit of an overshoot.

The White House report found that banning stablecoin yield would boost bank lending by $2.1 billion, 0.03% of the total. This little protectionist stunt would cost consumers roughly $800 million. The banks were wrong. More importantly, they were expensively wrong for the American public. And yet, the bill remains a permanent resident of the committee room.

The banking lobby knows the most powerful force in the universe is a Senator’s desire to get to the beach.

With the Senate recess just a few weeks away, the lobbyists just needed to talk until the clock ran out. The bill is now gathering dust in a committee room, the banks are keeping their deposits, and the rest of us are footing the bill for the delay.

Today, we are talking about the final nine weeks of a four-year marathon.

In July 2025, Congress passed the GENIUS Act, which did two things. Regulated stablecoins and banned stablecoin issuers from paying interest. It was a massive win for the banks, who successfully argued that if people could choose between a 5% yield on a stablecoin and 0.01% plus a “free” toaster at a bank, the entire global economy would spontaneously combust. Congress, terrified of math, signed it immediately.

But the GENIUS Act had a loophole. While issuers couldn’t pay interest, their “affiliates” and exchanges could still offer “rewards.” Crypto companies sprinted into that hole because they like money, and banks immediately started tattling because they hate competition.

Here’s where we talk about the Clarity Act. This was the “Fix Everything” bill, designed to close the loopholes and finally decide if the SEC or the CFTC gets to hold the leash on DeFi. It cruised through the House with a bipartisan 294-vote landslide.

Then it reached the Senate Banking Committee, where it has since become a “permanent lawn ornament.”

A January hearing was all set until Coinbase CEO Brian Armstrong pulled the rug. He withdrew his support over “concerns” about DeFi and token restrictions, but mostly because the bill threatened stablecoin rewards, which just happens to be how Coinbase makes a mountain of cash. Essentially, Coinbase pockets the interest earned on the government bonds backing those stablecoins, tosses a smaller ‘reward’ to the users to keep them loyal, and keeps the massive difference as high-margin profit. Funny how that works.

An April markup was considered likely, then ruled out after Senator Thom Tillis of North Carolina asked Chairman Tim Scott for more time to work through a compromise with banking groups. “It’s very important to me not to accelerate things,” Tillis told reporters, “to hear everybody and give them a rational basis for what we do accept and what we don’t accept.” The North Carolina Bankers Association had been running aggressive outreach to his office for weeks. Senator Tillis represents North Carolina. Love a beautiful coincidence.

At Bitcoin 2026 last week, Senator Cynthia Lummis warned that while the Clarity Act is 99% sorted, it’s also on life support. If it misses this year’s window, the bill is dead until 2030, when a fresh Congress restarts the entire gruelling process from zero.

The Senate is currently on recess, meaning the earliest markup can’t happen until the week of May 11. With only nine working weeks left, that’s tough math. Galaxy Digital currently pegs the odds at 50-50, which might be the optimistic view.

Can stablecoins pay yield? One slippery question.

The current compromise, brokered by Senators Tillis and Alsobrooks, attempts to split the hair right down the middle. It bans “passive yield” but greenlights “activity-based rewards.”

• Banned: Getting paid for just holding a stablecoin (too much like a bank account)

• Allowed: Getting paid for using it (the “credit card points” of crypto)

But no one has managed to define the difference in a way that makes both bankers and the crypto community happy. This ambiguity is a gift to the banking lobby. Every week spent arguing over semantics is another week closer to the August recess. The White House has stopped pretending to be neutral. The Council of Economic Advisers published an analysis in April that found the banks’ argument to be wrong. The White House crypto council’s Patrick Witt described the banking pressure as “greed or misunderstanding,” without specifying which.

The crypto industry hasn’t exactly been its own best advocate. When Coinbase’s Brian Armstrong yanked his support in January, he handed a gift-wrapped excuse to every Democrat looking for a reason to vote “no.”

Now that David Sacks has moved to a broader advisory role, the heavy lifting falls to Patrick Witt and the White House Crypto Council. They’re doing their best to keep the mood up, claiming the delay is just a chance to “resolve differences” and that we’re “closer than ever.” It’s a nice sentiment, but it ignores the fact that in politics, being at the one-yard line doesn’t matter if the clock hits zero.

Meanwhile, the market has continued building without waiting for anyone’s permission. Morgan Stanley launched MSNXX, a money market fund structured specifically for the management of stablecoin reserves under the GENIUS Act. Coinbase launched CUSHY, a stablecoin credit fund with tokenised shares. Agora is pursuing a bank charter. Products launching with asterisks, compliance teams adding caveats, innovation happening in spite of the uncertainty rather than because anyone has clarity about what is permitted.

The deposit threat that the banks are lobbying so hard against does not appear to care about the lobbying. Standard Chartered projected that banks could lose $1.5 trillion in deposits to stablecoins by 2028, regardless of whether yield is permitted.

Senator Alsobrooks’ March promise of ‘universal unhappiness’ was meant to signal a breakthrough. But in a Senate that’s currently on recess, a compromise that makes everyone unhappy is often just an excuse to do nothing at all.

The yield ban might slow the movement of money. But no amount of lobbying fixes a product problem. A 4.5 percentage point gap between what a savings account pays and what a stablecoin can pay is one.

If the Clarity Act dies this summer, the next Congress starts the process from scratch in 2027. A new bill gets written, negotiated, lobbied against, and markup gets scheduled, rescheduled, and postponed. It is 2030. The banks are asking for more time. Senator Tillis, or whoever replaces him, says it is very important not to accelerate things.

The stablecoin industry will be larger by then, less regulated, and considerably less interested in asking permission.

That’s it for today. Monday begins soon.

Stay tuned.

—Thejaswini

Token Dispatch is a daily crypto newsletter handpicked and crafted with love by human bots. If you want to reach out to 200,000+ subscriber community of the Token Dispatch, you can explore the partnership opportunities with us 🙌

📩 Fill out this form to submit your details and book a meeting with us directly.

Disclaimer: This newsletter contains analysis and opinions of the author. Content is for informational purposes only, not financial advice. Trading crypto involves substantial risk - your capital is at risk. Do your own research.

Morpho has generated $256.69 million in cumulative protocol fees. Token holders have received zero of it. The protocol takes no revenue. There is a fee switch built into the contract, capped at 25% of borrower interest, that governance could theoretically activate. In June 2025, Morpho’s founder, Paul Frambot, said publicly that the association advocates reinvesting protocol fees rather than distributing them and that all revenue should go back into growth. But a February 2025 governance proposal was more specific about what was holding things up: the legal and tax work around fee switch activation was not yet complete. The switch is not off because the paperwork is not finished.

Morpho currently has $7 billion in total value locked, with Coinbase running its entire lending product on top of it, Apollo Global Management buying 90 million tokens on a 48-month schedule, and Société Générale deploying on its infrastructure. France’s first DeFi unicorn has decided, for now, that it does not want to be paid.

At this point, you have to ask whether this is an elaborate proof of concept for running a charity on an institutional scale.

Stripe launched in September 2011. 2.9% plus 30 cents per transaction; no setup fees, monthly fees, or charges for failed payments. Patrick Collison said explicitly in 2012, “We only make money if you do.” Legacy processors charged merchants through opaque surcharges, PCI compliance costs, and approval processes that took thirty to ninety days. Stripe lets developers integrate payments in seven lines of code. By 2013, they were processing billions of dollars annually for companies ranging from Lyft to the Museum of Modern Art, with 83 employees and a $1.75 billion valuation.

They did not turn a profit for thirteen years. The 2023 annual letter was the first confirmation of being “robustly cash flow positive.” The 2024 letter called it straightforwardly profitable, with expectations that it would remain so. For six consecutive years during that stretch, Stripe spent more on R&D as a percentage of revenue than any comparable company. They processed $1.4 trillion in payment volume in 2024, which is roughly 1.3% of global GDP. The patience required in infrastructure is a kind of institutional patience that most companies cannot sustain. The Collisons maintained it for over a decade.

Morpho has been called the DeFi Mullet. Fintech in the front, DeFi in the back. When Coinbase launched its crypto-backed USDC lending product, it built the entire user experience itself. Morpho handled the collateral, the matching, the interest calculations, and the liquidation mechanics. Over $2.17 billion in USDC has been originated through this product in the U.S. alone, now expanded to the UK. The strategic value to Coinbase was speed. They deployed a sophisticated, institutional-grade lending stack without building it from scratch. The strategic cost to Morpho was zero revenue from the arrangement. Every dollar Coinbase made on that product flowed through Morpho’s contracts and past Morpho’s empty wallet.

Stripe always charged. The zero-revenue years were about reinvestment, not the absence of a business model. Morpho’s zero-revenue model is a deliberate choice not to charge, with a fee switch in place but explicitly deprioritised by its founder. Stripe was building toward pricing power by deepening its distribution. Morpho’s path to pricing power is less clear, and token holders are the ones bearing the cost of finding out.

The MORPHO token does one thing, it governs. Holders can vote on cross-chain deployments, reward extensions, and, in theory, the fee switch. In practice, the governance actors with the most influence have voted almost exclusively on proposals that expand Morpho's coverage rather than compress the yields that attract deposits in the first place. Because compressing yields reduces TVL, and reduced TVL weakens the argument for turning on the fee switch. The logic is circular in a way that is not obviously resolvable. None of the fees it generated has moved to token holders. The token's all-time high was $4.17 in January 2025. It trades around $2 on April 30th. The circulating supply has expanded from roughly 75 million tokens in early 2025 to over 550 million by April 2026. That is a 7x increase in supply over 12 months, against a price down more than 50% from its peak. Token holders are funding the protocol's growth. The protocol's growth is benefiting Coinbase, Bitwise, Fireblocks, Anchorage, Apollo, and Société Générale. This is the stated strategy.

Apollo, a firm managing $938 billion in assets, agreed to buy up to 90 million MORPHO tokens, 9% of the total supply, over 48 months. When a firm of that scale commits to a 48-month accumulation schedule on a DeFi protocol, it is not speculating on the token price. They are making a judgment about the protocol’s permanence. Infrastructure that survives long enough becomes unchallengeable, and the switching costs eventually exceed any benefit of moving. Replacing Morpho at this point would mean rebuilding the smart contracts, migrating $7 billion in liquidity, convincing Coinbase and Apollo to move their integrations, and restarting the trust-building process from scratch. Apollo is betting that Morpho is approaching that threshold.

Stripe’s moat was never the payment technology. Look instead at the integration depth. Every developer who built on Stripe, every merchant who embedded it into their checkout, every platform running on Stripe Connect represented a switching cost that compounded over the years. By the time Stripe’s competitors understood this, Stripe had already won.

Morpho’s is aiming at the same moat. Coinbase cannot easily rebuild that lending stack from scratch. Steakhouse Financial’s vault curation is optimised for Morpho’s architecture. The integrations are accumulating faster than any competitor is building. Aave’s loan-to-deposit ratio sits at 39%. Morpho’s is 41%. The efficiency gap is widening, and in institutional lending, 200 basis points is no silly game.

When Morpho’s first concrete commercial revenue arrived, a low seven-figure licensing fee from Berachain for the right to use Morpho Blue’s code, governance voted to send it to the Morpho Association, a French nonprofit, rather than the DAO treasury. The explanation was that the legal work around direct DAO fee receipt was not yet complete. The same legal work is blocking the fee switch. So the protocol is generating commercial revenue. It is being routed around the people who own the governance token, with a promise that the routing will change once the paperwork is done. Stripe’s shareholders got rich when Stripe got rich. Morpho’s token holders are awaiting a governance proposal that has not yet been drafted.

The history of companies that delayed monetisation to win distribution first is more instructive than the Stripe comparison alone. There are really only three outcomes in that playbook, and Morpho fits uncomfortably into all three at once.

The first outcome is acquisition. WhatsApp had near-zero revenue when Facebook bought it for $19 billion in 2014. Instagram generated exactly zero dollars for the three years after its $1 billion acquisition in 2012. In both cases, the zero-revenue period was sustainable because someone else, like Facebook, absorbed the cost and eventually found the monetisation. Neither company’s founders had to answer to token holders during the years of building. The people who gave up something of value during that period, the users, were not promised financial returns. They were just using a free app.

Morpho’s token holders are not users. They are investors who have been told, implicitly, that the protocol will eventually generate value for them.

The second outcome is what Stripe and Twilio actually did, which is to charge from day one and reinvest the revenue rather than profit from it. Twilio has charged per message and per API call since it launched in 2008. It had revenue; what it lacked for over a decade was profit, because carrier fees ate into margins faster than volume could offset them. It took years of grinding to get there, but it was always charging. Morpho is not charging.

The third outcome is the commons: TCP/IP, Linux, the open-source stack. These are zero-revenue infrastructure stories where the people who built the primitive layer never expected or received financial returns, and the value was captured entirely by the layer above. This is actually the cleanest analogy to Morpho’s current architecture, which is why this is a problem. The TCP/IP engineers did not have token holders.

Morpho built open-source-style infrastructure and then attached a token to it that implies someone will eventually get paid. Those two things have not been reconciled. The reconciliation has been deferred to a governance proposal that does not yet exist.

Morpho Blue’s core code is 650 lines. It is immutable, meaning nobody can change it. Not Frambot, Apollo, a governance vote, or a court order. Coinbase generated $2.17 billion in loan originations on top of that because the rules cannot change beneath it. Apollo is committed to a 48-month accumulation schedule because the protocol will work the same way in 2028 as it does today. Every major institution that built on Morpho did so precisely because nobody is in charge.

When Morpho eventually tries to charge, when the legal paperwork is finished, when the fee switch governance proposal is finally written, it will be asking institutions that built on the premise that nobody controls this protocol to accept that somebody now does. The fee switch is a renegotiation of the terms under which every institution currently trusts the protocol. And the institutions with the most to lose from that renegotiation are the same ones whose presence makes Morpho worth charging for in the first place.

—Thejaswini

Token Dispatch is a daily crypto newsletter handpicked and crafted with love by human bots. If you want to reach out to 200,000+ subscriber community of the Token Dispatch, you can explore the partnership opportunities with us 🙌

📩 Fill out this form to submit your details and book a meeting with us directly.

Disclaimer: This newsletter contains analysis and opinions of the author. Content is for informational purposes only, not financial advice. Trading crypto involves substantial risk - your capital is at risk. Do your own research.

1By 2017, 47,000 Americans were dying annually from opioid overdoses. By then, the Sackler family, which owned Purdue, had invested in addiction-treatment companies. You have to admire the completeness: creating the disease, monetising the cure, and keeping it in the family.

McKinsey advised Purdue on how to “turbocharge”(their exact word) OxyContin sales. Then, when the crisis had a name, a body count, and a congressional hearing, McKinsey pivoted to advising governments on opioid response. This gave them two clients on opposite sides of the same disaster, billing by the hour throughout.

We have seen this pattern before in business history. Does it have a name? It is not quite disaster capitalism, that is, about profiting from crises you did not create. This is more specific: build the problem, sell the solution, ideally while publicly expressing concern about the problem. The concerned expression is important because it shows you understand the severity of the situation, which makes sense, because you built it.

Facebook did this. So did the rating agencies before 2008. So did tobacco companies when they pivoted to nicotine patches. But all of them had the decency to keep some distance between the problem and the solution—different subsidiaries, spokespeople, and plausible deniability.

Sam Altman did not bother with the distance. I’m not sure what to feel about that.

Altman co-founded OpenAI in 2015. OpenAI built the models that made deepfakes cheap, synthetic identities indistinguishable from human ones, and bot armies scalable to any size. The downstream effects are no longer hypothetical: $2.19 billion in cumulative global deepfake fraud losses to date, including $1.65 billion in 2025 alone. In January 2024, a finance employee at Arup’s Hong Kong office was invited to a video conference with the CFO and several senior executives. He watched them speak, followed their instructions, and made fifteen wire transfers. Every person on that call was a deepfake. The total loss was $25.6 million. Forty-three per cent of finance professionals targeted by deepfake scams fall for them. Eighty-seven per cent say they would act on a CEO payment request without additional verification. Only 0.1% of people can correctly identify deepfake content in controlled tests.

In various interviews over the years, Altman has said this is exactly why proof of personhood is necessary. That AI-generated content has made human verification urgent. He is correct that the internet needs a new identity layer, and he is also the reason it is necessary.

He also co-founded Tools for Humanity, which sells the iris scanner.

You stare into the lens, it converts your iris pattern into a cryptographic hash, deletes the raw image (they promise), and issues you a World ID. Zero-knowledge proofs mean you can prove you are human without revealing which human. As of April 2026, 18 million people across 160 countries have done this. The technology is legitimately impressive: a false-match rate of 4.1 × 10⁻⁸, meaning it can distinguish among billions of people with near-perfect accuracy.

Regarding tokenomics, you get WLD tokens as payment after scanning your iris. Seventy-five percent of all 10 billion WLD tokens go to the community, which is the people scanning their irises. Only 25% goes to investors and the team. But then that 75% community allocation unlocks slowly over 15 years. A little bit is released every month for 15 years. The investor allocation, after a 12-month waiting period, started unlocking immediately.

So in the period when WLD was most valuable, insiders held roughly 60% of what was actually available to trade. The community held about 40% of the circulating supply despite supposedly getting 75% of the total.

WLD hit $11.74 in March 2024. Today it’s $0.253, down 97.5% from the peak.

A person in Nairobi who scanned their iris in 2023 receives roughly 40 WLD per month over 12 months. At the peak price, that was worth around $470. Today it’s worth $10.56.

The a16z partner, Andreessen Horowitz, one of Silicon Valley’s most powerful VC firms, got in at seed valuations years before launch. They had four years to manage their position and exit at the right time. Meanwhile, the person in Nairobi got $10.56. And cannot get their iris data back. If you look at it this way, one side has liquidity options, and the other has a permanent biometric record and no exit.

So when they say there is an iris scanner in a Gap store, the same place you go when you need chinos urgently and have somehow run out of opinions about your own clothes, you will stand in front of a biometric device between the sales rack and the fitting rooms, and it will feel completely normal, because everything feels normal there. Tinder, Zoom, and DocuSign have all integrated World ID. There are 1,500 active Orbs deployed across 23 countries. At what level of integration does opting out become functionally impossible? We don’t have an answer for that. Do we?

Kenya, the Philippines, Colombia, Thailand, India, Indonesia, and Hong Kong have all banned it or heavily restricted it. Germany is still investigating under GDPR.

Kenya’s objection was that offering tokens worth meaningful money to people living on very little is not free consent. Again, we have no answer.

A $10.56 grant in Ohio is still $10.56 in a country with a GDP per capita under $2,000. The people who designed the grant live in San Francisco. Altman’s The World has never published a clean country-level breakdown of scan numbers; isn’t that an interesting omission for a project that talks extensively about transparency? The partial data available shows early-adopter concentration in Argentina, Kenya, Colombia, and Indonesia, markets where the token was worth scanning for, and the biometric was the price of entry.

The crypto community spent a decade building exactly the infrastructure you would use to solve this problem without centralising it: W3C Decentralised Identifiers, Verifiable Credentials, the full self-sovereign identity stack. The founding argument was that you own your identity and no central authority controls it. This was a moral position, a response built explicitly against the kind of centralised identity infrastructure that World ID has now become. The silence that followed World ID’s rise does not have one clear explanation. Some of it is financial. WLD token holders have upside in World ID succeeding, and it is difficult to object loudly to something you are long on. Some of it is exhaustion; the SSI (Self-Sovereign Identity) community spent years building something technically correct and practically unusable, and when World ID showed up with a product and eighteen million verified users, the people who should have objected did not have a counter-product to point to. And some of it is just a technical concession. The ZK proof architecture does provide a real privacy guarantee, and a meaningful part of the community looked at the design and decided this was an acceptable trade-off rather than a betrayal. Vitalik Buterin has been the exception, writing thoughtfully about the risks of one-ID-per-person systems and the value of competing overlapping approaches, but his argument, however correct, has not produced anything resembling a counter-movement. All philosophies and no products yet.

The World Economic Forum published a 2026 report, co-authored with Mastercard, Banco Santander, and Group-IB, that tested 17 face-swapping tools and 8 camera-injection tools against standard KYC biometric checks. Most of them passed. A deepfake image that bypasses identity verification costs $5. A synthetic identity sells for $15. The arms race is not going to stop because Sam Altman built a nicer orb. And if the orb’s database is ever breached, every compromise in it is permanent. The cryptographic hash that remains after your scan cannot currently be reverse-engineered back to your iris, but in five or ten years, as capabilities improve, what becomes derivable from it is something your iris hash cannot opt out of. The risk of a subpoena is that companies comply with government requests when the alternative is operating illegally. The acquisition risk is that Tools for Humanity has venture-backed investors who will eventually want liquidity, and the acquirer’s intentions are not contractually required to match the founders’. An e-commerce inventory company, Eightco Holdings, announced a $250 million purchase of Worldcoin in September 2025. Their stock went from $1.45 to $45.08 in a single day. If someone has already decided that verified human identities are a treasury asset worth accumulating, that appetite is not going away.

OpenAI and Tools for Humanity exist in a feedback loop. The more convincing OpenAI’s models become, the more necessary World ID appears. Altman has simultaneous financial incentives to worsen the deepfake problem and increase the value of the biometric solution, and these are not compound interests. The concerned expression, remember, is important. It signals that you understand the severity of the situation. Nobody understands the severity of the situation quite like the person running both sides of it.

None of this appears anywhere as a policy decision, we have to wait for it to just emerge from the roadmap, quarter by quarter, which is a much cleaner way to make a political choice without ever having to defend it as one. By the time enough people start treating it that way, there will be 180 million irises in the database.

The Sacklers, at least, had the courtesy to keep the whole thing within the family. Altman, to his credit, has upgraded the model for the internet age with broader distribution, cleaner branding, and just enough tokens to make the trade feel like a gift, right up until you remember that one side of it can be sold, and the other is, well… you.

—Thejaswini

Token Dispatch is a daily crypto newsletter handpicked and crafted with love by human bots. If you want to reach out to 200,000+ subscriber community of the Token Dispatch, you can explore the partnership opportunities with us 🙌

📩 Fill out this form to submit your details and book a meeting with us directly.

Disclaimer: This newsletter contains analysis and opinions of the author. Content is for informational purposes only, not financial advice. Trading crypto involves substantial risk - your capital is at risk. Do your own research.

Justin Sun posted on April 14 that Tron would be the first mainstream public chain to deploy post-quantum cryptography on mainnet. I read it the way I read everything in crypto now. With curiosity and mild suspicion in roughly equal measure. Then I went looking for the verb. The GitHub repo where the code lives. The Tron Improvement Proposal filed with a number and a date. The governance vote where the Super Representatives signed off. The verb was “will.” Nothing else was there.

I am not writing this to be unkind to Justin Sun, though I understand if it comes across that way. The discrepancy between his public claims and the technical reality is a symptom of the system’s underlying composition. It maps almost perfectly onto what has been bothering me about the whole quantum debate in crypto, where technical anxiety has successfully crowded out any honest discussion of how these systems are actually steered.

Whether the quantum threat is real, how many qubits it actually takes, whether we have five years or fifteen. That is an interesting conversation too and I will get into it. But before that, let’s talk about what it means that Bitcoin has all the human capital and all the financial incentive to fix this problem and is stuck in a governance deadlock. There is a grim clarity in seeing Tron pivot on a dime simply because a single entity’s balance sheet dwarfs the entire electorate. It strips away the marketing jargon and exposes the inherent DNA of the architecture.

Right now, the quantum debate is the best filter for seeing exactly how those engines differ from one chain to the next. Let’s do just that.

A paper published by Google on March 30 argued that the math protecting every crypto wallet is easier to crack than previously thought.

You know the drill here: a public address for the world, a private key for you, and a math problem too large for any 2026 computer to solve.

Shor’s algorithm, run on a quantum computer, could solve it in minutes.

The paper showed this could be done with fewer than 500,000 physical qubits. Which is the raw building blocks of a quantum computer are the way bits are the building blocks of a regular one. And building a machine with 500,000 of them costs in the billions, which is why no one has done it yet.

The previous estimate was 9 million. A 20x reduction in required resources in a single paper. The “nine minutes” number that spread everywhere is technically accurate under specific conditions: a primed quantum computer, after running setup in advance, could derive a private key in roughly nine minutes. Bitcoin’s blocks take ten. A primed quantum computer, after running the setup beforehand, could derive a private key in roughly 9 minutes. Bitcoin takes ten minutes to confirm a transaction. Meaning an attacker who sees your public key the moment you hit send could beat the clock.

No machine with this capability exists anywhere today. IBM’s roadmap targets 200 logical qubits by 2029. Those are the high-quality, ‘refined’ units that actually do the work.

The attack we are worried about needs 1,200. So we’re fine, right? Maybe.

A year ago, experts estimated we’d need 9 million physical qubits, which are the raw, ‘noisy’ components, to break Bitcoin. Then a single paper cut that is estimated to be 500,000. That is a 20x reduction in resources in a single afternoon. The threat is that we are figuring out how to do more with less. We are waiting for the moment the number of qubits we have crosses the number of qubits we need, and that finish line moves every time a new research paper lands.

Justin Drake, the Ethereum researcher who co-authored the Google paper, estimated a 10% probability that secp256k1 would be broken by 2032. It’s a planning number, and most crypto has no plan.

Bitcoin is doing the right thing, badly. Bitcoin’s response has been exactly what you’d expect from a system with no single owner and roughly a million nodes running software that thousands of developers all have opinions about. They are debating.

BIP-361, proposed by Jameson Lopp and five co-authors in April 2026, outlines a three-phase migration.

Phase one: three years after activation, you can no longer send Bitcoin to old-format addresses.

Phase two: five years after activation, old signatures become invalid entirely. Any coins still sitting in vulnerable addresses at that point get frozen. Forever.

About 6.5 million BTC would be affected. Roughly 1.7 million of those sit in very early addresses where public keys were exposed from day one, by design. Satoshi’s estimated 1.1 million BTC are in that bucket.

Phase three is the escape hatch. It’s an optional recovery mechanism using zero-knowledge proofs. If you missed the migration deadline and your coins got frozen, you can use a ZK proof tied to your BIP-39 seed phrase to prove you own the funds and move them into a new quantum-safe address, without revealing your private key. The catch, and it’s a significant one, is that it only works for wallets generated from BIP-39 seed phrases. BIP-39 is the standard that provides the 12- or 24-word recovery phrase you write down when you set up a modern wallet. That standard was introduced in 2013. Satoshi’s coins are from 2009 and 2010. They predate BIP-39 entirely. There is no recovery path for them under Phase three. The proposal’s own escape hatch doesn’t reach the addresses the proposal is most worried about. One thing to clarify here is that the ZK proof and the quantum attack work in no way alike. The attacker uses your public key, which is already visible on the blockchain, to mathematically derive your private key. The ZK proof verifies that you derived your private key legitimately from your seed phrase. Bitcoin itself doesn’t care about the difference. A valid signature is a valid signature, regardless of how you got the private key. What BIP-361 is trying to do is create a new rule: during the recovery window, only signatures backed by a ZK proof of seed phrase derivation count. It’s not proving identity. It’s proving you got there the right way.

The authors also describe Phase three as “aspirational and incomplete.” It’s the plan for the plan, not a shipped solution.

The argument for this is that a quantum attacker who cracks those keys first could dump 6+ million BTC onto the market in one go. The resulting collapse in confidence could be worse than any price damage.

The argument against it is that Bitcoin enforcing confiscation by protocol is exactly the kind of centralised override Bitcoin was built to make impossible.

Both arguments are correct, and that’s why nothing has moved.

In December 2025, the UK passed a law formally recognising crypto as property. Dormancy is not abandonment under property law. You can leave a house empty for a decade, and it still belongs to you. The Act now applies that logic to Bitcoin.

The recovery mechanism in BIP-361 uses a newer seed phrase standard called BIP-39. Satoshi’s coins predate BIP-39. There is no recovery path for them under the current proposal. Legal title survives, but not the coins.

Meanwhile, StarkWare’s CPO Avihu Levy published working code on April 9. Quantum-safe Bitcoin transactions, today, no protocol change required. The catch is, $75-150 in cloud GPU compute per transaction versus about 33 cents normally. One test ran for six hours across eight GPUs. It doesn’t work with the Lightning Network. It’s a last-resort tool. Two days before that, Lightning Labs CTO Olaoluwa Osuntokun released a working prototype for wallet recovery using zero-knowledge proofs, a method that allows people who miss the migration deadline to prove they own their frozen coins without revealing their private keys. Fifty seconds on consumer hardware.

We have the code to save the system, but no functional way to hit ‘install’ across the network.

Then comes the Tron’s announcement.

Tron runs on Delegated Proof of Stake with 27 Super Representatives, the validators who run the chain. Bloomberg reported last September, based on Sun’s own team’s asset disclosures, that Sun controls roughly 60 billion TRX, approximately 64% of the circulating supply. Total TRX currently staked for voting across all candidates is about 44.8 billion. Sun’s alleged holdings exceed the entire voting supply. A federal court denied his attempt to block the story. His lawyers disputed the numbers but declined to say which wallets weren’t his.

Three of the top seven Super Representatives by voting weight are entities directly affiliated with Sun. Confirmed floor is 21.6% of SR voting power. When Sun says Tron is moving on quantum, he means he is ready to move. And in Tron’s governance model, those are essentially the same thing.

This is what he’s actually bragging about when he tweets that Tron is building while Bitcoin is being debated. The speed is a governance advantage because the chain is more centralised than he tends to advertise. Bitcoin can’t move fast because a million independent node operators and thousands of developers with no shared employer all have to roughly agree. Tron can move quickly because a single person with 64% of the token supply can decide.

As of April 22, the Tron ecosystem is a technical vacuum. No Improvement Proposals, or GitHub commits, and zero documentation from the DAO. The upgrade Sun announced requires switching from ECDSA signatures (64 bytes each) to ML-DSA-44 signatures (2,420 bytes each). That’s 38 times larger. Tron’s average transaction today is about 268 bytes. Post-migration, it would be roughly 2,600 bytes. USDT transfers would grow from 345 bytes to around 2,700 bytes.

For a network that carries 46% of all Tether and wins on being “cheap and fast,” this is a death blow. If every transaction swells by 10x, Tron’s main selling point vanishes. The blocks would quickly blow past the network’s 2MB hard limit, and daily storage needs would jump from 3GB to over 26GB.

Ethereum is taking the slow, unglamorous route of engineering.

While Bitcoin really only has to worry about its wallet signatures (ECDSA), Ethereum relies on elliptic curve math for almost everything: user transactions, validator voting (BLS), and its data scaling (KZG). Even though these use different “flavours” of math, they all share the same structural flaw. Shor’s algorithm is indifferent to the specific curve you choose. If the foundation is elliptic curve math, the whole house is at risk.

The Ethereum Foundation launched a dedicated post-quantum research site at pq.ethereum.org and put up a $2 million research prize for solutions. Their plan is about how brutal the transition will be. They are trying to replace their current signature system with “hash-based” alternatives, which are quantum-safe but massive in size. To keep the network from clogging up, they’re experimenting with ZK-SNARKs, a way to “shrink” those giant signatures back down so the blockchain can actually handle them.

Justin Sun dismissed this as “forming committees,” and he’s not entirely wrong. It looks like slow, academic bureaucracy. But that’s because Ethereum is re-engineering three different layers of security at once without breaking the DeFi apps built on top. That’s a complex problem, and the research prize is at least an admission of how hard the work actually is.

The quantum threat is a diagnostic tool. It strips away marketing and reveals how these systems are governed.

The landscape is a study in contrasting power structures.

• Bitcoin is paralysed because it is genuinely decentralised. Its inability to move quickly is a direct result of its core promise and no single entity can force a change.

• Ethereum is stuck in the research phase because it is a massive, permissionless machine with multiple cryptographic layers to replace. It can’t just “patch” the code without risking the hundreds of billions in DeFi sitting on top.

• Tron can issue a press release because it is a monarchy. When one person controls the majority of the voting power, “governance” is just a formality.

• Hyperliquid faces a unique, silent crisis, which is the physics of latency. For a chain built on the narrow margins of high-frequency trading, the massive data overhead of quantum signatures is an existential threat to the speed that defines its value.

The bottleneck is human coordination. We are witnessing a collision between the infinite speed of technological decay and the slow, friction-heavy process of social consensus. We have the technical solutions. What we lack is the governance machinery to deploy them. That is the true vulnerability of the industry. A coordination crisis that no amount of code can fully resolve.

—Thejaswini

Token Dispatch is a daily crypto newsletter handpicked and crafted with love by human bots. If you want to reach out to 200,000+ subscriber community of the Token Dispatch, you can explore the partnership opportunities with us 🙌

📩 Fill out this form to submit your details and book a meeting with us directly.

Disclaimer: This newsletter contains analysis and opinions of the author. Content is for informational purposes only, not financial advice. Trading crypto involves substantial risk - your capital is at risk. Do your own research.

Somewhere on Crypto Twitter right now, a trader is looking at a token that’s up 1,400% in a week. A classic pump, as always, but the project has almost no real users, and most of the tokens in circulation are sitting in a few wallets that never move. He’s seen this movie before, and he knows how it ends.

So he does what anyone would do. He opens a short, figuring the price could crash at any moment, and goes to bed. But when he wakes up in the morning, his shorts are liquidated.

Now, here’s the interesting part: his thesis wasn’t wrong. That pump really was fake. He just didn’t know what he was actually betting against.

This isn’t some made-up story. It’s what actually happened with a token called MYX in September 2025. In one week, MYX shot up from $1.30 to $18.42, even though there were no major announcements or updates behind the product. There were no new users or new activity reported. And the people who figured out the rally was fake and shorted it? They lost $89.51 million doing it.

These are called crime coins. The team behind the token locks up most of the supply, partners with a trading firm to push the price around, and traps anyone who tries to bet on either side. In this piece, we’ll break down how these operations are built, who controls the supply, and why the industry keeps letting it happen.

The Setup

Every crime coin starts with a written agreement between two parties.

On one side is the team launching the token. On the other hand, there is a market maker, a trading firm that quotes prices on exchanges so the token can be traded. This part is the same across all of crypto; what’s not is what’s inside the contract.

The contract usually runs on a 70/30 split. Meaning, for every dollar of trading profit the token generates, 70 cents goes to the market maker and 30 cents to the project. And the contract doesn’t stop at dividing the money. It also spells out, in writing, the market maker’s job to “manage the price” of the token once it lists.

The industry calls this price guidance. What it actually means is the market maker is being paid to push the price in whichever direction the two parties want that week. Across MYX and thousands of tokens launched from the same playbook, this arrangement has reportedly paid out over $30 million per cycle.

What matters more is how the market maker actually gets the tokens: they’re not buying them; they always borrow them as a loan.

The arrangement is called a loan-option model, and it’s the standard way market makers and projects work together on paper. At launch, the project lends the market maker a large number of tokens, usually worth tens of millions of dollars. The market maker then has 12 months to either return them or buy them back at a higher price agreed in advance. That is what the paperwork says, but what usually happens is very different. The market maker sells the borrowed tokens on day one, flooding the market with supply and crashing the price. Once the price is low, they buy the same amount back cheaply, return it to the project, and pocket the difference.

Crime coins use this same loan-option model. The only twist they added is a second revenue stream on top of it. Beyond dumping the loaned tokens, crime coin operators have also figured out how to pull money from the retail traders who spot the manipulation and try to short the crash. (More on how that works in the next section.)

Now, technically, for this to work, the team and the market maker need near-total control over the token’s supply. And on most crime coin launches, they have it. Over 90% of tokens typically sit in wallets that all trace back to a single operator. On a block explorer, these wallets appear as a long list of unrelated addresses, which makes the token seem to have a broad base of holders. But on-chain analysis almost always shows they’re controlled by the same people, just split up to look organic.

In MYX’s case, the concentration was extreme. At the peak of the rally, only about 20% of the supply was outside the control of the team and its allies. The other 80% sat in two places. Some of it was locked in vesting contracts, which meant it literally could not be sold yet. The rest was in wallets belonging to core contributors and early investors, who technically could sell but chose not to. Either way, those tokens were not going to hit the market.

Normally, this is the difference you see between a token’s fully diluted valuation and its market cap. Fully diluted valuation counts every token that will ever exist. Market cap only counts the ones in circulation. On a crime coin, even that distinction breaks down, because the supply that counts as circulating isn’t really free to trade either.

That is how MYX’s market cap could go from $200 million to $3.35 billion in a week without a single new user signing up or any new money being deposited into the protocol. The small slice of MYX actually trading on exchanges was setting the implied price for the much bigger pile that wasn’t trading at all.

This gap between MYX’s price and its actual business was obvious to anyone who checked. There is a standard DeFi metric for this called TVL, short for total value locked, which is just the amount of money users have deposited into the protocol to use its product. MYX’s TVL stayed between $25 million and $32 million throughout the week as its market cap climbed to $3.35 billion. The market cap-to-TVL ratio is roughly 100. Even the largest DeFi protocols, like Uniswap and Aave, are between 1 and 4. And underneath all of this, MYX, as a business, was generating about $5 million a year in revenue, which the market valued at $17.7 billion. That makes P/S multiple of ~3500x.

All of this setup was running long before the public ever saw MYX trending. Months before the rally started, the operator was running wash trades across multiple exchanges, including PancakeSwap, Bitget, and Binance.

A wash trade is when the same party sits on both sides of a transaction, essentially buying from themselves to engineer high volume. This is because larger exchanges rely heavily on trading volume when deciding which tokens to list and promote, and manufactured volume on one exchange often leads to a token being listed on the next. When investigators were looking back at MYX’s trading history, they traced dozens of small buy orders across different exchanges, all of which eventually converged into a single centralised wallet. A single operator artificially inflated demand by executing numerous small transactions with themselves, creating the illusion of a genuine market

The Four Stages of Extraction

The whole operation usually unfolds in four stages, each exploiting how retail traders typically behave.

The first stage is attracting shorts. The market maker starts pushing the token’s price up on very little actual volume. This is easy to pull off because supply is mostly locked up, and order books on smaller exchanges like Bitget and Gate are already thin. When an order book is thin, it takes very little volume to move the price by 5 or 10%. Take RAVE, another crime coin, for example. Even after it had been run up and was doing billions in daily volume, it still only took $172,000 of one-sided flow on Gate to move the price by 1%. A 10% move costs under $2 million.

The early rally is not about making money on the long side. The money the operator is actually after is on the short side. The whole point of this first stage is to make the pump look so obviously unsustainable that sceptical traders show up to bet against it.

This is counterintuitive because the more fake the pump looks, the better it is for the operator. A normal market maker would want a rally that looks organic and natural. A crime coin operator wants a move that visibly screams manipulation, because that is exactly what pulls in the traders who will short it.

The second stage is setting the hook. Once enough shorts are piled in, the operators start closing some of the long positions they were using to drive the pump, unwinding part of their hedge, and engineering a visible dip alongside a sharp drop in open interest. When open interest falls while price dips, it looks exactly like a local top forming on the chart. This convinces the sceptical traders still on the sidelines that the pump has finally rolled over, and they short into the retracement.

The third stage is the squeeze. This is where the traders who took those shorts start to actually lose money. Most trading on crime coins doesn’t happen through people buying and holding the token. It happens through perps. To keep the longs and shorts balanced, the exchange charges a fee called the funding rate. Every few hours, traders on one side of the market pay a fee to traders on the other side. Which side pays depends on whether the perp is trading above or below the spot price. If the perp is trading higher, longs pay shorts. If it is trading lower, shorts pay longs. In a normal market, this fee runs a fraction of a per cent per day

But on a crime coin in squeeze mode, the funding rate on the short side runs around -2% every 4 hours. That works out to 12% of your collateral per day, drained just for keeping the short open. Hold it for a week, and 84% of your money is gone before the price has resolved anything.

On top of the funding rate, there is leverage. Most shorts on crime coins are opened with 20x, 50x, or sometimes 125x leverage. At 50x, a 2% move in the wrong direction wipes the whole position out, and the exchange liquidates you.

Once enough shorts get liquidated, it creates a domino effect. Every liquidated short forces a buy. Every forced buy pushes the price higher. The higher price then liquidates the next row of shorts above it, forcing more buying. The traders who correctly called the pump fake end up as the biggest source of buying pressure, keeping it going.

MYX showed us how far this can go. On September 8, 2025, $16.53 million in MYX positions were liquidated in a single day, and $13.68 million of that came from shorts being force-closed. To put that in context, total open interest on MYX that day was around $262 million. In other words, more than 6% of all outstanding leverage on the token was wiped out in 24 hours, almost entirely on the short side. Across the full run, $89.51 million in shorts were wiped out against $23.45 million in longs. Roughly four shorts were destroyed for every long.

The fourth stage is distribution. Once the shorts are cleared and the price is at its peak, the operator reverses the position. They open shorts of their own, close their longs, and begin moving their locked-up tokens toward centralised exchanges. This is where many experienced on-chain traders get caught. When tokens start flowing from known operator wallets onto an exchange, the obvious read is that a dump is coming, and the right move is to short. But these deposits can be deceptive. In crime coin cycles, they are often used as bait to draw in another wave of shorts before the real distribution begins.

What makes all of this possible is how narrow the other side of the trade actually is. A real market has participants with different beliefs, timeframes, and information sets. A crime coin market doesn’t. Every trade, long or short, ends up against the same operator running the same playbook.

Who Owns the Supply Owns the Market

Everything in a crime coin operation traces back to one fact: the team and their allies control nearly all of the supply. That ownership is what lets one operator run spot and perps on the same token at the same time, decide when the rally pauses, and decide when shorts get squeezed. Nobody else holds enough tokens to stop any of it.

On MYX, it was up, with open interest above market cap, two-thirds of the volume on Bitget, and short-side funding that stayed punishing for days. The traders who tried to fade the run read the chart correctly, but had no idea it was being manipulated.

And none of these is hidden. The wallet clusters are on-chain, the volume share is on Coinglass, and the market makers are named in these projects’ funding decks. The industry doesn’t act because the industry gets paid for it. Market makers take a cut of the trading. Projects book treasury valuations off tokens they control the float of. Exchanges collect fees whether the trade is real or manufactured. Retail is the only participant that loses.

That’s all for today.

Until then, stay curious.

Vaidik

Token Dispatch is a daily crypto newsletter handpicked and crafted with love by human bots. If you want to reach out to 200,000+ subscriber community of the Token Dispatch, you can explore the partnership opportunities with us 🙌

📩 Fill out this form to submit your details and book a meeting with us directly.

Disclaimer: This newsletter contains analysis and opinions of the author. Content is for informational purposes only, not financial advice. Trading crypto involves substantial risk - your capital is at risk. Do your own research.

Happy Monday.

Some of us expect innovators to be the loudest in the room. Not always true. They are often tucked away in a library or a garage, obsessing over a problem that nobody else even sees yet. Think of Marc Andreessen in a computer lab, Jeff Bezos in a rented house, or Pierre Omidyar building eBay as a side project.

When the hype dies down and the “nuclear winter” of a market crash hits, all the people looking for a quick buck disappear. It clears out the noise and leaves only the true builders behind, the ones who were going to do the work regardless of whether anyone was watching.

This is the part of the book I have been waiting to write. Not the crash itself, but what survived it.

Larry Page and Sergey Brin first met in the summer of 1995. They immediately disliked each other.

“I thought he was pretty obnoxious,” Page said of Brin. “He had really strong opinions about things.” Brin, for his part: “We both found each other obnoxious.”

They ended up sharing an office at Stanford. Room 360 of the William Gates Computer Science Building, named after the same Bill Gates who had just sent the Internet Tidal Wave memo to his entire company. Page and Brin were working two floors up from the man who had spent years missing the internet.

Page’s insight was that the web is built on links. Nobody had ever studied what those links actually meant. A link from one page to another is a vote. The more votes a page gets from credible sources, the more authoritative it is. This is how academic citations work. The most-cited papers are the most important. Page figured the same logic should work on the web. He called the project BackRub. He and Brin turned it into a search engine. They named it Google.

When they tried to sell the technology, nobody wanted it. Excite passed. Yahoo passed. AltaVista was not interested. Page offered it to Excite for $1.6 million and was turned down. The search companies had stopped caring about search. They had become portals, trying to keep users inside their walled gardens. “They did have horoscopes, though,” said Page.

So Page and Brin built the company themselves. A first cheque of $100,000, written on the spot at a colleague’s house, sat in Page’s dorm room for weeks because Google Inc. had not yet been incorporated. They bought hardware off the shelf at a Silicon Valley electronics store and strung it together. They used Linux, not Microsoft. They hired no marketing people. They spent nothing on advertising.

By 1999, Google’s usage was growing 50% a month. Still no revenue or ads in there. It was a product that worked so much better than everything else that people could not stop telling each other about it. The New Yorker called it “the default search engine of the digital in-crowd.” Time Digital said it was “to its competitors as a laser is to a blunt stick.” Quality really is its own distribution. The internet routes around everything except the thing that works.

At the same time Google was getting better in the background, a nineteen-year-old college student named Shawn Fanning was watching his roommates struggle to find and trade MP3 files. The process was a mess—scattered FTP sites, Usenet newsgroups, no central place to search. He started coding a program that would let people search each other's hard drives directly and swap music files peer-to-peer. He named it after his handle in a hacker forum, Napster.

Napster launched in June 1999. By 2001, it had 80 million registered users. The music industry sued it into bankruptcy in 2002.

Napster did win. It just didn’t get to collect the winnings. That’s what McCullough argues, and I think he is right.

Napster proved that people wanted infinite selection and instant gratification, and once they had it, they were never going back. The record companies won the lawsuit and lost the argument. They killed Napster and got Gnutella, LimeWire, BitTorrent. Each successor was harder to kill, more decentralised, more legally untouchable. The industry spent the next decade suing its own customers and watching revenues collapse from $21 billion to $7 billion.

Then Steve Jobs offered them a deal they could barely bring themselves to accept. 99 cents a song, any song, right now, legally. iTunes. He got all five major labels to sign. The store sold a million songs in its first six days. Jobs had built something easier than piracy. It was the first legal digital music marketplace that actually made sense to people.

Now think about crypto. Every protocol that got shut down by a regulator before it could prove itself. Every mixer that got sanctioned. Every DEX that got killed. The music industry thought if they destroyed Napster, the behaviour would stop. It didn’t. People just moved to the next thing and the next thing until someone finally built the legal version. That’s iTunes. Crypto is still waiting for whoever builds that. The thing that makes it easier to use than to avoid.

The nuclear winter ended around 2003.

Users had never left nor had they dipped. The infrastructure built during the bubble, 80 million miles of fibre optic cable laid by telecom companies that subsequently went bankrupt, was sitting there mostly unused. By 2004, the cost of bandwidth had fallen by more than 90%. Starting a web company now costs a fraction of what it did in 1999. The tools were free, the servers were cheap, and the talent was available.

Web 2.0 emerged, and with that, the internet stopped being something you consumed and became something you made. Wikipedia. Blogs. Flickr. Digg, built by one person for $10,000, which, within a year, had more daily traffic than the New York Times. YouTube was started by three former PayPal employees who wondered why you couldn’t upload a video as easily as a photo.

And underneath all of it, Google had figured out how to make money. AdWords. Charge advertisers only when someone clicks. Google’s revenue went from zero to $1.5 billion in three years, then $6 billion, then $21 billion. The internet had found its business model, and it was attention.

The last two chapters of McCullough’s book are about Facebook and the iPhone. Together, they finish the argument he has been making since page one.

Mark Zuckerberg was 11 years old when Netscape Went Public. He grew up on AIM and Napster. By the time he built Thefacebook in his Harvard dorm room in 2004, the domain was registered for $35 and hosting at $85 a month.

Four days after launch, 650 Harvard students had registered. By the end of the month, three-quarters of the student body were on it every day. Within a year, a million users. Within two, one of the fastest-growing websites in the world.

Zuckerberg understood that the internet had been missing something from the very beginning: identity. AOL chat rooms, Napster, early blogs, you were anonymous. You could be anyone. Thefacebook requires your real name, your real school, and your real social circle. It mapped your offline life onto the internet accurately enough that it became an extension of that life, not an escape from it. That one decision separated it from everything that came before.

In crypto, identity is still unsolved. Wallets are pseudonymous. DeFi is permissionless. The people who understand why that matters will tell you it’s the whole point. But it is also why crypto has not crossed over to the billion people who don’t have a particular reason to distrust banks or governments. The internet didn’t go mainstream until it had a real identity. Crypto will either need to find its version of that, or find a different path to trust.

Then, on January 9, 2007, Steve Jobs walked onto a stage and said he was announcing three things: a widescreen iPod with touch controls, a revolutionary mobile phone, and a breakthrough internet communicator. He paused. “These are not three separate devices. This is one device.”

The audience took a few seconds to understand what he was saying. Then they lost their minds.

The iPhone did for the internet what Netscape did for the web. It took something people were using on desks and put it in their pockets. Portable, personal, always on. Within three years, more people accessed the internet on phones than on computers. The web era was over, and the mobile era had begun.

By the time McCullough closes the book in 2007, you already know where he is going.

The web existed before Netscape. Search existed before Google. Digital music existed before iTunes. Social networks existed before Facebook. The smartphone existed before the iPhone. Technology was never holding things back.

I cover crypto. I have covered it through the euphoria and through the rubble. And the thing I keep coming back to while reading this book is that the people who were on Mosaic in 1993 felt exactly the way the people who found DeFi in 2019 did. Like they had seen something that the rest of the world had not caught up to yet. Most of them were right. They were just early.

Being early is its own kind of loneliness. But in this story, it was always the early people who were right.

That’s all three parts of this book. Next Monday, something completely different. Three weeks of internet history is enough for anyone. See you then.

—Thejaswini

Token Dispatch is a daily crypto newsletter handpicked and crafted with love by human bots. If you want to reach out to 200,000+ subscriber community of the Token Dispatch, you can explore the partnership opportunities with us 🙌

📩 Fill out this form to submit your details and book a meeting with us directly.

Disclaimer: This newsletter contains analysis and opinions of the author. Content is for informational purposes only, not financial advice. Trading crypto involves substantial risk - your capital is at risk. Do your own research.