This way, you can put anything on-chain, but also, you can’t. We are not there yet; the relationship is still complicated.

A tin of tuna in a Lisbon supermarket now carries a code you can scan to see the boat that caught the fish, the date, the ocean, the whole line back to the water. But if the person unloading the boat mislabels the catch, the whole system fails. Tracking only works after the data is entered. The entire process depends on one human making an honest record, and we underfund this critical first step.

Similarly, in the case of blockchain chickens, it worked because it wasn’t a tradeable asset - the whole idea was an immutable provenance record. To prove that it was genuinely premium free-range poultry after years of food safety scandals.

Nobody should be asking whether you can put an asset on-chain. You can. But the ability to verify real-world facts is what tells you what a good and bad tokenisation project.

That’s why we should look at the environmental markets; that’s where I saw identical technology going two ways. One is to record verified truths, and the other to record claims that could be totally fabricated.

Renewable energy certificates work on a blockchain, and carbon credits do not. As the cost of addressing this reason decreases, the financial market for physical assets is opening up.

Start with what a token is, in economic terms, because the industry has spent a decade describing it wrong. A token is a claim. It asserts that some unit exists, and that it belongs to the holder. A blockchain is very good at enforcing the last part. It solves the double-spend problem for the token, so the claim cannot be copied or forged in transit and its ownership is legible to everyone at once. What it cannot do is verify that the underlying thing the claim points to is real, or is what the claim says.

A renewable energy certificate, a REC, says one megawatt-hour of clean electricity went onto the grid. This market is worth around $22-28 billion and grows 14% every year. Why are RECs easy to put on a blockchain?

Because all electricity on the power grid mixes together, it is impossible to separate clean energy from dirty energy once it is flowing through the wires. To solve this, physical meters are placed directly at solar and wind farms to measure the exact amount of clean power they produce before it enters the main grid. When the meter records a specific amount of clean energy being generated, it creates an official document called a REC. Companies use these certificates to prove that a specific amount of clean energy was actually created and added to the system.

This is how the grid works:

The economic significance is that the party certifying the unit is not a human with an incentive to shade the truth. It is a meter, and behind it, the grid, an involuntary third-party verifier with no stake in anyone’s sustainability report. Falsifying its reading would mean corrupting physical infrastructure the seller does not own. In the language of information economics, there is almost no asymmetric information about the core attribute. Buyer and seller face the same hard number, produced by a device neither controls.

When you tokenise a REC, then you are wrapping a unit whose quality has already been established by an independent measurement. The blockchain inherits a fact. And what it contributes is transaction-cost reduction, which is precisely what a liquidity technology is supposed to do.

Power Ledger, an Australian firm that spent years on peer-to-peer solar trading between neighbours, now runs a REC marketplace called TraceX, where generators and corporate buyers trade certificates on-chain instead of through weeks of bilateral paperwork and legal brains. It cleared over 1.2 million RECs in a single month in early 2025, and it plugs into the registries that actually issue the certificates. It connected to M-RETS, one of the largest voluntary REC registries in North America, and in mid-2025 to ERCOT, the Texas grid operator that issued more than 32 million RECs in 2023 alone. Power Ledger says users cut administrative costs by up to 72%.

Now the same idea didn’t work when it comes to carbon credits, which look like RECs. Each credit acts as a numbered, tradeable unit for one ton of avoided carbon. Many corporations promise their shareholders and the public that they will become “carbon neutral” or reach “net-zero” emissions. Because they cannot actually stop polluting entirely, they buy credits to mathematically cancel out the pollution they continue to create.

In 2021, a protocol called Toucan built a bridge to haul them on-chain as tokens, and a project called KlimaDAO offered a clever incentive to pour credits into its treasury, buying them up to prop the price and, in theory, make polluting more expensive. Money rushed in, and KLIMA, the token, touched $1 billion marketcap before people looked at what had been bridged.

The core problem with a carbon credit is that it measures a hypothetical scenario, like trees that were supposedly never chopped down. There is no independent hardware to verify this, making the units entirely subjective. Tokenisation could not fix this underlying flaw, but it did make it a bit more dangerous. By pooling these credits together on-chain, protocols like Toucan treated every asset as identical.

It wasn’t that the blockchain itself was bad, but the specific way they pooled these credits together caused a problem.

Then KlimaDAO manufactured demand for that pooled token, buying it up with its own freshly minted KLIMA, at a price propped far above what the junk inside the pool was actually worth. That combination sorted the market. If you held a good credit that fetched a fair price in the normal market, turning it into a pooled token priced at the junk level was a bad trade, so you stayed away. If you held a credit nobody wanted, the pooled token was priced above your credit, so you bridged it and sold. The pool filled with the lowest-quality credits, because those were the only ones the trade made sense for.

A researcher at CarbonPlan in 2022 found that most of the credits bridged to Toucan came from projects already excluded from serious offset markets on quality grounds. Because Verra’s retirements are public, CarbonPlan was able to read exactly which credits went on-chain. It found that 99.9 per cent of them came from projects too old to qualify for the standard aviation offset market, and that 28 per cent came from ‘zombie projects,’ ones that hadn’t sold a credit in years until crypto demand revived them. One hydro project in China logged its first ever retirement only when it bridged on-chain, fifteen years after it started.

A 2024 meta-analysis in Nature Communications looked at nearly a billion tonnes of carbon credits, about a fifth of every credit ever issued, and found that fewer than one in six represented a real emissions cut.

Verra, the largest carbon registry, watched its retired credits get turned into tradeable digital ghosts and, in May 2022, banned the practice outright. KLIMA fell from three thousand six hundred dollars to single digits. KlimaDAO spent over a million dollars of its own treasury retiring the worst of what it held.

Tokenising a broken unit doesn’t repair it, but industrialises it. You take a measurement problem, and you bolt a liquidity engine to it, and now the bad units move faster, price higher, and reach more buyers.

Which gives you a test you can carry to any tokenisation pitch, and it has nothing to do with the token. Is there an independent witness measuring each unit? Can that witness be faked? And, the one people forget, can the borrower own the witness? That’s the tough one.

If you tokenise a barrel of oil in a tank, or a tonne of grain in a warehouse, the whole thing rests on someone attesting the barrel is in the tank and stays there. If the person doing the attesting is the same person borrowing against it, you don’t have collateral.

Commodity trade finance has blown up this way for a century, warehouse receipts written against metal that had already left, or was never there.

Look at livestock, one of the biggest stores of value on earth for people with no access to credit. Banks have always hated lending against it, discounting a cow’s value by as much as 60%, because they had no way to know the animal was healthy, or where it was, or whether it was still breathing. A cow is collateral that can walk away, get sick, or quietly die in a field while the loan sits open. Who is watching the barn?

Three days ago, on a dairy farm in Paraná, Brazil, ten cows became the first livestock formally registered as collateral on the country’s stock exchange.

Each animal wore a smart collar from an agtech company called Cowmed. The collar tracked its health, behaviour and location, hashing that stream into an encrypted identity tied to the loan. The farmer borrowed around twenty thousand dollars against ten cows. The system can even tell if a cow dies, and let the farmer swap in a live one. Cowmed already monitors a hundred thousand animals valued at nearly four hundred million dollars.

If a lot of farmers adopt this, a smart collar tells you a collar is transmitting health and location data. It does not, by itself, tell you the collar is on the cow it’s registered to, that the animal is the one pledged, or that a real cow is on the other end at all. Every failure a farm inspector used to catch is still there. You can strap the collar to the healthiest cow and pledge a sick one. You can move the collar between animals. In principle, you could feed it fake data. So no, it doesn’t end the farm visit on its own. But it can turn one yearly visit into a daily record, so fraud has to be kept up full-time. The same cow now can’t be pledged to three lenders at once, since there’s a record.

A grid meter is placed at a fixed spot the seller can’t move. A cow is mobile, and the collar is only as honest as whoever strapped it on.

A cow with no collar is nearly unverifiable, so banks cut its value by sixty per cent. A cow with a collar isn’t perfectly verifiable either. But, verifiable cheaply enough that a lender takes the bet at a smaller haircut. So, the asset moved up the spectrum. This deal is also three days old, and we don’t have a lot of examples to argue for.

Successful tokenisation relies on a spectrum of measurement reliability. High-quality assets use automated, tamper-proof sensors like power grid meters, satellites, and weighbridges that sellers cannot alter. Low-quality assets depend on hypothetical projections and self-reported math, such as avoidance carbon offsets.

Even the metered end isn’t as settled as it sounds. A meter proves a megawatt-hour existed, but not that it existed near you, or when you needed it. For years, a company in Ohio could buy a certificate from a wind farm in Texas that generated at 3 am and claim to run clean.

So the market is now splitting the unit finer, stamping each certificate with the hour it was made and the grid it was made on. Water credits are following the exact same pattern. A watershed credit is easy to sell and hard to trust, because measuring that a river got cleaner is one thing and proving your project is why it got cleaner is another. The instrument decides how honestly an asset can be tracked. Each time the instrument improves, the market revises what the certificate was really claiming to be.

In a normal market, buyers and sellers will constantly argue over what an asset is worth, causing the price to fluctuate. That is perfectly healthy, but it only works if there is a hard, physical truth anchoring the asset. Like the actual revenue of a company or the physical weight of a barrel of oil. Eventually, the trading price will align with that reality. If you remove that physical proof (like with a subjective carbon credit), the price is completely disconnected from reality. There’s no baseline for the number to fall back on.

Tokenisation is just an accelerator. If the underlying asset is verified, tokenisation builds wealth. If the asset is unverified, it just accelerates a scam. The blockchain cannot check which is which. Crypto projects often defend themselves by pointing to their high liquidity and institutional capital as proof of their legitimacy. Liquidity does not create a true fact. You can build the most efficient, frictionless trading system on earth, but if the physical asset is of no value, the entire market is just a highly optimised fraud.

—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.

It was called remote disbursement. To stall a payment, a New Jersey buyer would issue a cheque from an obscure bank thousands of miles away in Montana.

By adding physical distance and extra handoffs between banks, the cheque took days to clear. With interest rates soaring past 10%, keeping that cash in your account for just a few extra days yielded free profit.

There were consultants who did nothing except maintain maps of which small-town banks took the longest to clear. It was the volume that made it work. Americans wrote 8 billion checks in 1970 and 16 billion in 1980. The Federal Reserve float, the money that exists for a moment in two places at once because a cheque has been credited and not yet collected, averaged about $3 billion a day in 1972. It more than doubled between 1975 and 1978.

The Fed was furious. It published a report on remote disbursement in February 1979, then a policy statement telling banks to stop helping. The Monetary Control Act of 1980 called for the practice to be eliminated. Check 21 finished it two decades later. Most cheques now clear in one business day and the Fed processes every one of them in a single building in Atlanta. That was the end of it.

Siemens employs more than 12,000 people across 80-odd countries in a division called Global Business Services, processing invoices and running payroll and reconciling statements. Airbus started one in Lisbon in July 2021 and now has more than 1000 people in Portugal across that centre and its industrial unit. Goldman Sachs, the entire firm, employs 47,400.

Then in September 2024, the same Siemens issued a €300 million bond that settled in minutes with blockchain.

Today I want to show you what connects those two facts. On to the story…

Start with the bond, since it’s the easiest

Siemens issued €60 million on Polygon in February 2023, waiting two days for the trade to settle. 18 months later, it returned with €300 million, swapping a public crypto network for SWIAT. Built by a consortium of European banks, SWIAT is a closed ledger designed specifically to play by institutional rules. That regulatory alignment allowed it to connect to the Bundesbank’s Trigger Solution, settling the entire €300 million automatically, in minutes, in central bank money.

Investors subscribed directly and could see their register entry the moment settlement completed. Peter Rathgeb, the corporate treasurer, ran both issues, and the second one removed nearly all the settlement risk from every party in the deal.

Normally, an institution like Clearstream (a massive Central Securities Depository) has to sit in the middle just to handle the certificate. Nobody’s job at Clearstream is safe or unsafe because of one German bond.

Secondly, the payroll department

Deel runs payroll for more than 40,000 businesses and 1.5 million workers across 150-plus countries, processing over $22 billion a year. Since January 2026, a company can fund its entire global payroll straight from a stablecoin treasury. In June 2026, Deel launched its own dollar-backed digital balance called DLUSD.

What leaves the employer’s side? Trapped liquidity in foreign bank accounts. Emergency troubleshooting for failed wires. Intermediary FX conversion fees. Manual account reconciliations.

What leaves the worker’s side? In Argentina, Turkey and Ukraine, a salary paid in local currency can lose 20 to 40% of its value inside a year. In 2025, 85% of Deel’s Argentine contractors chose to be paid in dollars. In June, the company gave them a dollar balance to hold inside the app they already use, built on Bridge and Privy and Tempo, with rewards on idle funds. In May, it started paying full-time employees in stablecoins on Polygon in the US and Eurozone, at 10 to 25% of net salary after tax.

From a treasury perspective, stablecoin utility is now backed by hard data. According to an EY-Parthenon survey of 350 executives, 13% of firms have already adopted stablecoins. Of those, 41% have realised cost reductions greater than 10% in cross-border B2B payments. This efficiency reclaims $5 million on a $50 million transfer program. This is capital previously lost to fixed operational costs. Looking ahead, 54% of surveyed non-users intend to adopt this infrastructure within a year.

Real end-user stablecoin payments hit roughly $390 billion a year by February 2026, double 2024, and about 60% of it is now B2B rather than trading. Hyundai Card settled a cross-border corporate payment in seven minutes. Visa is running a $7 billion annualised settlement rate across nine chains, up 50% in a quarter. None of these is crypto companies experimenting at the edges.

Businesses are using stablecoins to solve the mechanical problems of slow cross-border settlement and trapped capital discussed earlier.

Third one, the biggest, least discussed - Collateral

JPMorgan’s Kinexys runs about $5 billion a day and has cleared $3 trillion in total, with its intraday repo alone passing $1.75 trillion. BlackRock has posted tokenised money market funds to Barclays as derivatives collateral. Broadridge moves $354 billion a day. CME is building the same thing with Google Cloud, aimed at a $15 trillion collateral market.

I want to explain why collateral is the one to watch if you’re building.

A hedge fund holds a position it doesn’t want to sell and wants to borrow against it. Neither side trusts the other, so the collateral goes to a triparty agent who sits in the middle and holds it. Somebody negotiates the agent’s contract first. Then the asset travels for days, and fees stack.

Semi Liquid secures the collateral in place and changes what the borrower can do with it. Freeze in place, adjust the spending conditions, skip the transfer. Just like Kinexys, it’s going at the same $15 trillion. Look now at the jobs that survive the introduction of software; the logistical aspect of moving the collateral disappears, and what remains are only the difficult human judgments involved in assessing the value of the asset and making the late-night default calls. Which is the line I’d hand to anyone deciding where to point a budget.

Which means technology is not eliminating the cost of misjudging a counterparty. This explains why some of these projects lived and the rest died.

Between 2018 and 2020, big companies including HSBC, Maersk, and BNP Paribas launched four trade-finance blockchain networks. None survived past 2023.

• June 2022 - we.trade went bankrupt.

• November 2022 - Maersk and IBM shut down TradeLens.

• Early 2023 - Marco Polo shut down, $4.6 million in debt.

• November 2023 - Contour closed after averaging only 60 to 70 trades a month (later acquired by XDC Network).

Faster digital letters of credit improved document speed, leaving the core cost of credit underwriting untouched.

Komgo is the only survivor of the group. It got there by dropping the blockchain. Launched in 2018 by major banks, it originally offered both a digital letter of credit and a document workflow tool. The credit product died; the paperwork product continued. While competitors collapsed trying to digitise trust, Komgo survived by pivoting strictly to mechanical plumbing.

Tether proves what happens when a financial company refuses to judge counterparties. By operating strictly as a mechanical conduit, taking dollars, holding Treasuries, and issuing tokens while exchanges handle the actual customers, it removes the operational cost of human trust. This perfectly automated asset-moving model allowed a 300-person team to clear $10.09 billion in profit in 2025. Generating $33.6 million per employee, Tether massively outperforms traditional risk-managing banks like JPMorgan and Goldman Sachs.

Then it stepped into credit. Its secured loans went from $14.60 billion on September 30 to $17.04 billion on December 31, almost $7 billion added in six months. Protected by a $6.34 billion safety net, though the borrowers are kept secret.

The second Tether started lending money, it had to hire humans to evaluate collateral and make late-night margin calls.

Two details heavily distort Tether’s profitability. First, nearly half of its profit came simply from the surging prices of its massive gold and Bitcoin reserves. Second, it pays zero interest on the billions of dollars of tokens it has issued. If a traditional bank held that much customer cash, it would owe billions in interest payouts every single year.

Circle shows the same boundary from the other side. It booked $1.4 billion in distribution costs tied to Coinbase in 2025, up from $924.5 million, per its 10-K. That’s 51% of its $2.7 billion revenue going to a company that neither issues USDC nor manages the reserves. Coinbase runs the onboarding, the KYC queue, the fraud team. Verifying a customer’s identity still requires human judgment. Circle keeps its headcount low by outsourcing this process entirely, turning what should be an internal payroll expense into an external contract line.

So, where to build.

Enterprise blockchain and stablecoin technology solve three operational problems. It automates the movement of cash, collateral, and certificates. It frees up idle capital trapped in payroll floats and escrows. And it removes cross-border friction in Latin America, Africa, and Southeast Asia, the exact use case driving the 10% cost reductions reported by executives.

Corporate issuance, which almost nobody is working on. Germany’s eWpG made Siemens possible. Total eWpG digital securities issuance was around €236 million by June 2024, with KfW adding €150 million across two deals. That’s nothing against the European corporate bond market.

Treasurers used to profit from delayed settlements. When the float died, they shifted to managing pre-funded accounts. Now, tokenisation has vaporised that too.

—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.

Each of the twenty districts is run out of a dzong, a whitewashed fort with sloping walls and a monastery in the courtyard, usually planted on high ground above a river. Many are three centuries old. They were built to be hard to reach, because they were built to survive sieges.

Which worked beautifully when the job was defence.

Bhutan has around 786,000 people scattered across a very steep country. Roughly eight in ten work the land. Roads follow ridges instead of cutting through them, so two villages facing each other across a valley can be half a day apart. To get a document, you went down the mountain and up to the fort, and you waited, and you carried photocopies of everything you owned. Forget one paper, and you came back another day. The cost to the citizen is the day of farm work they lost.

In October 2023, Bhutan switched on the National Digital Identity (NDI), built to delete the whole tiring process the people went through. Two weeks ago, Bhutan started exporting it. We need to understand if the success belongs to the tech, or to Bhutan’s society.

The National Digital Identity puts your credentials in an app, locked with your face or your fingerprint. The records are on citizens’ phones, without a central vault holding everyone’s records. The state can confirm a credential is real without keeping a copy of it. A highly practical use case of zero-knowledge proof.

Say a service needs to know you’re over eighteen. The old way, you hand over a card carrying your birth date, your address, your ID number and your face, all of it, to settle one small fact. The NDI lets you prove the fact alone. Yes, over eighteen. You prove that is true without showing the file it came from.

Bhutan moved rails three times. It started on Hyperledger Indy, a closed ledger, the cautious institutional pick. In August 2024, it moved to Polygon for speed.

In October 2025, with Vitalik Buterin and the Ethereum Foundation’s president standing in the room, Bhutan anchored the system to Ethereum itself. First country on earth to put a live national identity system on a public chain, with full migration due this year.

Zero personal details are stored on-chain; it acts purely as a cryptographic ledger verifying that a credential remains valid and unaltered. The chain is a notary that no ministry can rewrite, including Bhutan’s own. The GovTech secretary’s reasoning was that Ethereum is spread across so many machines worldwide that it’s very hard to knock over. Bhutan selected a widely distributed public network to ensure its identity system stays online even if foreign infrastructure crashes. Because if a small nation like Bhutan relied on infrastructure controlled by a superpower (like US-based AWS or Google servers) and those servers suffered an outage, cyberattack, or political sanction, Bhutan would be left completely helpless. Since they cannot fix another country’s or company’s servers, they would just have to wait in the dark.

In February 2024, four months after launch, about 24,000 people had signed up. A year later, that was 219,506 registered users, out of 280,757 attempts, so not everyone who tried got through.

By the end of March 2025, the count of foundational IDs issued reached 234,568, covering citizens and foreign workers. A ninefold jump in twelve months, and it’s also roughly three in ten people. There’s been no official update published since, though the system has kept adding pieces.

On March 13, 2026, the Bhutanese Ministry of Foreign Affairs and External Trade, in collaboration with the GovTech Agency, issued an official press release announcing the integration of the Bhutan NDI wallet and eKYC with the national passport system. This means citizens no longer need to fill out physical forms or scan documents when applying for a passport. Instead, they can log into the G2C (Government to Citizen) website, scan a QR code using their NDI app, and their verified details, photograph, and e-signature are instantly synced to the passport office.

How did the 3 out of 10 that signed up get there? They got there because things people already had to do were routed through the app. Civil servants were required to register. Enrolment for Gyalsung, the national service programme every young Bhutanese goes through, runs on it. Three of the country’s banks, BOB, BNB and BDBL, wired it into remote account opening, so you can open an account without the trip to the branch. Telecom SIMs carry verifiable credentials. Passports followed.

High trust is what let Bhutan try. Ordinary usefulness, plus a few mandates, is what moved people. They might have downloaded it because the bank won’t open the account otherwise, and because the alternative is climbing up the fort.

But then, plenty of governments issue mandates. Very few get obeyed without a fight.

That’s where I get to romanticise this whole thing. Bhutan measures itself by Gross National Happiness, an idea its fourth king put forward in the 1970s that the constitution now instructs the state to pursue. Easy to file that under the tourism brochure that we all see and get excited. In practice it works as a test that proposals have to pass, and the test is whether the thing improves how people live, not only what it earns. A digital ID that saves a farmer two days of walking clears that bar without question.

The country also has a strange history with power. The monarchy handed authority to an elected parliament in 2008, and the king spent years persuading a reluctant public to take it. Because the monarchy willingly surrendered its power to the people, the Bhutanese public does not have an ingrained fear of an oppressive government. They do not automatically suspect that state tech initiatives are secret surveillance plots designed to track them.

When that state introduces a digital ID and promises it is actually better for privacy than a physical ID card (since you only share the necessary data, not the whole card), the public takes them at their word. The NDI Act passed both houses unanimously.

The same sovereign fund that built the identity system, Druk Holding and Investments, has been mining Bitcoin with surplus hydropower since around 2019. Holdings peaked near 13,000 BTC in late 2024 and have since been sold down to somewhere around 3,100, with hundreds of millions moving out to wallets nobody has identified. In May, the fund’s chief executive said he did not recall selling any. On-chain trackers say otherwise. A citizen-controlled identity system and an opaque treasury at the same time. After all, it’s a small state making bets, some of them as murky as anyone else’s.

Compare it with the United States, which for the last twenty years has been trying to standardise a driver’s licence.

REAL ID went into a wartime spending bill in 2005. It was a law requiring higher-security driver’s licenses to board domestic flights or enter federal buildings. Twenty-five states passed statutes or resolutions rejecting it. Fifteen wrote laws forbidding their own officials to comply. The opposition coalition ran from the ACLU (American Civil Liberties Union) to Gun Owners of America, groups that agree on nothing else, and they held for two decades. Enforcement finally came into force in 2025. Basically, the country took twenty years to settle the format of a card everybody already carries.

Because there’s a belief most Americans hold without being taught it. A government that can identify you reliably will eventually use that against you.

Barry Goldwater said it in 1974, warning that a universal identifier would leave a trail behind a person for life, ready to be reassembled and used to press on them. That instinct comes from a country with a long memory of files kept on the wrong people.

In fact, in America, proving who you are is an industry. Credit bureaus, verification vendors, background checkers, brokers reselling the fragments. When we talked about the paperwork in Bhutan, it was the effort and the time they lost on it that made it a problem. In this case, it’s the product itself.

So the same thing looks like a bridge over a mountain in Thimphu and as the first page of a dossier in Washington.

On July 5 this year, Sierra Leone’s technology ministry signed an agreement with Bhutan NDI and the SIGN Foundation to build the same kind of system there.

Bhutan is sharing its battle-tested, open-source digital wallet architecture, prioritising the training of Sierra Leonean developers so the country can manage its own W3C-standard system instead of paying perpetual vendor fees.

Sierra Leone has almost none of Bhutan’s advantages. It’s far bigger, far poorer, with a civil war inside living memory, and a long public record of corruption. If the theory is that only small, stable, high-trust countries can pull this off, Sierra Leone is the case that tests it. If it works there, the theory was wrong, and the useful ingredient was not Bhutanese serenity. The true formula came down to open tech standards, targeted state mandates, and a government with everything to gain by lowering bureaucratic hurdles — if it works.

The dominant narrative around this technology is still the one that promises escape; dodge the state, route around the rules. Bhutan used it to run the other way and make the state easier to reach, and it’s now sharing that with whoever wants it, while its own rollout is still in progress.

The result is mundane, people don’t NEED to climb the mountains too much.

—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.

Paradigm, which is one of the largest dedicated crypto funds ever assembled, recently raised $1.2 billion to start backing startups in AI, robotics, and aerospace. They have also entirely scrubbed “crypto” from their website! Their thesis is that crypto was their first frontier, but there’s so much else happening right now that they can’t ignore it.

Framework Ventures also closed a $400 million fund in June to start investing beyond crypto, and they’re far from alone. Nearly every major crypto-specialist fund has started drifting towards broader themes and mandates over the past year. In Q1 2026, we only had eight new crypto-only venture funds launched, which is the lowest since 2020.

Today I will dig into whether crypto-specialised VC as a fund category is actually dying. If so, how does the shakeout map into the lifecycle of these funds, and what does it mean for crypto startups, which will now be competing for attention inside multi-sector portfolios?

The Specialist Fund Lifecycle

The reason crypto-specialised funds became a thing was that they were willing to spend time building the competitive advantage and were the only people willing to underwrite the risk at all. Understanding how Solidity contracts actually work and forging relationships with pseudonymous developers building in Discord channels was not something accessible to a growth equity partner at Tiger Global in 2017.

And to understand whether crypto VC is dying as a category, it helps to look at how specialised fund categories have lived before, because this has happened more than once in the past.

Between 2006 and 2011, climate tech went mainstream as an investment thesis, and VCs started raising dedicated clean energy funds for the same reason crypto VCs raised dedicated blockchain funds: thinking they spotted a generational technology shift before the generalists caught on, and wanted to build entire franchises around that conviction.

They poured in over $25 billion into clean energy startups and lost more than half of it. What’s Interesting is that the technology actually worked, and today’s clean energy market is enormous. It led to solar costs falling 85% over that period. But what VCs misinterpreted was that they applied the same software-company model and wrote $5 million seed cheques into companies that, in reality, needed $200 million of project financing and fifteen years to reach profitability.

MIT’s Energy Initiative ran a post-mortem and found that the venture model was fundamentally flawed for the sector. The specialist VCs who funded the experimental phase by taking the technical risk funded the early R&D and gave the sector the credibility to attract bigger capital, but once the technology matured enough for infrastructure lenders and project finance facilities to underwrite it, the specialists’ informational edge had evaporated.

SPACs played out a similar pattern. For context, a SPAC is a blank-cheque company that raises money through an IPO without having an actual business. It then later uses that cash to merge with a private company to take it public faster than a traditional IPO would allow. In 2020 and 2021, some investors saw this as a repeatable vehicle and built entire firms around them.

Chamath Palihapitiya had raised $1.6 billion in dedicated SPAC capital. But by 2022, two-thirds of 2021’s SPACs hadn’t completed a merger, and Chamath had to return the money he had raised. All of this happened in less than twenty-four months, which can give you a sense of how fast things change once the specialist edge disappears.

And there’s a big reason this keeps repeating itself across completely different industries. Carlota Perez documented it across 250 years of technological revolutions, calling it the techno-economic paradigm. She stated that every major technology goes through an early phase where only insiders understand it, and the people closest to the technology become its most valuable investors because they’re the only ones who can tell what’s real. Then the technology matures and starts integrating into existing institutions.

At that point, the insider edge that built these specialists doesn’t matter anymore because the asset class becomes legible to generalists with bigger balance sheets. Fred Wilson saw this coming for crypto specifically. He wrote about it in 2015, predicting that crypto would hit a major “financial break point” as it crosses from what Perez calls installation into the deployment phase.

That break point is hitting now, and you see what the deployment phase for crypto looks like. We had fintech like Stripe acquiring Bridge, and launching their own stablecoin chain. Institutions like BlackRock and Fidelity have launched their own tokenised money market funds. Even traditional payment giants like Visa and Mastercard are now building settlement layers on top of stablecoin rails.

These companies don’t need a crypto-specialist VC to explain MEV extraction and Validator economics because the specialist crypto knowledge doesn’t matter to them. What they need instead is regulatory approvals, distribution, and banking partnerships. Something that any other fintech would need to scale. Today, a generalist at Sequoia or Founders Fund can evaluate a crypto deal the same way they’d evaluate Stripe or Plaid.

The Barbell and the Drift

So if the specialist edge has dissolved, what actually happens to the funds that were built on it? Their fate depends entirely on the economics of their fund size.

See, venture capital has been splitting into a barbell for years. At one end, you have mega-platforms like a16z, Sequoia, and Founders Fund that can absorb entire asset classes as verticals inside their portfolio. On the other end are tiny cottage-industry funds that write small conviction-driven cheques into things they understand very deeply and can return the whole fund off a single breakout. Everything in between has become a kill zone, and that’s where most crypto-specialist funds sit right now.

A $500 million fund needs to generate roughly $1.5 billion in total exits to return 3x net to its LPs, so you can’t get there from seed cheques alone because no seed portfolio produces enough breakouts at that scale. And you also can’t compete for growth-stage deals against $5 billion megafunds that are willing to write $100 million cheques without thinking twice about it. For example, in 2025, Founders Fund alone raised 1.7 times what every emerging manager combined raised in the first half of the year. The capital is concentrating at the extremes.

This barbell is also why Framework Ventures and Paradigm look like they’re doing the same thing, but in reality they are very different. Framework sits at $400 million, which is too large to return the fund off a few seed bets and too small to compete with megafunds for growth deals. At that size, crypto alone doesn’t generate enough exits, so they had to broaden. But Paradigm, at $1.2 billion, is large enough to try becoming a multi-sector platform altogether, which is strategically different. In simple terms, the size of your fund determines which end of the barbell you land on, and that determines what options are available to you.

Even the crypto VCs that say they’re staying on the course have completely redefined what ‘crypto’ means. Dragonfly had raised $650 million in February, which was oversubscribed by 30%. Still, they have been explicit that non-financial crypto has completely failed, and the firm is betting exclusively on stablecoins and prediction markets. A16z’s latest $2.2 billion crypto fund was raised in May 2026, and it’s half the $4.5 billion they raised in 2022; not just that, Chris Dixon has even shifted his framing from crypto being a new computing paradigm to finance being the foundation for everything in the space.

What these firms now call ‘crypto-only’ investing is a bet on financial infrastructure built on blockchain rails, which is exactly what the generalist funds with bigger cheques will fund too.

A major force driving all of this is also the LPs’ behaviour. The Venture industry is currently in a DPI crisis where 2021-vintage funds have returned roughly 0.08x. And LPs who got burned during the 2022 crypto crash now have an obvious alternative in AI, which is absorbing 70% of global funding this year. So when your LPs have been sitting on dead capital for four years and see AI companies generate the returns they once expected from crypto, the fund manager has no choice but to create exposure to AI forcefully.

For crypto founders who are still building, this is concerning because it leaves a shrinking pool of investors who actually understand what they’re working on and are committed to backing them. The obvious response to the shrinking is to say, founders should just raise from generalists instead. And on paper that might sound accurate - Sequoia and Founders Fund can write bigger cheques, and they also bring in distribution that no crypto-native fund can match.

But there’s a problem with that. As AI companies are sucking up the majority of the dealflow today, a crypto founder inside a generalist portfolio will be competing for attention against that, and the deal would need to be exceptional just to get on the agenda, which is a very different game than pitching to a crypto-specialist who eats, sleeps, and breathes this stuff.

Another problem is the growth of the crypto ecosystem as a whole. Specialist VCs didn’t just write cheques but also funded the infrastructure layer that made the next generation of applications possible. Paradigm funded research into MEV, Dragonfly backed cross-chain tooling, and all this obviously doesn’t have the commercial returns on a deal-by-deal basis, but it builds the commons that the whole ecosystem runs on. A generalist fund will never fund that because they evaluate deals on standalone returns.

I think in a few more years, calling yourself a Crypto Investor will feel the same as calling yourself an internet investor. Crypto has now become plumbing, a set of rails that enable financial products. And you don’t build a fund thesis around the plumbing; you build on top of it. And if Perez’s framework holds, this is exactly what’s happening - Crypto is no longer the thing you invest in; it’s the infrastructure underneath the things you invest in.

But that doesn’t mean specialist crypto funds disappear entirely. As tokenisation and onchain securities like new categories keep emerging, there will be niches that generalists can’t touch, and small funds will keep forming around them every cycle. What’s dying is the current generation of large dedicated crypto funds that are too big to survive on niche crypto deals alone. The category will keep on reshuffling around the barbell, where generalists will take the big cheques, and small specialists take the frontier bets.

The early specialist funds of 2017-18 financed Uniswap and Ethereum, the tooling that made stablecoins work. But this era is changing; there are also cases where few of the most successful recent crypto projects, like Hyperliquid and MegaETH, have raised entirely through community rounds and zero-VC models. The specialist funds made crypto legible enough for generalists to move in, but there are also founders realising they can do it without them.

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.

It feels like a modern sickness, this premium we pay just to hold something from somewhere else. Until now, I looked at it as a globalised supermarket problem. Then I read the book, A Splendid Exchange by William Bernstein. Published in 2008. This book is about humans, so is every book.

This one follows it all the way down to a single piece of glass, carried across a Stone Age sea. That’s where we want to start.

Twelve thousand years ago, long before the invention of the wheel, the first writing, or the concept of a sovereign state, human beings were already risking their lives to trade. Deep within the dark recesses of Greece’s Franchthi Cave, archaeologists unearthed shards of obsidian. It is a razor-sharp, volcanic glass that could only have come from the island of Melos, over a hundred miles away across the open waters of the Aegean Sea. Why did these early humans risk drowning in primitive boats just to fetch shiny black rocks?

We know where it came from because obsidian keeps a secret. Bernstein notes that atomic fingerprinting can trace these obsidian flakes as they travel thousands of miles from their volcanic origins.

Every volcano lays down its glass with a slightly different mix of atoms, like a chemical signature. You cannot fake it or lose it. The archaeologist Colin Renfrew followed those signatures across the Fertile Crescent and found the pattern that gives the whole game away. Two hundred and fifty miles from an Armenian source, half the cutting stone in a village was obsidian. Five hundred miles out, it was two per cent. The farther the glass travelled from the mouth of its volcano, the rarer and dearer it became.

I find this almost mind-blowingly beautiful. A sliver of glass, sharper than a modern scalpel, moving hand to hand across a roadless world, each village keeping a little and passing the rest along. A peer-to-peer network running eight thousand years before we decided we invented the idea.

Trade is older than farming and older than the written word. It might be the oldest thing we do that isn’t eating or sex.

Why did we do it? Because we are a little bit insane.

Pepper grew on one coast of India, and nowhere the Romans could reach. To get it, ships had to catch the monsoon at the right week, run open ocean with no sight of land. They trusted the wind to carry them home, hoped for it, but it turned and drowned plenty of them. The ships went, again and again, because a sack of peppercorns bought at the source came back to Rome multiplied many times over. This might sound silly for all the effort, but a rich Roman just wanted his meat to taste of somewhere far away.

Pliny grumbled that pepper had no real merit beyond its bite. Romans crossed the earth and paid for it by weight, like gold. When Alaric and his Goths finally had Rome by the throat in 408, part of the ransom they demanded was three thousand pounds of pepper. The city that ruled the world bought its life partly in spice.

When Alaric and his Goths finally had Rome by the throat in 408, part of the ransom they demanded was three thousand pounds of pepper. The city that ruled the world bought its life partly in spice.

I have to stop here because this one is personal. That coast the Romans were dying to reach, the one place their pepper grew, is my coast. I’m from Calicut, on the Malabar shore, the pepper capital of the medieval world. The black spice Alaric took as ransom for Rome grew in the hills behind my hometown, cheap as weeds, close to free where it stood.

Everything that made it precious got added on the way out, by distance and danger and the men who owned the ships.

Bernstein tells this story from the buyer’s chair. He talks about how ancient Rome, old Europe, and big markets always wanted rare things from far away. Funny to read, the view from here is uglier. Not to be all emotional about it, but markup was never really the reward for the risk. It was the prize for controlling the gap between the source and those who really wanted it. And sooner or later, the people who controlled the gap will come and take the source itself. In 1498, Vasco da Gama’s ships reached the coast, and he sailed back a few years later with cannons. But the coast just got a new owner. For a thousand years, the world paid huge premiums for goods from far away. When the buyers finally reached the actual source of those goods, they brought warships to take control of it by force.

Setting the Indian textbook aside and going back to Bernstein, he has a chapter he calls “The Captives of Trade.” The same ships that carried the pepper carried people. Between about 1200 and 1500, Italian merchants became the richest slave traders on earth, buying human beings on the shores of the Black Sea and selling them into Egypt. People far away would pay money for them, and the middlemen who controlled the narrow routes kept the extra profits.

Trade routes also acted like an early version of the internet. They carried things that moved much faster than regular cargo, like new ideas, religions, and deadly diseases. For example, religions like Islam and Buddhism spread across the world along the exact same paths used by merchants. The roads moved silk and spice, and they moved gods. Then, in 1347, the network delivered its most efficient shipment ever. The Black Death rode the same Silk Road caravans and Genoese ships that carried the luxuries, and it killed something like a third of everyone between China and Ireland. Bernstein calls the plague a stowaway, and the word is perfect if you ask me. Mind-virus and the actual virus riding along beside trades, unpriced.

For a while, we thought crypto was that. We really believed, a few years ago, that these networks were the modern Silk Road (not that one) and that the ideas riding them, open money, code instead of banks, ownership without a landlord. We said it would spread like a faith and rewire how the world thinks about value. Tough to believe now, and reading Bernstein is part of why.

I see crypto as the pepper, not the plague, in this set-up. Exciting, a niche luxury product traded by a small group. Most of the planet has never been touched and is not being thought about.

Then what is the thing that spreads like Islam down the incense road right now? It’s the dollar, slipping into every weak-currency country in the world through a stablecoin, so that a shopkeeper in a collapsing economy saves in someone else’s money. And the thing spreading like the Black Death is whatever is loose in the models now, the synthetic flood, the machine-made everything. Crypto flatters itself that it is the contagion. On the evidence, it looks like a spice. Is that a win-win?

What crypto did keep, perfectly, is the old shape. It promised to kill the markup, to cut out the merchant and the king and the temple and let value move hand to hand like obsidian. All we could end up doing was creating new markups, such as for early investors and token issuers. Regular retail investors take all the financial risk, and when markets crash, they lose their money while the new middlemen keep the profits.

William Bernstein trained as a neurologist and then became one of the most trusted investing writers alive. He once said that all the things that make us human make us terrible investors, and you have to learn what they are and how to avoid them. The fear, the greed, the herd pull, the certainty/uncertainity. A Splendid Exchange is the long proof of that message.

The people who win in investing are the ones who stay completely calm when everyone else panics or gets greedy. Money always moves in one direction, which is away from normal, emotional people and into the pockets of the few who control their feelings. Holding power is about controlling your own humanness. The small group of people at the top stay rich simply because they keep a clear head while the rest of the world loses theirs.

This is where part one leaves me, halfway through a book about how trade built the world. The stone changes. Whether it is obsidian, pepper, silk, or crypto tokens, the specific item being traded doesn’t matter anymore and that’s evolution. We realised that tokens and all the shiny things were a distraction from the system, its flaws, and its corruption.

Being human is expensive, and someone standing at the narrow point has always known it. The second half of the book is where that machine goes industrial, where companies get their own armies and free trade gets a philosophy, and the whole thing scales up into the world we live in now. That’s part two. I will look more closely at the insane psychological wiring that drives us. Until then, hold this thought.

—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.

Kalshi just launched markets on GPU compute prices, meaning you can now bet on where GPU rental costs will be three months from now. And for the first time, compute has a forward curve. In Part I & Part II, I covered how compute is becoming tradable and how it is being financed underneath. Today, I want to complete the picture by walking you through every layer of the market forming around it.

Each layer of compute being built is either held by the oldest financial institutions or crypto-native startups that didn’t even exist a few years ago. We’ll map each layer to see where crypto infrastructure is genuinely load-bearing versus where traditional players have an advantage. Let’s dig in!

The Pricing Stack

Let’s start with price. All commodity markets in the world begin here because everything that comes after futures contracts and lending covenants gets settled against this one reference number.

In oil, for instance, the number is published by firms called price reporting agencies. The largest of these agencies today is Platts, which is now part of S&P Global. What Platts basically does every day is survey traders and refineries, collect the prices at which crude is sold, and publish a single reference price. This price is then used in all futures contracts and cargo deals. Warren Platt started this back in 1909 as a trade magazine with $2,500. And over a century later, Platts remains the dominant reference price for physical oil.

Platts survived for a century because once a benchmark is set for the first generation of contracts, switching away from it later becomes almost impossible. You would be surprised, but even Platt’s most vehement detractors still use it because they have legacy contracts and other derivative instruments based on Platts, and adopting any other price source would create a basis risk across their entire book. The benchmark is more permanent than even the companies trading around it.

In Compute Markets, two firms are racing to become the Platts of compute. The first is Silicon Data, which I introduced in Part I. It collects rental-price data from over 30 GPU cloud providers, covering 3.5 million data points, and publishes a daily index, called SDH100RT, on Bloomberg terminals. CME has announced that its compute futures will settle against this index.

The second firm is Ornn, which, instead of surveying providers on what they charged, only counts prices from transactions that actually completed on its own exchange venue, where over 400 data-centre operators and AI companies trade. This is very different from how Silicon Data does it because benchmarks built on surveys can be gamed. In oil, traders have manipulated Platts assessments for decades by submitting inflated quotes during the assessment window to push prices in directions that benefit their positions.

Ornn’s approach of using only completed transactions makes it far harder to pull off, and crypto infrastructure gives an edge: if those transactions settle on-chain, the raw trade data becomes verifiable to anyone who wants to check the ledger. You never have to trust that the benchmark provider filtered the data correctly.

This model of verification does not exist anywhere in traditional commodity markets and could help make compute benchmarks more trustworthy. Ornn has over 400 data-centre operators on its platform and has also announced an integration with ICE to settle their compute futures on Ornn’s index.

Now, the exchange side of this layer is barely developed. The first-ever compute swap was cleared in December 2025, and that too on Ornn’s own venue. The biggest reason is that 78% of compute supply is still concentrated among the four hyperscalers, and there is no standardised, tradable unit. A rating system for GPU clouds was tested that grades providers on networking quality and uptime. Because buyers are willing to pay significantly more per GPU-hour for higher-rated clusters, the annual revenue difference between the lowest and highest tier on the exact same hardware can reach up to $15 million. This tells you how far off compute still is from being a fungible commodity where one unit equals any other.

DePIN platforms like Akash and io.net sit on the other side of it. They provide spot markets where compute can be traded at hourly rates, for a fraction of what hyperscalers charge. But the problem with these is that they are not yet very reliable. A developer who had tested four DePIN providers had two of them shut down on him in just eight months of setting up. But trades here are visible and on-chain, so if DePIN continues to grow, these platforms could become another raw source of verifiable price data, making compute benchmarking more credible.

The Intermediary Stack

Once the pricing is in place, someone then has to sit between the GPU owner who wants to sell capacity and the AI lab that wants to buy it. In commodity markets, these middlemen are called dealers. Their job is to take a custom request on one side and match it with the available supply, by managing the price risk in between. In compute markets, the dealer layer has split into two tiers with completely different economics.

At the top, we have the hyperscalers who charge a hefty premium for immediate GPU capacity. For example, Google’s deal with SpaceX runs at $920 million a month for 110,000 GPUs; that pricing is roughly 4x what a typical neocloud will charge for the same 5-year contract. The premium is for getting the capacity now rather than in eighteen months.

Google needed capacity immediately to meet its surging demand for Gemini Enterprise while its own data centres were still being built, and that urgency is what creates the 4x premium. Only a handful of companies can serve this tier because offering large-scale compute on short notice requires owning the hardware outright, without carrying any debt against it.

Next below the hyperscalers, we have the commodity tier, where all the neoclouds and brokers compete on price. Neoclouds have to cover every cost from GPU-hourly revenue alone. This means everything from data-centre construction to daily operations, and for many of them, the unit economics just don’t work without raising additional capital.

Brokers, on the other hand, run live order books without owning any hardware. The biggest example of this currently is SF Compute, which manages over $100 million in GPUs and processes orders as low as $0.25 per GPU-hour from clusters that would otherwise sit idle.

And all the financial infrastructure being built today for Compute is to serve this same commodity tier. Because if you are Meta and your customer is Google, you don’t need a futures market or a clearing layer; Google’s credit is sufficient to make that deal happen. But if you are a neocloud borrowing against hardware that depreciates each year, you need to lock in revenue months in advance and prove that the hardware is performing as promised.

Once a trade is matched, there are two more things to take care of. First, someone has to guarantee that the trade will settle regardless of any party defaulting, which is what a clearinghouse does. Second, someone has to prove that the physical assets exist and are not encumbered. This is done through custody and verification. But in Compute, the clearinghouse does not exist yet because the futures themselves are not live.

For verification, we discussed in Part II how a unique cryptographic key can be used to prove that the chip is genuine. But since compute is a flow good, verification cannot just be a one-time inspection, as it is for metals. GPUs switch workloads every hour, so verification must also be performed continuously. This is why on-chain infrastructure becomes critical here: it allows the asset to self-report its status directly to the ledger, where it can be verified instantaneously.

At the bottom of all this is the Lending layer built around the same collateral bifurcation — whatever the lender actually underwrites as collateral is completely different depending on which tier the borrower sits in. With a hyperscaler, the lender is not really underwriting the GPU. It is underwriting the revenue contract behind it, which is essentially project financing. But with a Neocloud at the commodity tier, the lender doesn’t have that luxury. It has to underwrite the hardware itself and explain why you need everything else above in the stack being built.

Every commodity market that has gone through this cycle has eventually concentrated most of its durable value not in producing the commodity but in the companies that control the financial instruments around it. The oil majors built the wells and the refineries, but the exchanges and the clearing firms ended up with the better economics because they earn on volume and do not have to worry about the price moves.

Compute is still early in that transformation, building all of its financial infrastructure at once, some of it on traditional rails and some of it on-chain. The on-chain part comprises three distinct layers. First is pricing: on-chain transaction settlement makes benchmark data verifiable without trusting that the provider filtered the numbers correctly. Next is on-chain verification, which gives a shared ledger combined with the GPU’s own cryptographic identity, allowing it for continuous, real-time proof that the hardware is genuine and unencumbered.

And lastly, the settlement layer, where stablecoins are the only payment rail that works for AI agents buying compute per inference call, since it doesn’t require them to have a legal identity. At every other layer of this stack, the lenders and traditional infrastructure have an advantage.

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.

That’s how the platforms extracted, taking the decision away. Couch lock – one of my favourite words from Tim Wu’s Age of Extraction.

Can’t pick stock, sure, index funds, S&P…

Not sure about lending. Call it a savings account, be proud.

They all just charge you for removal. Removal of decision-making, also sometimes the removal of the best benefits. People didn’t care.

What DeFi did was add. Which chain, which pool, what rate, when to move, where’s the bridge, is this even the real app or made with Claude Fable before July 12? Aave has 2.5 million users and six years of history. Revolut has 65 million users. Is it fair if we say Aave has to be the smarter one here?

Between January and July, the rate on Aave’s USDC pool wandered between roughly 2% and 9%.

Bouncing around is normal in DeFi. You watch the rate go up and down, and you move your money whenever it makes sense.

But that structure is unsellable everywhere else. Neobanks can’t explain to a user that the rate is set by borrowing demand and could collapse to 2% when it clearly advertises a savings account. People don’t put money in chaos. That is why people never start using crypto apps, let alone crypto savings apps, every day.

On July 9, Aave Labs found a workaround called Stable Vault. Today, I want to show you how it works, who is making money off it, and why I think regular people are going to use it anyway.

Stable Vaults lets any business plug in once and start offering a savings account. A neobank, a wallet, a payroll processor. Deposits route into Aave’s lending markets. The user sees the rate in the app they use every day; if it seems like a good deal, they’re in.

The rate is fixed. I know it’s a tough thing to say in crypto.

Aave’s market pays whatever borrowers are paying that week. Stable Vaults sits on top and hands the operator a dial. The app types in a number. 4 per cent, say. From then on, the vault pays four per cent per second, no matter what Aave is doing underneath. That is the app’s problem, not yours. Anything the strategy earns above that number goes to the operator.

POV: the person putting money in

They get rated insurance. When Aave’s USDC pool paid 2% this spring, a vault promising 4% still paid 4%, and the operator covered the difference.

Handing off risk has a price everywhere else, and it has one here. Look at a fixed-rate mortgage. It costs 50 to 100 basis points more than a floating one, and that premium is the borrower paying for a fixed number.

The person doesn’t have to create a wallet, keep the seed phrase, bridge, or choose a chain. They get a support line, account recovery, Face ID, and a company with an address if something breaks. Aave’s own app runs SOC 2 certification and 2FA and markets both, because those are what the customer is buying.

The top of the range is what the user loses. When the pool paid 9%, they got 4%. When it paid 6%, they got 4%. It’s a fixed rate fixed by the app, based on what tier the user is. A moving rate is a live measurement you can actively check; an unmoving rate completely hides the middleman’s cut.

The user also picked up a second counterparty. By using this setup, they add two new failure modes to their balance sheet. The financial health of the fintech company running the app, and the code quality of a hidden, private script that shuffles their capital around behind the scenes. In plain DeFi, your only risk is the core protocol code. Here, if the middleman company goes bankrupt, or if their private backend script crashes and misplaces the funds, their money is just as gone, even if Aave itself works perfectly.

In a real swap market, the fixed rate is driven down to something close to fair value because both sides can shop. Here, the operator sets the number alone, and the customer has nothing to compare it against. User isn’t comparing 4% to Aave’s 6%. They are comparing it to their bank. Aave’s app page puts its own rate next to the FDIC national savings average of 0.40%, and against that, anything looks generous.

POV: The Operator

Take a neobank with $200 million of idle user stablecoins. It already has the money and users. It paid to acquire them. It wires in one integration, advertises 4%, and if the strategy returns 6%, it books $4 million a year on a balance sheet item that was previously a cost. Low effort 2%, not bad.

Rise is a payroll company. It pays contractors across 190 countries and has processed over $1.5 billion. When a company pre-funds payroll a week early, that USDC used to sit dead, so Rise built Rise Earn to park it in Aave’s USDC pool on Arbitrum until payday.

Rise takes 1% of the interest earned and nothing else. On a 6% yield, that’s six basis points. The worker gets 5.94%, and they just see Aave’s live rate.

A Stable Vault operator on the same money takes 200. The middleman’s cut went up thirty-three times.

POV: Aave and Stable Vault

Aave sells the vault’s ability to set per-user rates for loyalty, campaigns, or tiers, so think that a premium subscriber gets 5% and everyone else gets 3.5% from the same pool of borrower interest. A fintech issuing its own stablecoin can register it as a deposit asset and run a closed loop. And a balance that earns doesn’t leave, which means the yield is also a retention line.

Is the operator getting this for free? No. It’s short the spread in both directions. When Aave paid 2% this spring, every vault promising more than that had to write a cheque.

Which brings us to April 18, when the Kelp DAO bridge exploit on April 18 triggered a massive run on Aave, pool utilisation hit 100%, completely freezing all withdrawals and trapping both the operator’s paper profits and the user’s money in the same locked queue.

When utilisation hits the ceiling, nothing comes out for anyone, including the vault. The surplus stacks up on paper next to the user’s principal.

If liquidity returns, the operator sweeps a surplus that piled up during the days the user couldn’t leave. That surplus is the fee the market was paying for illiquidity. The person supplying the illiquidity was the user. If liquidity doesn’t return and bad debt hits the pool, the vault is short, and Aave’s docs say an authorised party can top up the system. No reserve behind “can”.

Aave will say its contracts were never exploited, that Kelp’s bridge failed and not Aave’s code, and that rsETH was frozen within hours. All true. They had previously voted to accept the risky collateral at a dangerous 93% loan-to-value ratio right before their risk manager resigned, leaving everyday users to bear the brunt of a broken app.

Now Stable Vaults look like the last piece?

Rise runs payroll floats through Aave. Kraken white-labelled Aave v3 into a protocol called Tydro on its own L2, then pointed its retail Earn product at it, so a Kraken customer tapping earn is an Aave user too. Cap Finance parks stablecoin reserves in it.

Horizon lends against tokenised treasuries with Circle and Franklin Templeton in the room. Aave App goes straight at consumers. Stable Vaults lets everyone else in and calls it diversification.

Aave doesn’t need more deposits. Kulechov told The Block in March that there’s a liquidity surplus in DeFi and that the focus has to shift to the borrowing side. He’s right, and it’s why USDC has been paying 2-3% instead of 8%. What Aave has always faced is that DeFi capital is mercenary and leaves for fifty basis points. Routing them through an app that owns the relationship, like the payroll, turns the flightiest money in finance into something that acts like a deposit base.

Aavenomics 3.0 went live on June 27 and now automatically buys AAVE from revenue. Revenue needs to flow through the pipe whether or not anyone is bullish. In a bear market, sticky deposits are what keep the buyback running. How do you get them? Stable Vaults.

Coinbase pays around 4% on USDC balances. Robinhood opened Earn on July 1 at roughly 7%, with 28 million funded accounts. Both call it savings.

Coinbase runs on Morpho and Ethena. Robinhood runs on Morpho and Maple, with the risk parameters set by a firm called Steakhouse.

Both of them had to build that. Custody deals, curators, risk teams, months of lawyers. Aave’s contribution is to make all of it unnecessary. With one integration, any app on earth can put a number on a screen and keep the distance between that number and whatever borrowers happen to be paying.

A bank gets to use it because a century of law stands behind it. Reserve requirements, examinations, deposit insurance, and a regulator who can walk through the door unannounced. All of that was built because everyone agreed long ago that the bank lends your money out and somebody has to be standing there when the lending goes wrong.

Everything Stable Vaults delivers is available to anyone with 20-30 minutes. But you make a wallet, bridge some USDC. Then supply it to Aave. What you don’t have to face in this case is KYC, an operator, or a rebalancer. And you lose nothing to spread. You get 6% instead of 4%, and you can see the pool the whole time.

I understand that the system thinks much longer term than my logic does. Also, when people don’t do that, I don’t think they are stupid.

Iyengar and Huberman’s work on retirement plans found that as the number of fund options went up, the number of people who enrolled went down. Faced with more choice, people picked nothing. Every consumer finance product since has been built on that result.

Self-custody has been right for fifteen years, and everyone has been told. Most on-chain card spending still routes through custodial platforms anyway. It’s a preference, expressed repeatedly, at scale. And their safety maths is better than ours. For someone with $2,000 and no crypto background, the likeliest way to lose it is by losing a seed phrase or sending it to the wrong address. An app with Face ID and account recovery removes the failure mode that would have got them. They’re paying 200 basis points for insurance against themselves, and that’s a rational purchase.

So Aave did the right move. It’s what a business with liquidity and no loyalty is supposed to do, and every consumer app in crypto is racing to the same place because the same logic is sitting everywhere we go.

In the end, this is an acceptance of human nature. People want safety, predictability, and most importantly, ease. Building a life is hard enough without having to run a personal central bank. They just want to close the app, look at a number that sits perfectly still.

—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.

This is BlackRock in crypto right now, even though all you can think about when I say this are ETFs.

BlackRock launched one; it swallowed something like a hundred billion dollars. Every outlet on earth ran the same piece about how the suits finally came for crypto. Fine, true. Also, the decoy.

An ETF is a product. You buy it, you sell it, it is right there in your brokerage account next to your index funds. If BlackRock’s Bitcoin fund vanished tomorrow, crypto would shrug and keep moving.

But today, I want to talk about the BUIDL playbook.

BUIDL is BlackRock’s tokenised money fund. It’s a pile of short-term US Treasuries and cash, chopped into on-chain tokens, that pays yield straight to whoever holds them. It launched in 2024, holding about $2.5 billion, which sounds small and is small. But I was looking at what it was turning into.

BUIDL holds Treasuries, so what’s inside the fund is the safest thing you can own. The danger is how you get in and out of it. BUIDL is permissioned. Only wallets that BlackRock and its partner Securitize have approved can hold it or move it, there’s a whitelist, which BlackRock controls. You can only cash it out during set hours, on BlackRock’s terms. BlackRock, or a regulator leaning on BlackRock, can freeze a wallet or halt redemptions. An ordinary crypto token moves the instant you tell it to. In BUIDL’s case, someone else can switch off for you. That is the reason it’s a risk, and it’s why everything built on top of it inherits that risk.

BUIDL is becoming the collateral the rest of on-chain finance gets built on. Today we are talking about that very sovereign takeover.

Start with Ethena and USDe. It is a synthetic dollar that’s now the 3rd-largest stablecoin. Ethena runs a second dollar, called USDtb, and more than 90% of its backing is BUIDL. When USDe’s engine sputters in a rough market, USDtb is the shock absorber it leans on. So BlackRock’s Treasuries already sit a layer below one of crypto’s largest dollar assets.

And a couple of weeks ago, it went further. BUIDL became the main reserve for the stablecoins that Ethena now builds for other companies, a core ingredient in a whole product line.

Let’s be precise about how much of USDe leans on BlackRock, though. On a calm day, barely any of it. USDe is backed mostly by crypto held against short futures, with only about 7% parked in plain stablecoins, including USDtb.

Ethena has $6 billion circulating supply, but only an $65 - $80 million native reserve to absorb market losses. When the funding trade stops paying, Ethena can’t rely on its own tiny buffer. It has to aggressively dump its capital into the Treasury fund.

So BlackRock becomes load-bearing right when things are breaking, the one moment you’d least want to learn your floor belongs to someone else.

In a standard crypto panic, if you want to exit a position, you interact with an automated smart contract. It doesn’t care who you are. But BUIDL is a permissioned, whitelisted token. If BlackRock’s compliance desk flags an address, or if regulatory pressure forces it to pause redemptions, that liquidity instantly freezes.

If the collateral under a stablecoin, or under an exchange’s margin system, can’t be sold at the exact moment everyone needs to sell, then every loan and every leveraged bet resting on it is trapped too, and they all unwind together. Traders think moving into a Treasury-backed asset is the safe play. They’ve traded the risk of crypto prices swinging for the risk that one company, on the worst possible day, won’t let them out.

Since April, big traders on the OKX exchange can post BUIDL as margin, the money they put up to back their positions. Meanwhile, the tokens sit off to the side in custody at Standard Chartered, a proper global bank. The collateral earns Treasury yield while it works. It’s the first time a bank that size has done this for crypto. Idle margin became a paying asset, and the asset is BlackRock’s.

A couple of weeks ago, BlackRock wired USDe into Aladdin.

Aladdin is BlackRock’s risk management system. It’s software that monitors portfolios and models what could go wrong. By adding USDe to Aladdin, BlackRock is now closely tracking it. The system is named after Aladdin, the story where a lamp grants wishes, but in this case, the benefits mostly go to whoever controls the lamp.

BlackRock isn’t doing crypto a favour with the USDe integration if you think so.

What they are doing is using their software to pull the inner data, risk profiles, and structural behaviour of the on-chain financial system into their own brain.

They know which players are overleveraged and roughly where prices have to fall before those positions are forced sold. So when a sell-off starts, BlackRock already knows which dominoes go first. Even if you think you are an independent crypto founder running a sovereign protocol, you are ultimately forced to price your risk using BlackRock’s proprietary math.

USDC, the second-largest stablecoin with around $78 billion in market cap, keeps the bulk of its cash reserves in a fund that BlackRock itself manages.

When you add it all up, BlackRock is a big buyer in crypto and also, crypto is being built on top of BlackRock’s collateral and margin, with its risk system overseeing both.

What does this look like from inside? A trader holds USDe, which leans on USDtb, which is mostly BUIDL, and then borrows against that USDe to place another bet. That’s what collateral does. It disappears into the foundation, and everything heavier gets stacked on top of it, meaning the loans and the leverage other people pile on those dollars, all of it assuming the bottom layer will hold and will let them out. If it doesn’t, they all have the same problem at the same moment.

Ethena may be loud, but other exchanges have started taking tokenised Treasuries as margin too, some from BlackRock, others from rivals like Franklin Templeton. BUIDL is the biggest one in the room. The biggest usually keeps winning. Traders want to hold the same collateral everyone else holds, because that’s the one that’s easiest to borrow against and sell in a hurry. So money keeps flowing to the biggest option, which makes it bigger still.

BlackRock has done this before and won it.

Index funds were cheap and slow, the oatmeal of investing. BlackRock rode its iShares brand as that oatmeal ate the world, and today the three big index shops, BlackRock alongside Vanguard and State Street, are the largest shareholders in close to 90% of the companies in the S&P 500. And for that, they just stayed under everything as the money piled in.

Aladdin did the same on the operational side. That risk platform now runs the books on north of $20 trillion in assets, somewhere above a tenth of all the financial assets on the planet. Much of that money belongs to firms that compete with BlackRock and still pay to use its system. So even the competition uses BlackRock’s math to measure its own safety. That means BlackRock sees every card on the table.

You want the measure of how deep that goes; look at 2008. When the crash hit and the US government found itself holding a mountain of toxic mortgage assets it couldn’t price, it hired BlackRock, asking it to put a number on it.

It happened again in 2020. When COVID froze the bond market, the Fed hired BlackRock to run the rescue, buying corporate bonds and bond funds by the billion. A large share of what BlackRock bought, on the government’s behalf, was BlackRock’s own funds. The money wasn’t BlackRock’s. It was public, the Treasury put up the cash, and taxpayers were on the hook for any losses. The government let BlackRock decide where the money went, and a chunk of it went into BlackRock’s own products. Its flagship credit fund swelled from $28 billion to $47 billion in two months on the back of it. BlackRock ran the buying and supplied the thing being bought, and when people cried foul, the reply was that BlackRock was just the agent and the conflicts were handled with care. BlackRock issues the collateral and runs the system that prices it, all handled with care. Look carefully, and that is the seat BlackRock is building in crypto now.

If tokenised dollars keep settling on BUIDL, and DeFi keeps parking its reserves there, and risk keeps getting priced through Aladdin, BlackRock becomes the operating system of crypto.

So why does this matter more than the ETFs? Because an ETF and collateral are different things. An ETF is just demand. People buy it for exposure, and if they change their minds, they sell, and no one else is touched. Collateral is the opposite. Once protocols use BUIDL to back their dollars and their loans, they can’t pull it out easily, because pulling it out means unwinding everything sitting on top of it, every dollar it backs and every leveraged bet resting on those dollars. So nobody pulls it. It stays put by default. People keep it because removing it has become too dangerous to try. You can sell an ETF in a second. You can’t rebuild the foundation of a system while everyone is still standing on it.

But this is the early days. That $2.5 billion in BUIDL is a small next to the stablecoin market, which runs past $300 billion, most of it Tether and Circle. On today’s numbers, BlackRock isn’t the floor of anything.

And you could argue this is good news. Crypto’s whole history of blowups runs on junk collateral, made-up tokens propping up other made-up tokens until the morning they all fall down together. Putting US Treasuries, watched by the most trusted name in the business, under your synthetic dollars is how you stop that. More trust.

But first of all, size is the wrong lens. The right one is position. Ask how much is built on top of BUIDL, not how big BUIDL is on its own. A $2.5 billion fund that a dozen other products lean on matters far more than a $50 billion fund that nothing else touches. BUIDL is the largest single fund in the tokenized-Treasury market, a roughly $15 billion category that has more than tripled in a year, and Boston Consulting Group puts tokenised real assets on a path to $16 trillion by 2030. It’s small today. But it’s the base everything else is starting to settle on.

Secondly, the control. It’s the same deal every industry has already signed with BlackRock. You take the stability, hand over the control. It feels like a steal because the stability is the part you can see, and the control is the part that doesn’t bother you yet. And BUIDL is a permissioned, whitelisted, redeemable fund on BlackRock’s terms.

To grow and avoid crashes, crypto protocols needed ‘safe’ assets to back their tokens and support their trades. That’s why they started, or may start, using BlackRock’s BUIDL fund.

If the whole industry relies on BlackRock’s models to measure risk and its assets to back their dollars, changing the system becomes impossible; any change could destroy the market.

The industry gets the stability it wanted, but at what price?

That’s all for today. See you soon with another one.

—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.

“Man is small, and, therefore, small is beautiful. To go for gigantism is to go for self-destruction.”

E.F. Schumacher wrote that in 1973, in a book called Small Is Beautiful. He was an economist who had stopped trusting economists.

The rich world decided that the big problem, which is making enough stuff, was solved. And that all that was left to do was make more of it, faster, forever.

The new game plan was to make more of it, faster, until the end of time.

Schumacher thought this was a fatal misread. Because nothing in nature grows without limit, and the one system pretending it could was ours.

It was a strange read. A man asking the world to stay small and human and humble. And here we are, at the far end of the experiment he watched begin. A whole world order built on one unspoken promise, that the line can go up forever and nobody ever has to pay for the climb. Every country, every company, all running the same one-word instruction. More, more, and MORE.

“Gigantism”- his word for the sickness. Schumacher thought size itself could be a disease. A city, a company, a machine, a government, past a certain scale, stops serving the people inside it and starts serving itself. “Any activity which fails to recognise a self-limiting principle,” he wrote, “is of the devil.”

All I could think was, isn’t crypto one of the loudest rooms in that building? It is the kind of technology that might resonate with this the most. Small, peer-to-peer, built so no giant sits in the middle. You can guess where this goes, so I’ll keep it short, because we have had this conversation so many times and have moved on. Made to escape the big players, ended up growing its own, the way anything successful does.

The more useful thing Schumacher hands us, along with a scorecard on how decentralised we really are, is a way of seeing what scale costs. And that shows up most clearly in what the network eats.

The book is not a perfect prophecy. Schumacher got plenty wrong, and some of his solutions have aged like milk. But I am choosing this one because his diagnosis of what happens when systems grow too massive hits close to home.

Every business owner knows the line between income and capital. Income is what you can spend. Capital is the thing that makes the income, and if you start spending that, you’re dying slowly, still calling it a good year. Schumacher’s charge against modern economics is that we make this error with the one kind of capital, the kind man never made and only found. Fossil fuels, yes, but also the soil and the water and the living systems underneath everything. We book them as income and spend them like they refill. Once they’re gone, they’re gone for good.

Now Bitcoin enters the frame, doing its best impression of what it was talking about. Proof of work turns electricity into digital scarcity, on purpose. Cambridge reckons the network burns around 138 terawatt-hours a year, roughly half a per cent of the world’s electricity, on par with a mid-size industrial country. Whatever you make of the exact figure, the shape of it is pure Schumacher.

We take a capital asset, energy, much of it still fossil, and convert it into a thing whose entire value is that it was expensive to make.

Now, AI runs the same trade with another kind of capital. Which is not oil, but commons. The whole written and drawn record of human beings, scraped into models as if it were free and endless. Researchers at Epoch AI put the usable stock of high-quality human text at a few hundred trillion tokens and expect the models to chew through it somewhere between 2026 and 2032.

That’s why Elon Musk told us that “the cumulative sum of human knowledge has been exhausted in AI training.”

This was capital all along. The labs are now paying for it. News Corp handed its archive to OpenAI for more than $250 million. Reddit is selling its comment history; yeah, that’s why AI loves Reddit. You want AEO? I heard that you go sit on Reddit. Data that was once booked as free income but instantly gained a price tag once it became scarce.

This capital might not survive being spent. When models train on the output of other models, the quality rots, a failure that researchers named “model collapse” in a 2024 paper. If we flood the internet with AI-generated garbage, the AI starts learning from its own fake data and rots from the inside out. We’re basically poisoning our own water supply while we’re still trying to drink from it.

And the power bill always matters. Electricity doesn’t come from nowhere. It gets made, mostly by burning something, and there’s a fixed amount to go around at any moment. The IEA expects the world’s data centres to climb from about 415 terawatt-hours a year in 2024 to roughly 945 by 2030, close to the entire electricity use of Japan. By the end of the decade, data centres in the US alone are set to pull more power than all its heavy industry combined, the steel, cement, aluminium and chemicals that build the physical world. Much of it still comes from the same fossil capital Schumacher begged us to protect.

Why can’t we stop?

The whole machine runs on greed and envy, and runs on them by design.

That’s why there’s no stop.

Growth economics needs you to want more, always, or it stalls. Schumacher thought a system built on farming those two feelings would slowly make its own people stupid, unable to see plain things in front of them. A collapse of intelligence.

A market powered by FOMO and green candles, maybe, feeling a little called out? But in crypto, for long stretches, envy has been the product.

Think about how most people got pulled in. Someone they knew turned a small bet into a noticeable number, and they didn’t want to be the fool left standing outside. That feeling, the fear of watching someone no smarter than you get rich, is what the thing ran on. Although that’s not the case anymore, since numbers don’t impress much now. But back in the peak, entire coins were launched with no purpose except to manufacture them, to make the next person afraid of missing the next one. Schumacher says that a machine built on that feeling takes your money and, if you give it long enough, your judgment.

I have to be ME about the book here because it would be easy to make Schumacher a saint, and he wasn’t one (RIP). His cure is where he loses me. He wanted “intermediate technology,” small tools, local making, “production by the masses” in place of mass production. I think it’s a beautiful idea. I also think it’s mostly lost, and it’s only fair to say why.

Schumacher pictured a village weaving its own cloth on simple looms, staying whole and self-reliant. What lifted people was the opposite of that. China, then Bangladesh and Vietnam, put millions of people into enormous factories making clothes and phones and steel for export to strangers far away, the thing he called uneconomic.

It was often grim and underpaid. It also pulled roughly a billion people out of extreme poverty between the 1980s and now, the fastest fall in human history. China’s own poverty rate fell from 88 per cent in 1981 to under 1 per cent by 2015. The world’s extreme-poverty rate dropped from about 38 per cent in 1990 to under 8 per cent today. The huge, ugly, centralised machine did something the gentle local one couldn’t. Which is to feed and power billions, fast.

Coming to the fight he’d have cared about most, the climate one. The reason clean power is finally cheap is a handful of colossal Chinese plants that, in 2024, produced about 86 per cent of the world’s finished solar panels and more than 90 per cent of the cells and wafers beneath them. They drove the price so low that a panel in Europe now costs less than the fence you’d bolt it to, and pushed solar to roughly four cents a kilowatt-hour, by most measures the cheapest electricity humans have ever made. The same machine built 16 million electric cars in 2025, three of every four on the planet. Smallness couldn’t have done that.

Food is the same story. The Green Revolution that saved India from famine ran on industrial fertiliser, big irrigation, and lab-bred seeds, not on small self-limiting farms. Schumacher would have hated the runoff and the debt it left behind.

The scale he warned against is also the scale that fed the world. And some of his vision has aged badly, like the passage where he calls mothers working in factories “a sign of serious economic failure.”

For a lot of the women he was pitying, that factory wage was the first money that was ever theirs to keep, the first exit from a father’s house or a husband’s whim. The man who wanted everything human-scale couldn’t see that the big, anonymous factory was the very thing setting some of those humans free.

All of these brought me back to things we read before.

We spent three weeks on Tim Wu’s The Age of Extraction, and its whole argument is that a platform, once it becomes the unavoidable host for everyone else’s business, starts skimming. The app store takes its 30 per cent. Amazon takes half of what you sell. If you think about it, both men are saying the same thing over fifty years, that once a thing grows past a certain size, it no longer serves the people inside it; instead, it feeds on them.

In the oldest fight there is, who owns the ground everyone else has to stand on. For Schumacher, it was soil. For Wu, it’s data and rails. Wu still believes a working government will show up and prune the weeds.

So I went back to the chicken farm. Xiaowei Wang put a fitness tracker on a chicken in a Guizhou village so a farmer could prove the bird was real, that it walked around and had a life before it became dinner. It’s the smallest, most human use of a blockchain I’ve ever come across. And how did that small world go?

GoGo Chicken, the company behind the anklets, once talked about tagging 23 million birds. It never got near. In Wang’s own telling, the Guizhou farmer’s orders dried up right after the first sale, the fate of most beautiful small things. What scaled were the giants, Alibaba and JD, running food tracking for the whole country. So eventually, the big machine keeps the parts it likes.

So where does that leave me, closing a very old book in the middle of a hideous market?

Wanting to believe him, mostly. I’ve always thought people are better than the systems they build, that we’d choose enough over endless if someone just showed us the door. Schumacher shows you the door? Not sure about that, but he writes with a warmth you don’t always find in a book with “economics” on the cover. Reading this felt like being handed permission to want a smaller, kinder world.

And then the grown-up in me says it out loud. It’s too good to be true. Greed and envy don’t lay down their arms so easily. The world he wanted needs everyone to want less. For the big machine, it may already be too late to turn.

But maybe the big machine was never the place to look. It’s hard for you to change the big machine now. Change keeps coming from the edge, from the small and the overlooked, from a phone in Kenya doing what a bank branch wouldn’t, from a stablecoin holding its value in a country whose own money won’t. Crypto has been flunking Schumacher’s test. But it’s also one of the few rooms left where people still ask what he asked, with a straight face. What is money for? What do we owe the people who come after us?

Small is beautiful was a dare. Most of the world assumes that the era has passed and that giant machines have won. But in a few stubborn corners, we don’t have to listen to them. We can choose to stay small anyway. And let the mothers keep their jobs.

—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.

AI is believed to be the next big bubble. A bubble is when people throw wild leverage at it, build their entire business on a house of cards, ignoring the flaw in their system, and then it fails; they get to blame the bubble.

I looked at Bittensor, which pays people to make AI (good thinking, love the thought behind it). The system is split up into a hundred or so little ecosystems called subnets. You build something, it gets scored, and you get paid instantly in a crypto token called TAO.

Full disclosure before we start. I have a defect. I can’t look at anything shiny without going straight for the crack in it.

Someone shows me a miracle; my first thought is, who’s footing the bill? I’m working on it. Sorry in advance. But most times, I would love to be proved wrong.

So dock me a few points for temperament, then keep reading. Because Wall Street is racing to wrap in an ETF, Bitwise and Grayscale have filed for a Bittensor ETF with the SEC. The crack this time is just right there for you to see.

Bittensor tries to build AI the way Bitcoin builds security. Pay people in tokens to compete, let the competition sort the good from the junk. The network is cut into about 128 subnets, and each one is a small market for a single job such as inference, training models or scraping data.

Miners do the work, validators grade it. TAO pays Miners based on the quality determined by the validators. Validators are paid based on how closely their grading matches the other validators’ grading, weighted by stake. So a validator earns by scoring in line with the pack, not by being right in some absolute sense.

How much TAO each subnet gets is set by the subnet’s alpha token price, not by quality at all. Also, the subnet owner skims a fixed 18% off the top before anyone else is paid.

TAO is a roughly $2 billion token, with about $690 million of it staked into the subnets, the pools that decide which AI work gets funded.

To make things more complicated or awesome (depending on the way you look at it), every subnet has its own token. It’s called alpha. When you invest your TAO in a subnet, you are buying its specific token and pushing up its price. That price determines how much of the new TAO a subnet gets compared to the others. And it’s the price averaged over time, not the price right this second.

So one quick pump won’t get you a bigger cut. You have to keep the price up, and that means you have to keep buying.

As always, cracked markets run on loops. Buying alpha drives up the price, which increases that subnet’s share of new TAO. That TAO goes straight to the people already holding the alpha, giving them cash to buy more. Money coming in makes the number go up, and the number going up pulls more money in. The only brake is that the network constantly prints fresh alpha, forcing miners and validators to sell it to cash out, which drags the price down. A subnet only stays funded if new buyers keep arriving to absorb that supply. Which is on purpose, and how the system is designed to work.

Now the good thing about this is that, because each subnet carries its own token, you can bet on a single slice of AI by itself. Buy the inference subnet and skip the training one, or vice versa. Capital gets to pick a lane inside the machine, which no stock market lets you do.

The chain can only see tokens moving, rather than AI being used.

There’s no clean tally of who really paid for what. The price is free to drift wherever the money leans. A stock can’t do that. Markets usually price real value, sure. Nvidia’s price has real sales sitting under it, sales you can look up. A subnet’s price has nothing under it but the buying. So when money coming in is the only thing you can measure, money coming in is the whole price.

Validators are supposed to score miners honestly. The system that runs it, Yuma Consensus, does fight cheating. It clips scores that stray too far from the pack, so a validator who over-rates a friend earns nothing for the excess. Good design. But the anti-collusion math holds only while the cheaters control less than half the validator stake in a subnet. Past that line, a validator and a miner can agree to swap high scores for a cut of the TAO, and the chain pays them for it. There’s another leak too, weight-copying, where a validator skips the grading entirely and copies everyone else’s scores off the public ledger, collecting rewards for doing nothing. The patch for that, commit-reveal, seals the scores in an envelope for a while so there’s nothing to copy. It only works where miner quality keeps shifting fast enough that stale scores go wrong. Where the work is steady, the copying pays again.

Now take that cheating threshold and look at who holds the power. One team, Rayon Labs, runs three of the top subnets that together take roughly a quarter of all the TAO the chain prints each day. Around two-thirds of TAO sits staked, plenty of it in a few hands. You can stand in either of two places here.

Stand in one spot, and Bittensor is a market doing its job. Instead of a closed-door committee deciding which AI deserves money, thousands of people bet openly on which kind of intelligence will matter, and funding flows toward their bets. Money leaning toward a thing is often the first sign the thing is promising.

Stand in the other spot, and a price means something only when it’s tied to the world outside it. A customer. A sale you can point to.

Doesn’t the tie look loose? Bittensor’s top earner is paid far more in printed tokens than it takes from real customers, and the people who can turn the rewards on and off are still a small core, close enough that when the founder switched emissions and sold token piles mid-fight this spring, the network’s largest operator, Covenant AI, packed up and left.

Early mistakes are swiftly made and then swiftly corrected; this token already bears the benefit of a hard fork to right a wrong. On Optimism, crypto-native venture capital got so frustrated with the endless rounds of fundraising designed to anticipate future developments that they instituted a form of retroactive funding. Payment is made for what has proven valuable, rather than a speculative bet on what might be. Rewards are given after verifying that the work has delivered on its promise, rather than as an a priori subsidy for a token’s promise. Both Gitcoin and Filecoin have experimented with different genres of this idea.

Bittensor’s entire wobble is a result of rewarding in the currency in which value is realised rather than the much more reliable double-spending of the underlying proof.

The network has changed how it decides a subnet’s pay twice a year. First, it went by the token’s price. Then, from November, by net staking flow, the money moving in minus the money moving out. In June, it switched back to price after the flow method ran into its own problems. Both were just stand-ins for the thing no one can measure, which is whether anyone actually used the AI.

A network willing to destabilise its very existence by uprooting its own fundamental law twice in a short span is arguably more dynamic in its capacity for change than most. That said, the sobering reality of this hard fork is that the three sets of criteria all failed to take into account a single vital metric - the willingness of any one individual, outside of the subnets, to pay for the work being done by others. By rewarding one group for bringing in money, they are all trying to let money follow money, rather than value follow demand.

Even financial waste builds infrastructure. Just like the dot-com bubble created the fibre optic backbone of the modern internet world, the Bittensor bubble is creating computer silicon and AI training runs that can possibly outlive the hype.

And the prize for this particular investment has enormous value because it involves distributed AI. The way to beat chip monopolies is to create open-source alternatives, as Linux did for operating systems and Wikipedia did for encyclopedias. This same kind of disruptive innovation is happening right now on this network. The Covexus team trained a model on 70 different devices that exceeded Meta’s Llama-2 and received praise from Nvidia’s CEO, but it gets buried under the din of tokens.

That’s also why the ETF is more than a harbinger. Grayscale and Bitwise both anticipate the answer from the SEC later this year, around August. If approved, they would slip a defective machine into the middle of America’s retirement funds. That does represent risk to any observer who wants to buy in blindly, but it also represents two things you want to see in a nascent ecosystem. Infusion of real money and exposure to daylight. Few things are more efficient at accelerating a network’s incentives than a regulator giving its blessing and millions of new shareholders watching every payout unfold under a prismatic lens. With this last note, note that the scrutiny that comes with the territory is what ultimately makes it grow up.

I, with my newfound optimism, would say you should keep a close eye on the right stuff. The plumbing is still young and leaky, and it will be tightened, as with all young, leaky systems. The promise underlying it all is what I want to highlight. Open, multi-participant, and nonproprietary AI, as opposed to the walled gardens of the major cloud providers that happen to own the largest server farms in the world.

We spent this whole piece describing weaknesses, but the reason we did that is that the object is so worthwhile that improving it should be a priority. That’s why you should hope that subnets will eventually be able to support themselves outside the patronage of the foundation, if and when that happens. That would be a crucial statement that the most powerful technology of our age does not have to be controlled by the fewest possible entities.

It would be worth waiting for.

—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.

Meta is building a data centre in Louisiana to run its AI workloads. To finance its costs, they set up a subsidiary and loaded it with debt. Pimco took on that debt, repackaging it into $30 billion in bonds that will mature in 2049 and placing them with pension funds, insurance companies, and retirement advisors. But the problem now is that the GPUs inside these data centres become e-waste within four years. People’s retirement money is being used to back hardware that will be obsolete even before the bonds mature.

In last week’s piece, I showed how GPU compute is becoming a tradable commodity with its own pricing indices and futures contracts. Companies are borrowing billions of dollars to build GPU clusters and selling capacity in advance to lock in revenue. Still, there is no public way to verify whether the same rack has been pledged to multiple lenders or to check its real utilisation.

Today, I will dig deeper into the financial layer, and how computing is actually being financed, and how they are pricing the risk and lending against it.

The Compute Lending Structure

So the question is, how do you lend against hardware that will be obsolete in four years?

The Meta deal is a good example of how the structure works. A data centre is built and then filled with GPUs worth approximately $27 billion. A lender puts most of the capital; in Meta’s case, it was Blue Owl. After this, Meta signs a take-or-pay contract, meaning it commits to pay a fixed rate for the compute capacity the facility produces over the full term of the financing, whether or not it actually uses the GPUs.

The lender here is betting entirely that Meta will continue making those payments, and as long as it does, the lender gets repaid regardless of what happens to the chips underneath. Pimco then takes that $27 billion in loans, repackages them into $30 billion in bonds, and places them with pension funds and insurance companies.

Every major GPU financing deal is structured the same way: the credit decision sits on the customer’s balance sheet, and the hardware comes second. CoreWeave, which builds GPU clusters that companies like Microsoft and Meta rent for AI workloads, has used this approach eight times now. Their latest facility raised $8.5 billion on this model.

And this approach isn’t new. Wall Street used this trick for mortgages twenty years ago, and it ended in one of the worst financial crises since the Depression. For the CDO machines of the 2000s, rating agencies did not examine the individual mortgages within a CDO; instead, they focused on the legal wrapper, assuming the mortgages would fail independently of one another.

Think of it like insuring a hundred houses spread across the country; each of them looks fine individually because a fire in Texas would not cause a fire in Maine. But what if all one hundred houses are in the same flood zone? When the flood comes, every single one will file a claim at the same time, and the insurer will be wiped out. The agencies assumed the mortgages were spread out across neighbourhoods. But when the housing crash occurred in 2008, prices dropped everywhere at once, and the safe layers that pension funds had been buying turned out to be the most exposed.

This is exactly what these GPU-backed SPVs are doing: lenders take a complex, rapidly depreciating physical asset that nobody can reliably price, wrap it in a legal structure from an investment-grade counterparty, and produce a debt instrument that pension funds and insurance companies can buy.

But GPUs are worse than mortgages because every single GPU in these financing vehicles depreciates with Nvidia’s product release cycle. When Nvidia ships a new generation, every chip from the previous one loses value on the same day. And every one of these deals depends on a handful of buyers. Five hyperscalers account for 70% of all AI compute demand, and their capex budgets are the only variable that determines whether these contracts will be renewed.

Mortgage defaults at least had geographical variation — a housing crash in Florida didn’t automatically trigger one in Oregon; GPUs have no such cushioning, and yet lenders are still underwriting every one of these deals as if the risks are independent of each other.

The biggest counterargument to this is that depreciation doesn’t matter as long as the customer’s contract repays the full loan amount before the GPU becomes obsolete, which might be true at the individual-deal level. But that also puts the operator on a non-stop treadmill, where every new GPU generation that makes the current lending ‘safe’ also forces the operator to raise fresh debt for the next one. Nvidia has recently moved to a one-year product cycle, which only makes the treadmill faster. The whole market is making the same bets, just through separate legal entities that need to be renewed every year.

The rating agencies understand this well. Estimates on such new hardware are meaningless, which is why they price the contract rather than the chip.

But Can You Verify & Price the Chips?

Pricing the contract solves the lender’s immediate problem, but the underlying problems remain untouched. If no one is checking the chips, does that mean no one can tell whether those chips actually exist? If yes, are the chips running? What’s the utilisation so far? And what if the same chips have been pledged to multiple lenders?

None of these questions can be answered without verifying the underlying asset. The verification problem has a long history in commodity markets. In 1963, a vegetable oil trader in Bayonne, New Jersey, convinced banks he had 900K tons of salad oil stored in tanks in his facility. When the inspectors checked the tanks, they found them full, but with water at the bottom and only a thin layer of oil on the surface. Banks had lent this trader $180 million against warehouse receipts for oil that did not even exist. The shortage turned out to be 1.8 billion pounds, causing two brokerage firms to collapse and nearly bringing down American Express.

Fifty years later, it happened again in Qingdao, when a Chinese trading firm used duplicate warehouse receipts to pledge the same metal to multiple banks simultaneously. A single batch was pledged an average of ten times. And it was only discovered when a Communist Party corruption probe caught the company’s chairman.

Nearly every physical-asset market that scales fast enough eventually produces this kind of fraud. The documents suggest that the collateral exists, but because no one verifies its physical reality, companies misuse it to make multiple claims against the same asset. The current verification infrastructure for GPU-backed lending has the same leaks. In traditional lending, when a bank finances a physical asset, it files a public notice claiming the asset as collateral so other banks can verify it before lending against the same asset again.

In the US, this runs through UCC filings, but these filings often contain errors. A study of the Florida system found that nearly a third of all filings there were legally void because of typos in the borrower’s name. A single wrong character can easily erase the lender’s entire claim.

Now imagine an on-chain version of this registry, which is immutable and resistant to those errors, designed specifically for GPU financing. USD.AI, for example, has built a system that turns any physical GPU into a tokenised financial instrument with real legal enforceability.

The borrower can sell the hardware to a tokenising agent, who creates an NFT that functions as a digital deed of ownership under UCC Article 7, the same law that governs warehouse receipts for physical commodities such as oil and copper. This NFT gives the holder a legally enforceable claim to the physical GPU, just as a warehouse receipt gives its holder a claim to metal in a vault.

The borrower gets the NFT back under a bailment agreement that lets them operate the hardware while it serves as collateral for an on-chain loan. These loans are set up inside bankruptcy-remote Delaware SPVs with perfected liens. The hardware is also monitored in real-time, and the default triggers are hardcoded into the smart contract itself. This means if telemetry ever goes dark for more than 24 hours or if a hardware attestation check fails, the contract freezes the position and routes it into a liquidation waterfall.

Another protocol, GAIB, takes verification even further by continuously running several cryptographic proofs. It verifies that the hardware is where the borrower says it is, confirms whether it has been moved or swapped, and checks that the GPU is actually running compute jobs and generating revenue.

Validators can stake tokens to participate in this verification, and if they attest to something false, their stake is slashed. This is the same security model that secures hundreds of billions of dollars in value across Ethereum and its restaking ecosystem.

The verification with GAIB is also embedded at the hardware level. Every H100 chip that ships from Nvidia’s factory has a unique cryptographic key physically burned into its silicon, serving as a fingerprint that cannot be changed or forged. So when a lender or a protocol wants to verify that a specific GPU is genuine, they just have to send a challenge to the chip, and the chip signs a response with that key. The signature is then checked against Nvidia’s certificate authority, which verifies that it is a real H100 chip, that it is running authentic firmware, and that it has not been tampered with.

Researchers have further pushed this to software-only fingerprinting that can measure the entire latency profile of a chip’s processing units and produce a physical signature unique to each individual die, much like a human fingerprint is unique to each person. This can distinguish individual GPUs with 100% accuracy and pin their physical location to within 44 kilometres of a claimed datacenter.

Last week, I argued that the compute market needs a settlement layer in which collateral is verifiable by anyone so that the forward curve becomes a public good. That layer is now being built, and it runs on crypto rails because no existing financial infrastructure can settle a compute credit between a lender in Tokyo and a GPU rack in Virginia in under a second. No paper-based filing systems can provide cryptographic proof that a GPU exists, is operational, and is not already pledged to someone else.

The question is whether it will be adopted fast enough. Over $200 billion in private credit has already been deployed for AI hardware, and that debt does not remain with the original lenders. It flows through insurance companies and target-date retirement products until it reaches the savings of people who have no idea their 401(k) is just four layers of intermediation away from a rack of GPUs.

Nvidia sits at the epicentre of all this. Their standing offer to buy back compute at pre-agreed levels is what makes these clusters bankable in the first place. But the floor is set by the same company whose next chip release lowers it.

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 how do you reliably sell something the law doesn’t let you sell? Money travels downhill to the kind word. A payday advance is a loan that can get someone sued. Not licensed to write becomes credit default swap insurance.

Last week, Vlad Tenev switched on Robinhood’s own blockchain and Stock Tokens at a keynote named “The World Is Flat.” Cute, because so is what we’re buying.

When you grab the Nvidia token, you get the price of Nvidia stock and none of the rights. If the company blows up, you have no claim on it. Well, that’s a risk that something tokenisation comes with. A work in progress, Ondo has filed an SEC filing, but as of today, unresolved.

In Robinhood’s case, what you are owning are debt securities.

The debt securities are issued by Robinhood Assets (Jersey) Limited. You lent money to a shell company on a tax island in the English Channel, and it pays you whatever the stock does.

I want to take you back to a year ago, in June 2025, in Cannes. Robinhood handed European users $5 in free “OpenAI” and “SpaceX” tokens to hype the launch. Private companies. You can’t buy their shares at any price unless they invite you. OpenAI saw the tokens with its name on them and posted a warning saying, “We did not approve any transfer. Please be careful.” Elon Musk, who cofounded OpenAI and now hates it, replied by calling the company “fake.”

Robinhood’s CEO, Vlad Tenev, then admitted that the tokens aren’t “technically equity”; they just give people “exposure.”

So why wrap a stock as debt and run it out of Jersey? Because the SEC has a map.

Equity is when you own a piece of the company.

Debt is when the company owes you money.

Robinhood stock tokens give you debt that looks like equity. You’re not a shareholder. If you buy Nvidia through this instrument, Nvidia doesn’t know you exist.

You’re buying a debt instrument from a Jersey company. That Jersey company owes you a payment based on Apple’s price movement.

If Apple goes up 20%, the Jersey company owes you 20%. Good.

If the Jersey company goes bankrupt, you’re a creditor. You join the queue of people waiting to get paid. The actual Apple shares the Jersey company holds might cover you. Might not. Depends on how messy the bankruptcy is.

If Apple itself goes bankrupt separately, you’re even further removed. You don’t hold Apple shares. You hold a promise from a company whose value was tied to Apple shares that are now worthless.

So why is Robinhood doing this at all? Start with the worst day in its history.

In January 2021, GameStop was squeezing, and Robinhood users were piling in. Then Robinhood had to turn off the buy button. The two-day settlement system behind US stock trades hit it with a collateral bill in the billions that it couldn’t post overnight, so it froze buying on the hottest stock in the country. Retail felt sold out. Congress hauled Tenev in to explain. The trust never fully came back.

This could be Tenev’s fix, five years later, by deleting the two days. A stock that settles on a blockchain clears in seconds, not T+2. No settlement lag means there will be no giant collateral call. This gives zero reason to kill the buy button again. That’s the case he’s made since January 2026, and it’s why Robinhood handed the SEC a 42-page proposal in 2025 asking for a tokenised-asset rulebook.

In January 2026, three divisions of the SEC released a joint statement on tokenised stocks. They split the mess into buckets. The real stuff is in one bucket. A company puts its own shares on a chain, and you own real equity. Another was called the ‘linked security.’ That’s a token that some third party issues to mirror a stock's price. It doesn’t carry any of the company’s obligations, and it doesn’t give you any rights as a shareholder. The SEC explained that a linked security could be dressed up as debt, for example, as a structured note. The holder assumes a risk that a typical shareholder would never take. If that outside party goes bust, that’s your problem now.

Two months later, in March, the SEC and the CFTC issued a joint release that left the January rules in place.

Robinhood built the linked security out of Jersey on purpose.

The same SEC memo described a cousin product, the “security-based swap,” which is a side bet that tracks a stock. Trouble is, you can’t sell one of those to normal people. Federal rules block offering a swap like that to anyone who isn’t an “eligible contract participant,” which means a rich or institutional investor. A debt security has no such wall. Slap the words ‘structured note’ on it, and you can hand it to a 19-year-old with ten bucks. Robinhood chose the wrapper that appeals to the largest crowd with the least hassle.

They locked Americans out.

Stock Tokens went live in more than 120 countries. Not the US. Also shut out Canada, the UK, Switzerland, and the UAE.

Europe gets a different wrap. Robinhood’s first set of tokens, now called Classic Stock Tokens, launched in Cannes in June 2025 and is subject to EU MiFID II rules. Each token is issued by a European Robinhood entity and is 1:1 backed by a real share held in custody. Since then, the catalogue has grown from 200 to over 2,000 assets, with a minimum buy-in of 1 euro. So Robinhood already runs the backed, real-share version in Europe.

This whole model relies entirely on the transparency of public markets. Because a public stock at least has a price set by the market every second. Anthropic and OpenAI don’t trade. So the token’s “value” leans on somebody’s guess about a company that publishes nothing and answers to no outsider.

Now, the 7% yield comes from a separate product called Robinhood Earn. US-only, oddly enough. It pays you for lending out a stablecoin called USDG. Your dollars flow into a Morpho lending vault run by a shop called Steakhouse Financial, then get pushed further downstream into other protocols like Ethena and Maple. Morpho holds around $6.6 billion in deposits. The rate floats with demand. Robinhood bought insurance through Lloyd’s of London in case a smart contract gets drained.

Count the hands your money passes through there. Robinhood, then Morpho, then Steakhouse, then whoever Steakhouse lends to. The insurance covers a hack. It does nothing when the yield drops to zero. Rates move, and a lending yield drops when borrowing demand dries up, the same as any money-market rate. The thing to watch is your principal sitting four counterparties deep. The Lloyd’s cover pays out if a contract gets hacked. It does nothing if USDG loses its peg, or if borrowers default in a wave the vault can’t cover. That’s how money like this usually goes missing.

Because the tokens sit on a blockchain, you can post them as collateral and borrow against them. Handy. But a smart contract can’t look up Nvidia’s price, so it asks an oracle, a price feed, usually Chainlink. Feed the contract a bad number, and it’ll happily liquidate people or release funds it shouldn’t. Rigged oracle prices were one of the top ways DeFi got robbed in 2024 and 2025, with tens of millions lost across dozens of separate hits.

The only piece of real, voteable, own-a-slice equity anywhere in this whole setup is Robinhood’s own stock, HOOD, trading on Nasdaq the old kinda way. They kept the ownership for themselves.

Why? Every trade runs through Robinhood’s spread; the chain is Robinhood’s, and a busy new product overseas keeps HOOD’s story growing without a US regulator in the room. Ownership-free tokens are a cleaner business.

Robinhood Chain is an Arbitrum Orbit network that runs on ETH for gas and skips a token of its own, which kills the usual speculation and also means Robinhood earns nothing from issuing one. The plan is to make it the settlement rail for tokenised everything, stocks and ETFs, stablecoins and commodity perps, eventually private-company shares, all in one place, trading around the clock. If that works, Robinhood, a brokerage that routes your order to somebody else’s exchange, becomes the exchange, the clearinghouse, and all at once.

In Robinhood’s defence, the rules are changing fast. Atkins has buried the sue-first era at the SEC and is drafting an innovation exemption, a sandbox that lets firms run tokenised products under lighter rules while the permanent ones get written. The CLARITY Act has been sitting on the Senate calendar since June. Once cleared, the grey zone Robinhood is renting starts to shrink.

That’s likely the whole plan. Coinbase ran for years before it got its blessing. Kraken did the same. Build the thing in a soft spot in the law, prove people want it, then bring it home once the rulebook catches up and you can screw the protections back on.

Seen that way, the token you buy today is a rough draft of a better one.

A few days before Robinhood’s launch, on July 2, Ondo put real tokenised stocks on Ethereum through an SEC-registered transfer agent, with actual shares in custody with ownership behind them. On-chain voting for 250-plus companies. Coinbase also sells these, with dividends paid straight to you. Both are legal for Americans right now.

Going full SEC-registered in America costs more and puts a regulator in the room.

Robinhood already eats that cost in Europe, so the Jersey wrapper is a choice. It upgrades the day the law forces it, or the day a rival hands Americans the real share and walks off with the customers.

Build now, regulate later, right?

—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.

Google is one of the three largest cloud providers on Earth, and right now it’s buying $920 million worth of compute per month from SpaceX. A rocket company.

That is how broken the GPU capacity market is. There is no benchmark for pricing or a way for lenders to hedge the hardware they are financing; everything runs on blind capital allocation. And this is about to change because the CME and ICE have announced they will list futures contracts for GPU compute time.

Compute is being pulled into a full-scale capital market, the same way electricity was in the 1990s, and today I will dig into what this new, liquid forward curve, powered by stablecoin settlement, could unlock for the largest infrastructure buildout since the railroads.

Making Compute Tradable

When I say compute is following electricity’s path into a capital market, I mean it very specifically, and understanding it will tell us exactly how this market can be built.

There is a huge distinction in commodity markets between what traders call stock goods and flow goods. Oil, for instance, is a stock good because you can store it in a tanker until there is a buyer. You can stockpile crude when prices are low and sell when prices spike. Compute, on the other hand, is a flow good because you rent a GPU for a period of time and pay for access to its computational power over that window. And any capacity that goes unused during that time is gone permanently.

A GPU sitting idle in a rack is not a “stored compute” any more than a disconnected power plant is “stored electricity” because in both cases the valuable product is the flow - GPU hours or kilowatt-hours and not the physical machine producing it.

This is a critical difference for pricing because stock commodities have a built-in stabiliser in the form of inventories that flow commodities lack. The inventory can be released during periods of high volatility to counter price rallies. Flow commodities don’t have such a buffer; that’s why compute spot prices often swing hard.

In mid-2025, compute spot prices had collapsed by 70% in 18 months as a lot of new supply flooded the market alongside Nvidia’s next-generation Blackwell chips, making the H100s less desirable. But then they again surged 48% in just four days this year when demand spiked due to HBM manufacturing, and there was no inventory anywhere to absorb it. For AI companies, whose training runs cost tens of millions of dollars, and lenders who have financed over $120 billion in data-centre credits for this hardware, the volatility with no hedging instrument is an existential problem.

And there is a second problem that sits on top of this. A barrel of oil is always identical to every other barrel anywhere in the world, which is exactly what allows it to be traded on an exchange without inspecting the physical product. But an H100 in Virginia is a measurably different product from an H100 in Iceland because the chip, cluster configuration, and the neighbouring workloads will all affect its real-world performance.

Benchmarking data across global GPU providers has shown up to 38% performance variance for nominally identical hardware. Electricity had this same problem in the 1990s: power on the Texas grid was a different product from power in the mid-Atlantic because transmission and local demand created different conditions at every point in the network. The only way for them to solve this was by giving each node its own price, which can be quoted as a spread against a reference benchmark, and that reference benchmark is exactly what the compute market is missing right now.

SF Compute has built a live order book for GPU time, where buyers and sellers trade hours, much like you would trade any commodity on a spot market. The logic is that once you have a liquid spot market, you can use it to derive an index price from trading activity. And that same index price could then be used to build a cash-settled future on top of it.

Once a data centre can sell a futures contract and lock in revenue months ahead, it can approach its lenders and show them that its revenue is hedged, thereby securing lower interest rates and building more capacity. Which, in turn, can bring overall compute prices down for everyone.

Another company, Silicon Data, has built a daily index called SDH100RT, which has been live on Bloomberg terminals since May last year and has so far aggregated 3.5 million data points from global providers into a single benchmark for the cost of an hour of H100 GPU time. The CME’s newly announced futures contract will settle against its index. There are a few more firms racing to build this now because becoming the reference price allows them to capture a thin layer of every transaction in the market for as long as it exists.

The electricity market went through this exact phase too: when, in 1993, Nord Pool opened the first electricity futures exchange, more than 200 new power marketing firms launched. It took the industry a decade to debate whether electricity was even a commodity in a legal sense, but today it’s a $6 trillion annual market. Compute is following a similar timeline right now.

The Dealer in the Middle

So we have what we can all call the first Compute spot markets with price indices in some form, and the exchanges have announced intent too. But between an index price on a Bloomberg terminal and a functioning capital market, there is another critical layer that makes it all work, and it looks very different from trading.

A compute futures market will not function the way a stock exchange does, in which standardised shares change hands between anonymous buyers. It will be driven by dealers who sit between GPU owners wanting to lock in revenue and AI companies wanting to lock in costs.

For example, let’s assume a data centre has plenty of H200s available in the US starting in October. There is a startup that needs 500 GPUs but only cares that the interconnect is InfiniBand (a communication medium for GPUs), not where the machines are located. Now, this is a very specific ask, and someone will need to take the custom order on one side and hedge the exposure against the standardised index on the other.

And it’s not something new; every commodity in the past has needed such a layer, where someone can absorb the messy linkages of the physical product and translate them into fungible units that an exchange can trade. An H100 sitting in a rack is just a bespoke contract that nobody else can price. It generates revenue for only one party under a private agreement, and the rest of the financial system cannot even touch it. But if you can wrap it with an index price and a public settlement layer, it can become a live commodity that a lender can hedge.

In 2023, CoreWeave borrowed $2.3 billion using just Nvidia GPUs as collateral, the first time H100 hardware had ever secured a loan. Its most recent facility got an investment-grade rating from Moody’s, based on Meta’s creditworthiness, not CoreWeave’s, because Meta signed a take-or-pay contract that requires payment whether or not they actually use the compute.

This is also where crypto rails become load-bearing. The buyers and sellers of compute are global, and none of them can open a US commodities exchange account with CFTC approval. But a crypto wallet can settle a stablecoin payment, and any wallet can hold a tokenised compute credit.

GPU export controls already reveal how compute access is geopolitically stratified, as Nvidia has been unable to ship frontier chips to China and dozens of other countries. A compute futures market settled in stablecoins can enable researchers and startups outside the export-controlled tier to access compute pricing and hedge their costs through infrastructure that routes around constrained access, much like stablecoins already do in Argentina and Nigeria.

The Liquid Forward Curve

Right now, building a GPU cluster means borrowing millions of dollars against revenue you cannot lock in, because no instrument in global financial markets exists to do so. But a liquid forward curve allows companies to borrow against hedged revenue at a lower rate than they would against an unhedged position. This means you get a cheaper cost per compute hour. So who is going to build the settlement layer for the forward curve?

The only solution needed right now is a settlement layer in which the collateral is verifiable by anyone and the forward curve is a public good. Currently, there’s no way to verify the condition of the pledged hardware, whether it has been double-collateralised, or its actual utilisation rate. But if the GPU and its revenue stream are tokenised as on-chain assets, every lender can verify the collateral in real time, making the forward curve publicly visible rather than trapped in bilateral negotiations.

And then there is also an upcoming generation where AI agents will buy compute per inference call, and they cannot open bank accounts. Crypto is the only payment gateway that can clear a micro-transaction between an agent in Tokyo and a GPU rack in Virginia in under a second.

Now there are severe counterweights to this, because the GPU supply is brutally concentrated right now. The top hyperscalers in the market control 78% of global IT power. Nvidia holds over 80% of the high-end AI chip market, and its product launch schedule can move the entire market. The standardisation is a bottleneck, but financialising an asset class during a buildout boom can make it more contagious.

More than $120 billion of AI infrastructure debt has been moved off balance sheets into Wall Street-funded SPVs, and much of it into corporate bond funds sitting inside target-date retirement products, where the person whose savings are exposed has absolutely no idea. I believe there’s a high likelihood that the financing models used to build the infrastructure embed assumptions about residual hardware values that available data does not yet support.

The electricity market did not stop at the generator. It went all the way through to the wall socket and into the price of everything that uses power. The compute market has a lot of wiring left to string too!

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.

In the 1970s, Bruce Bent and Henry Brown created the first-ever Money Market Fund. The idea was stupidly simple. Banks in America were capped at a 4.5% interest rate on savings due to a depression-era rule. While Treasuries were paying over 9%, you needed a minimum investment of $10,000 to buy them. So Bent and Brown decided to pool small deposits, buy Treasuries at scale, and pass the yield back. Today, money market funds are roughly an $8 trillion product.

Stablecoins have been running a similar play, except that the asset this time is private credit, a $2 trillion market where you need at least a million dollars just to get through the door. Yield-bearing stablecoins are being used to pool small deposits and channel them into credit.

Today, I will dig into how this happened and how Goldfinch, the first-ever attempt to do this with real-world money, just shut down, leaving $56 million in depositor funds stuck in Kenyan motorbike loans.

How Stablecoins Became the MMF of Private Credit

Banks in America used to provide roughly half of all debt capital to businesses and consumers in the 1990s, but today that number is roughly 20%. This was because, after 2008, new capital rules came into effect, making it too expensive for banks to hold leveraged loans on their balance sheets. Due to this, they withdrew from mid-market lending altogether, and private credit funds took over.

Apollo, Blackstone, and KKR raised capital from pension funds and insurance companies to start lending to companies that banks had walked away from, charging a steep premium because these borrowers had no other option.

The market has grown from under $200 billion in 2008 to over $2 trillion today, and almost all of that capital has come from institutional investors writing checks of $5 million or more.

One of the biggest reasons why there is a million-dollar minimum for access to private credit loans is that they are hard to manage. Every deal needs due diligence, restructuring, and years of monitoring. Running a fund with ten institutional LPs who each put in $50 million is far more manageable than one with retail investors, each putting $500, and often even the economics don’t work at scale. That’s why, for the last decade, only pension funds and insurance companies had access to this type of yield, earning anywhere between 8% and 12%.

This is when yield-bearing stablecoins changed the game, much like when Bent and Brown opened Treasury access in the 1970s. The paperwork still happens at an institutional scale, with funds like Apollo underwriting and managing risk, but now a tokenised feeder fund can accept deposits of any size and channel them into the institutional strategy without needing to manage thousands of individual investors.

Apollo recently launched a tokenised fund, ACRED, and has already seen $109 million flow into its diversified Credit Fund. Investors can even deposit it as collateral on Morpho to borrow against and loop for leveraged yield.

Figure has built an entire on-chain lending stack that has originated $21 billion in loans, secured a Nasdaq listing, and launched YLDS, a yield-bearing stablecoin with $376 million in circulation. Other protocols like Pyse and Glow went even more granular, tokenising solar energy projects so that investors can invest a few hundred dollars to fund installations in developing markets and earn APYs paid from monthly electricity bills.

This does not mean minimums have disappeared from the funds themselves. ACRED still requires $5 million for direct entry. But once the fund is tokenised, the token trades on secondary markets with no minimum, and it works with DeFi in ways traditional fund shares never could.

In traditional private credit, your capital is locked for years with quarterly redemptions capped at 5%. But on-chain, it’s composable and liquid around the clock. And for firms like Apollo and Figure, this gives them access to $315 billion in stablecoin capital actively chasing yield, and tokenising their funds provides direct access to that pool as a new distribution channel without having to build retail infrastructure from scratch.

A year ago, total on-chain private credit was just $400 million; today it’s $5.87B, a 15x jump in twelve months, which is still just 0.30% of the $2T global private credit market with half of all the new stablecoin supply in Q1 2026 being from yield-bearing, which means the bulk of new stablecoin capital is now chasing active yield instead of just a dollar-denominated peg.

And because every dollar of on-chain credit can be used as collateral and looped through DeFi protocols, the effective financial activity it generates is multiples of the dollar amount.

Take ACRED, for example. An investor deposits $10,000 on Morpho, borrows $7,000 in USDC against it, buys more ACRED with that USDC, and deposits it as collateral again. That one deposit will now be generating over $17,000 in credit exposure. Compare that to traditional private credit: the same $10,000 will just sit in the fund for 5 years with nothing. But on-chain, it compounds across multiple layers simultaneously, which is why the market can scale faster than its raw dollar size suggests. But it also means that when a loan underneath goes bad, the losses can ripple through every layer of the loop.

Tokenising doesn’t mean the underlying risks are also mitigated. Often, these risks go unnoticed as new money keeps coming in, and these fresh deposits fund any redemptions. But as inflows slow, the differences between token promises and what the loan can deliver start to show up. Investors try to exit, but liquidity is lacking, or the token’s price disconnects from its underlying value.

Something like this happened with Goldfinch; it was one of the first protocols launched in 2021 to put private credit on-chain, and it recently had to shut down after $56 million in depositor funds was trapped across Kenya and Nigeria.

What Goldfinch Got Wrong

Goldfinch raised $25 million from a16z in 2021 to route capital from crypto, which was earning barely 2 to 3% in DeFi lending pools at the time, into businesses across Africa and Southeast Asia whose borrowers were paying 15 to 25% on loans because their local banks wouldn’t serve them.

The idea was to let anyone with USDC deposit into a Goldfinch pool, and the smart contract would then route it into a borrower tranche within seconds. But underwriting a loan to a motorcycle-financing company in Nairobi means someone has to understand Kenyan transport economics and verify the borrower’s books in person. They also might have to physically show up at their office if repayments stop.

But these things are not possible on a blockchain. Once USDC was converted into Kenyan shillings and deployed in the lending book, depositors had no way of knowing how their money was being used, whether the borrower was financially healthy, or whether the loan terms were even being honoured. All the information that mattered for the loan’s performance was off the chain and held by borrowers in countries most depositors had never visited.

Which is why it took months for anyone to notice that Tugende Kenya moved $1.9 million from its $5 million facility to Tugende Uganda in 2022 without authorisation. Nearly 40% of the loan had been redirected to a different legal entity in a different country. At the same time, depositors continued earning what they believed was 10 to 12% interest, completely unaware that the capital backing their returns had gone somewhere the loan agreement never covered.

A traditional private credit lender finding a covenant breach this severe would call the loan and force restructuring within days, but Goldfinch depositors found out through a governance forum post, and their only option was to vote on a proposal that had no legal authority to seize assets or audit what was left.

By 2023, Tugende had defaulted in full and gone completely dark. Of the 24 pools that Goldfinch originated over its $113.3 million lifetime, only 13 ever fully repaid. The remaining 8 held $53.82 million in outstanding loans, and not a single one was performing on its original terms. Most were in restructuring and paying less than $51,000 per pool per month, which means that recovering the full $53.82 million at that rate would take between 8 and 15 years.

Goldfinch had absorbed all the risks of currency volatility and limited credit histories in emerging markets, while having almost none of the infrastructure that traditional lenders spend decades building and managing to mitigate those risks. For example, a bank lending in Kenya has local offices and regulatory relationships that give it leverage when deals go sideways.

But Goldfinch had routed capital from anonymous wallets across the globe into the same category of borrowers without any of that scaffolding, widening the information gap between lender and borrower beyond that in a conventional deal and leaving depositors with close to zero ability to intervene when things fell apart.

Moving money on-chain accounts for only about 10% of what lending requires. The other 90% is underwriting and capital recovery, which is hyper-local and expensive to build. Underwriters in this category need to build the credibility floor for an entire asset class that is still earning its right to exist. Every dollar lost to weak underwriting makes the next institutional partner harder to bring on-chain and the entire asset class less credible. And the hard part of credit has nothing to do with what happens on-chain, and anyone building in this space who doesn’t understand that is building the next Goldfinch.

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.

A ten-dollar note from the Bank of Cincinnati was not worth ten dollars everywhere. Maybe nine. It can be any number. Maybe nothing, if the bank had failed and the news hadn’t reached his county yet. The most famous of those books came out of Philadelphia, Bicknell’s Counterfeit Detector, a price list for money itself, reprinted on a schedule because the worth of a dollar changed depending on whose name was on it.

That was America from 1837 to 1863. Any bank with a state charter printed its own notes, and getting that charter was easy. Michigan kicked it off in 1837 by letting you open a bank with barely any requirements, no act of the legislature needed. Thousands of different notes ended up moving around the country at once. Around a third of the paper in circulation was outright counterfeit.

Each note was a bet on the bank behind it. The system died during the Civil War, when the government printed one national dollar, partly to fund the war, but also because trusting eight thousand private dollars had worn the country out.

On June 22, the Senate overwhelmingly voted the 21st Century ROAD to Housing Act (85 to 5), and the House passed it the very next day. Hidden inside this housing bill is a rule that bans the Federal Reserve from creating a CBDC for the public until 2030.

That is why we are looking back, digging up the ghosts of the Free Banking Era.

The dollar in a Venmo balance is a promise from a bank to hand you one, and if that bank dies, your money is at risk past the FDIC line. A CBDC (Central bank digital currency) cuts the bank out. It allows the holder to possess state money directly, digitally.

The Senate killed that idea for two reasons.

A government digital dollar could track every cent you spend and freeze your wallet on command, the way China’s digital yuan already does.

Secondly, the banks fought it hard because money sitting directly at the Fed is money that will never be in their deposits. They’d lose the float their whole business runs on.

Now, even the man holding the pen looks unsure about what he’s signing. On June 24, an hour before the signing, Trump cancelled the ceremony and demanded a voter-ID bill the Senate had already killed. The ban likely becomes law anyway.

Fine, the government won’t mint a digital dollar. At the same time, it hands the job to private firms instead. That means 1840 walks back in the door.

Even now, with the GENIUS Act signed in July 2025, the bar is mostly about reserve quality, rather than hard gatekeeping. More than a dozen firms are lining up for charters to print their own dollar. Every fintech wants a branded coin.

The market is about $312 billion now. Tether’s USDT and Circle’s USDC are roughly 80% of it. Then PayPal’s PYUSD, Ripple’s RLUSD, and Paxos minting white-label coins for whoever wants one. Each one says the same thing the Bank of Cincinnati said. Trust us, it’s backed.

It’s not 1840, and Jim Carrey is out here struggling to prove he isn’t his own clone. We believe in nothing. Do we? Which is good and bad. The distrust is what makes issuers prove their backing, and also what they’re counting on staying shallow, since a crowd that suspects everything but verifies nothing is exactly the crowd you can hand a private dollar to.

A wildcat bank said its notes were backed by silver in the vault. The vault was often a barrel of nails parked in the woods where no inspector could reach it. A stablecoin backs its tokens with US Treasuries and posts the receipts every month. On the collateral, that is a big difference, and it runs in the stablecoin’s favour.

Leave the collateral part aside, and then we have a problem. A dollar equals a dollar because everyone agrees on three things at once. The issuer is good for it. The reserve is real, and you can reach it. And if the thing starts to tip, someone catches it. Hold all that together, and a dollar is just a dollar, and you stop thinking about it.

When one of them wobbles, money gets priced issuer by issuer, the way it was in 1840. It is how money works.

Even the checking account works on the same principles. The government underwrites all three at once, so you don’t see it. Stablecoins are where you can watch the machine with its cover off.

Tether is the largest issuer in the world and the least transparent. Its reserve reports have been picked apart for years. It settled with the New York Attorney General in 2021 and admitted it did not always fully back its reserves the way it claimed. It had been lending billions of its backing to affiliated companies. Circle is the good guy, the regulator’s pet. It is audited monthly and has been public since 2025. But then what happened in March 2023? Circle had $3.3 billion of USDC reserves frozen inside Silicon Valley Bank when SVB collapsed. USDC dropped to 87 cents over the weekend until the government backstopped SVB’s deposits. Good reserves, and yet, one dollar was worth 87 cents for about 60 hours. Because what cracked was the belief that it was reachable.

Every name in payments now wants its own dollar. PayPal has PYUSD. Ripple has RLUSD. There’s USDG, run by a consortium, plus a wave of bank tokens coming from the likes of JPMorgan and Western Union. In December 2025, the OCC handed trust-bank charters to Circle, Paxos, and three other crypto firms, and the line behind them is long. Tether, shut out as a foreign issuer, spun up a separate US coin called USAT just to get back through the door.

Read: Charters, Keys, and Control 🔐 - by Prathik Desai

Read the words on the tin, though. These are “trust bank” charters, not insured-bank charters. The Fed is holding out stripped-down accounts for them that don’t have overdrafts and don’t have access to the emergency lending window, which is the thing that actually saves a bank when the run starts.

And the names you’re handing the money to have records.

Paxos, the company issuing PYUSD and a half-dozen other branded coins, was ordered by New York’s regulator to stop minting Binance’s stablecoin back in 2023. Some of the newer tokens aren’t even backed by cash. Ethena’s USDe leans on a derivatives trading strategy to hold its peg.

The Act bars these issuers from paying you interest, so they route around it. Coinbase pays “rewards” on USDC. PayPal dangles 3.7% on PYUSD. The yield reels your money in. Unlike standard bank accounts, which are backed by FDIC insurance, this money has zero safety net.

Since stablecoins back their tokens with US Treasuries, Washington still sets the money supply. The interest on those Treasuries flows to the issuers. Tether cleared around $10 billion in profit in 2025 with about 100 employees and holds more US Treasuries than Germany does.

But if it all goes bust, the loss is all yours. A run is when everyone leaves at once through a door meant for a trickle. Stablecoin redemption is a narrow door. Most people can’t redeem with Tether at all. They sell to a handful of arbitragers, and Tether averages around six a month, with a $100,000 minimum to redeem at the source.

If Tether breaks tomorrow, a hundred billion in tokens reprice to 0. Tether owns so many of the Treasuries that when it panics and dumps them, it shakes the Treasury market itself.

Washington steps in because the alternative is a coordinated credit and liquidity freeze.

And the government doesn’t even need a global meltdown to do it. It saved Lockheed in 1971 with a $250 million loan guarantee, over a tie-breaking vote, to protect 60,000 jobs and the Pentagon’s biggest supplier. An admiral watching it happen called it “a new philosophy where we privatise profits and socialise losses.”

When Penn Central, the country’s largest railroad, collapsed in 1970, the government let the company die and then spent public money building Conrail to absorb the tracks anyway, because the trains had to run.

A dollar token used as money by 250 million people clears that bar with ease. The state has a long history of refusing to guarantee private risk right up until the private risk threatens to disrupt public infrastructure.

There are ways to close this. What if, instead of a bank like SVB, which can collapse, they make the reserves of the issuers park directly at the Fed? The risk of a run largely disappears because the Fed can’t fail like a regional bank can. Or make them buy real deposit-style insurance, funded by the issuers, so the backstop is paid for before the crisis.

The government could try to tax the yield on the Treasury they earn on public debt and return it to the public that bears the risk.

But we would hate all of these, wouldn’t we? Reserves at the Fed mean the Fed is now the backstop. Insurance means a government insurer standing behind it, the way the FDIC leans on the Treasury when things get bad enough. Taxing the yield means treating these firms as public utilities. We don’t want to drag the government back into the middle of the money. That’s why the ban on CBDCs in the first place.

Going back to human history again, in the Athens marketplace around 375 BC, the city paid a public slave to sit among the bankers’ tables and decide whether your silver was real. He cut the plated fakes in half and handed the good coins back. He even waved through foreign knockoffs of the Athenian owl, as long as the silver was solid, and a law forced every merchant to take whatever he approved.

Twenty-four centuries later, have we arrived right there? The pleasing answer is that big shifts always come with holes, which get filled with time. Maybe give it a decade for the reserves to get cleaner, the rules get teeth.

But the trade-off here is what you want to look at.

Look at what you’d be giving up. Right now, your dollar sits behind FDIC insurance and a Federal Reserve that can print its way out of a panic. It is slow and one of the most backstopped things you will ever hold. The bank lent it out the day it landed. The Fed dropped reserve requirements to zero in 2020, so by law your bank doesn’t have to keep a single cent of your money on hand. What stands behind it is FDIC insurance, and that fund holds about $154 billion against the country’s insured deposits. Round it out, and that’s roughly a penny and a half set aside for every dollar it promises to cover. It still cannot handle a synchronised, systemic panic. If the fund is wiped out, the FDIC must immediately activate its backup line of credit with the U.S. Treasury or coordinate with the Federal Reserve to print emergency liquidity.

In a single month of 2023, three of the biggest bank failures in American history hit at once, Silicon Valley, Signature, First Republic. To stop the run, regulators tore up their own $250,000 insurance cap and made every uninsured depositor whole, the exact emergency move they’d run for a stablecoin. It cost the fund about $20 billion.

Stablecoins are cheap and fast. They are open all night, and the reserves sit in Treasuries you can watch on-chain, and the best issuers post the receipts every month, which is more than the bank reveals about where the deposit went.

Would you swap the first for the second?

I would, without thinking twice. Maybe you would too. You read this article, you already live in this stuff, you know what a depeg is and where to look when one starts. But the test is on normies. The person is getting paid in USDC because it’s the easiest dollar they can hold. The shop is taking PYUSD because the fee is lower. Most people never read the risks behind an easy system.

After all, the convenience spreads faster than the understanding does.

Your money is only as good as its issuer, and the issuer is only as good as the state that agrees to stand behind 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.

While doing that, would you build physical branches? Paper money? A “wet” signature on a form, witnessed and filed, to prove who you are? Passbooks, plastic cards, cheque books, and a mortgage that needs 17 - 20 pieces of paper.

This is what Brett King asks you in his book, Bank 4.0. Today, I am reviewing his thesis to see what parts of his future showed up.

“No, I’m sorry,” King writes. “That’s just plain crazy talk.”

He’s right, that’s why I kept reading after the 50 pages (that’s my rule of thumb to decide whether to drop or go on with a book). The bank you use is a stack of medieval artefacts, now made with software on top of it. The Medici built the shape of it in Florence. Today’s payment networks are another version of the Knights Templar moving gold for royalty in the 1100s. For those who don’t know, the Knights Templar created the first international banking system to protect wealth. Instead of carrying physical gold through dangerous areas, rulers deposited their assets in Templar fortresses and travelled with coded letters of credit.

An 1850 passbook in plastic is now your “debit card”. Apple Pay is that plastic card copied again inside a phone. And the branch? It hasn’t really changed since Monte dei Paschi opened in Siena 750 years ago and never left.

King borrows a word from engineers for how this happens. Design by analogy, which is when you solve a problem by looking at how something completely unrelated works and stealing their idea.

You build the new thing by copying the old thing and improving it a little. Von Braun’s V-2 became NASA’s Apollo became every rocket since. Carl Benz, in 1885, instead of trying to build a faster horse carriage, took an existing engine and built the first car. Musk did the same with SpaceX and dropped the cost to orbit by more than 90 per cent. Jobs carried a block of wood around the office to remind his team that a phone could start from nothing.

When the web showed up, King says, they took the branch and retyped it onto a screen. Online banking is a branch form, but it is in a browser. Mobile banking is the same form, but smaller. Did we ask if the form needed to exist?

A quarter century after the internet went commercial, there are still places you can’t open an account without going there in person and signing your name with a pen. You prove your identity with a first-century artefact that isn’t unique and is easy to forge.

If you build a bank from scratch today, you don’t build a bank at all. You just want your money to move from point A to point B. So, the whole thing is a tool of utility. When you remove the cards and signatures, you can see how they were blocking what you actually wanted to do without that friction; the “bank” would be swallowed up, and only the function would remain.

Most of the book is about the companies King thought would do the stripping. Neobanks and Fintechs. Also, the Chinese giants like Alibaba and Tencent.

The heroes of Bank 4.0 mostly did the analogy thing, too. Revolut, Monzo, N26, and King’s own Moven built a beautiful app and a faster sign-up. But they are still routing your money through the exact same slow, fifty-year-old banking pipes as Everyone Else.

That means a neobank is an app, not a bank in the deep sense. When you tap a Revolut card or send a transfer, the app hands the job to the old systems everyone uses. Visa and Mastercard clear the card tap. A network called ACH handles bank-to-bank transfers in the US, and it settles in batches.

King’s test requires abandoning assumptions and rebuilding infrastructure from zero, but his chosen heroes fail that benchmark. They only make small changes to old systems instead of building a new foundation.

If you do not touch the rail, you are iterating on the branch. Seventy-six per cent of neobanks still lose money. I have written about this last week. The average one earns about $45 a year per user. A normal bank earns $350. The surviving neobanks did it by reaching for the oldest trick banks have. Lending. Nubank makes its money on credit cards and loans. Revolut turned its first real profit once its loan book grew. To sum it up, the neobank that lived became a bank.

Read: Neobanks Have to Be Banks - by Thejaswini M A

King writes, is “what inevitably emerges when you have to retrofit money, value stores and payments systems to a real-time world sitting on the IP layer, directly accessible by the user rather than through a gatekeeper.” Read that again with stablecoins in mind.

I meant to say take the volatility out of his Bitcoin, and you land on the thing his own logic was pointing at. He couldn’t say it in 2018, so he kept the chapter at arm’s length, probably why he called it “a futurist’s perspective,” and moved on.

The action is in emerging markets, where legacy is poor. Then he tells the M-Pesa story. In 2005, seven in ten Kenyans had no bank account. Today, nearly every adult uses mobile money, and around 40 per cent of the country’s GDP moves across M-Pesa’s rails(at the time when King wrote this, now it’s more). The banks saw it coming and lobbied the finance ministry to investigate. They were already too late. First principles win first in the places the old system never bothered to serve. Stablecoin adoption is loudest where the local currency is shaky, the banks are slow, and a phone is the only branch anyone has, which is the same shape as M-Pesa, one layer down.

In this book, the crypto pages are a time capsule. Back then, ICOs were the “killer app.” John McAfee promised Bitcoin would hit a million dollars. I want to believe him still. A little embarrassing to read now, but it remains an essential read because we look back to “First principles.”

Eight years on, that’s what the phrase has come to mean. Crypto, for all its dead tokens and noise, is one of the few places running the experiment for real, asking what money even is when you start from zero. Some of the answers are junk, but we see that for what it is. King dared banking to picture itself starting today. Almost nothing around us would survive the question. And we’d keep it anyway.

First principles is a beautiful idea. But using it means you look at the thing you’ve served your whole life and admit you might not build it again. So the idea stays admired and unused.

That’s it about Bank 4.0, see you next week with another one.

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.

That’s how financial companies make a good amount of money purely off your procrastination.

Two economists, Richard Thaler and Shlomo Benartzi, figured out years ago that convincing people is a waste of time. Instead of trying too hard on winning arguments; you change people’s behaviour by rigging what happens when a human does precisely nothing. “opt-out” link is apparently a tough one to press on.

When companies make people manually fill out a form to save for retirement, barely half do it. Flip the default so they have to manually opt out, and participation instantly clears 90%. It’s the same story with subscriptions. Over 50% of people paying for auto-renewing apps don’t even use them. The subscription I took last week to watch FIFA WC, which I am completely sure I will forget forever, after the game.

The catch in this system is that a default only works when the person whose money it is and the person who picks the fund are two different people. Your employer picks the 401(k) menu, and you just land in it.

On June 18, Franklin Templeton filed for two ETFs designed to wire that default into bitcoin

In the grand scheme of things, the “capital flow moat” is tiny.

Deciding to buy Bitcoin is one of the big barriers to adoption. We almost overcome it when Trump says hi at Bitcoin conferences, but the barrier still stays after all.

Somebody has to choose to hold it, defend that choice to a client or a compliance desk, and eat the downside when it cracks in half. For an advisor who is career risk, most of them route around it and never put it in front of you.

An advisor builds a model portfolio. They pick the funds, and the client holds whatever got slotted in. The client opens the statement, sees “US large-cap equity, 40%,” and doesn’t bother with the holdings underneath. If the advisor used the DRIP version, that client now owns Bitcoin. The client inherited the choice without noticing one had been made.

But they aren’t trying to trick you, the retail buyer, into being lazy. They know you actively look at your portfolio. Instead, they built this infrastructure for your financial advisor.

That is the trick, and it is the same one that built the $4 trillion target-date pile. The default becomes the product, and the product wins by being the thing you get when you do nothing. The DIY buyer who typed the ticker is the one person this doesn’t describe, and Franklin doesn’t need them. The money these funds want sits downstream of someone else’s click.

A dividend reinvestment plan is the most set-and-forget thing in all of investing. Your stock pays you a dividend, and instead of the cash coming into your account, it buys you more of the same stock. You compound deeper into what you already hold and hardly think about it. That is a DRIP.

Franklin just flipped it. The two funds, the Franklin US Equity Bitcoin DRIP Index ETF and the Franklin US Innovation Bitcoin DRIP Index ETF, take the dividends your stocks pay and use them to buy Bitcoin (not more stock).

For the Bitcoin allocation itself, the funds plan to use spot Bitcoin ETFs, futures, and options. Franklin built in a quarterly rebalancing mechanism with asymmetric rules. If Bitcoin rallies hard and its weighting drifts above the 5% target, the fund trims it back to 4.5% at the next quarterly rebalance. There’s also a hard cap at 20% Bitcoin allocation.

So the Bitcoin portion starts at 5%. Every quarter, dividends flow in and are used to buy more Bitcoin. If the Bitcoin price goes up and Bitcoin becomes a bigger percentage of the fund, they trim it back down to 4.5% at rebalancing. The excess goes back to the equity side.

The maximum Bitcoin can ever be is 20% of the fund. Even if Bitcoin goes completely parabolic between rebalancing dates.

To bypass a bunch of regulatory red tape, the actual coins are tucked away in a Cayman Islands subsidiary that Franklin owns, which buys a mix of spot crypto, futures, or options.

The whole thing tracks specialised indexes built by a firm called VettaFi, and Franklin is aiming to launch them on September 1. We don’t know how expensive they’ll be yet, because they left the fee section blank on the paperwork.

Now I stop selling you the dream. A new buyer, hardcoded into Wall Street’s plumbing, sounds enormous.

Run the real numbers, and the bid turns into a drip.

The broad fund tracks an index yielding 1.05% a year. The innovation fund tracks one yielding 0.52%. Both start at 95% stocks and 5% bitcoin, and only the dividends from that stock sleeve get turned into bitcoin. So the broad fund converts about 1% of its assets into bitcoin a year. The innovation one around half a per cent.

These funds haven’t launched, so use Franklin’s existing bitcoin ETF as the yardstick. It holds about $359 million. One per cent of that is ~$3.6 million of bitcoin buying spread across a whole year. Bitcoin trades tens of billions of dollars a day, around $36 billion. At that rate, the market chews through a full year of this fund’s buying in under a minute.

The innovation fund is the joke inside the joke. Its largest holdings are Nvidia, Apple, and Microsoft, names that pay tiny dividends or none at all. Since the fund relies on those dividends to buy bitcoin, it has almost no fuel to work with. Then the rebalance turns the engine backwards. Bitcoin gets trimmed back to 4.5% every time it grows past 5% at the quarterly reset. So the better Bitcoin does, the more the fund sells. In a real market rally, that constant selling will easily wipe out any gains the tiny trickle of buying managed to create. It is built to keep you from ever actually holding onto a winner.

On a good day for Bitcoin, this fund turns into a forced seller.

How? Index funds have been forced on buyers. Traders make a fortune front-running them because everyone knows exactly what stocks the index has to buy and when. Franklin’s new funds create the opposite problem of a forced, automated seller. Because the fund must buy bitcoin the morning after a dividend and sell it at every quarterly rebalance, any fast trader can step in front of the fund and squeeze it for a profit on both ends.

One fund this size is a mosquito; a whole category of them is weather. Get enough of this capital in one place, and you create a ceiling that presses down on bitcoin every time it rallies.

Apart from the default trick, the filing leaves a few smaller ones open, too.

The mandate dodge - An advisor whose rulebook bans crypto can still drop a client into a fund labelled US large-cap equity.

The offshore wrapper - The bitcoin sits in a wholly owned Cayman Islands subsidiary, the standard rig for holding a commodity-type asset inside a regular fund without breaking its tax status. Legal, and common.

And the tax wrinkle- If a dividend turns into bitcoin before it ever reaches you, you owe the money, but the fund already locked it up in crypto. To pay the tax bill, you are forced to dig into your own pocket and find outside cash to cover a dividend you did not hold.

For all of this to work, A fund like this has to become a default, or sit right next to one. In 2006, the Pension Protection Act gave employers legal cover to enrol workers automatically and have them in a default fund.

Back then, only 5% of 401(k) plans offered a target-date fund. Today, 96% do, and total assets exploded from $100 billion to $4 trillion.

In August 2025, an executive order was issued by Trump that formally opened the way for cryptocurrencies to be used in 401 (k) s. Then, in March 2026, the Department of Labour published an administrative proposal giving fiduciary services liability protection for putting alternative assets such as cryptocurrencies on their plan menus.

The period for submitting comments on this proposal was closed as of June 1, which means that if a final administrative rule is to be issued prior to calendar year-end, it must be done before then. Creating a default alternative investment (target-type funds) is more difficult than creating alternative investments for participants with funds in which they have something invested; thus, regardless of the content of this new administrative rule, corporate attorneys believe that most employers will continue to “wait and see” until a court allows them to proceed under the proposed safe harbour.

None of this runs on convincing anyone. Human attention is the scarcest resource on earth, and whatever automates the friction of thinking wins.

The system just needs your inertia.

—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.

Do you know that a partnership called Cede & Co. in New York is the registered legal owner of around 83% of all publicly traded stocks in America? When you buy Apple shares through Schwab or Robinhood, the legal owner of those shares is still Cede & Co.

Cede is the nominee of the DTC (Depository Trust Company), and their own representation letter filed with the SEC says word-for-word that it “has no knowledge of the actual Beneficial Owners of the Securities.” So basically, Apple does not know you own their stock. Only your brokerage does, because you are their customer. But in the actual ownership chain, you don’t even exist. The word “Cede” itself comes from Latin, and it literally means “giving up power or territory.” It couldn’t be worse.

So I went down looking into how this happened, and what it actually means for you, and whether tokenisation, the thing crypto built to fix all of this, is any different?

The Chain, and the Crisis

So what you actually hold when you buy a stock through a U.S broker is technically called a “Security entitlement,” which is just a polite way of saying you have a contractual claim against your broker.

What essentially happens when you buy a stock is: the DTC updates its records to show that your broker now owns the stock, and your broker updates its records to show that you own the stock. There are three layers of IOUs between you and the actual stock, and what you have is a promise from someone who might have the real thing.

Because of how this chain works, your broker also has the right to lend your shares to short sellers without your consent. Meaning, someone could be actively betting against the stock you bought, using your shares, and you would never know. This also means you don’t actually vote at shareholder meetings; your vote goes through the same chain of middlemen as your ownership. And because these securities are reused as collateral throughout the trading day, multiple institutions end up reporting ownership of the same asset simultaneously.

By some estimates, only about one in three parties actually owns a US Treasury bond. The other two are just holding claims on something that’s already been pledged to someone else. And the wild part is how we even got here.

In the late 1960s, the entire American stock market was still based on physical paper. Stock transfers used to mean moving actual certificates between firms, and a single transfer could require up to 33 different kinds of forms. Each afternoon, hundreds of couriers, many of them retired cops and firemen, used to carry suitcases and steamer trunks stuffed with stock certificates through lower Manhattan, walking them from one brokerage to another. At one point, there was a firm that Merrill Lynch later absorbed, but it had 600 employees whose entire job was handling certificates.

By 1968, the trading volume had hit 20 million shares a day, which was enormous for that time but roughly 1% of what we do now. This led to the back offices being completely broken down, and the NYSE started closing on Wednesdays and cutting hours for the rest of the week just to catch up on all the pending paperwork.

The pressure was so much that it even broke firms like Goodbody & Co., which had been on Wall Street for decades. And with that much chaos, a lot of organised crime emerged. The Attorney General told the Senate in 1971 that more than $400 million in securities was stolen in just three years. A 22-year-old stock clerk at a firm was indicted for stealing $900,000 in IBM certificates.

At one point, the Congress threatened to federalise the entire post-trade operation, so they decided to stop the current way of operating entirely. Instead, they created a centralised vault, locked all the certificates in it, and updated ledger entries when ownership changed, rather than physically moving anything. This was called immobilisation, and the DTC was set up in 1973 as that vault.

But there was a second option, called dematerialisation, which could have removed paper certificates entirely and given every stockholder direct electronic ownership of their shares. They went with immobilisation instead because it was faster to roll out during crises, and it was supposed to be a temporary fix. Then, in 1994, a UCC amendment made it a permanent law in all 50 states, and since then, it’s still running that way.

Phantom Shares, Old and New

This new way of tracking ownership through ledger entries rather than actual certificates introduced new kinds of errors. One where two different parties can claim ownership of the same shares. For example, when a short seller borrows a share and sells it, the buyer holds what looks like a real share in their brokerage account, but the original share still exists in the lender’s records, too.

So both show up as owned and can be lent out again. And if you do this enough times, the number of claims on a stock can exceed even the number of shares that actually exist in the market.

In 2017, when Dole Food was taken private, its shareholders filed claims for 49.1 million shares against only 36.8 million that actually existed. That’s 33% more claims than the number of shares that existed. These are called ‘phantom shares,’ and they’re not due to fraud or manipulation but to the design of the current settlement system, which runs through the same Cede & Co. structure. The DTC’s own ledger could not track all the trades that had passed through it until the company decided to go private.

Something similar happened with GameStop on a much larger scale. In early 2021, the short interest in its stock had exceeded 140% of the float, meaning more shares had been sold short than existed. The squeeze made headlines, and at the peak of the rally, brokers like Robinhood froze the buying while still allowing sells. Reddit’s GameStop community started asking how their brokers even had that kind of power, and that was when they found that their shares were not registered in their names but held by Cede & Co., and were being lent to the same short sellers they were betting against.

So they started pulling their shares out of brokerages entirely and transferring them to Gamestop’s own transfer agent, to put their names directly on companies’ books. By 2023, around 76 million shares had been registered directly, roughly a quarter of the entire company.

Coinbase recently launched tokenised stocks, claiming “true equity ownership” with full shareholder rights and dividend payouts. But if you look at the mechanics, a token backed one-to-one by shares held at a third-party custodian is still a claim against the custodian. The only thing that changed is the database it runs on, from a ledger at DTC to a ledger on the blockchain, but the number of intermediaries between you and the actual shares is still the same. If anything, you added one.

In the traditional system, the issuer’s register lists Cede & Co. as the legal shareholder. And Cede holds on behalf of the DTC. The DTC records which broker owns how many shares. And the broker records your name against a portion of that. There are four layers between you and the company whose stock you bought, and the company has no idea you exist.

In the tokenised wrapper model, as built by Coinbase and Robinhood, nothing changes either. A custodian still holds shares through the DTC, registered in the name of Cede & Co. The token platform issues claims against the custodian’s holdings. And you hold a token that represents a claim on the platform.

Last year, Robinhood began offering tokenised exposure to OpenAI in Europe through this model. The tokens did not represent direct ownership of the company’s shares. They represented ownership in an SPV that holds the shares on your behalf. So you own a piece of the vehicle, and OpenAI would have no idea you exist. Within hours of its launch, OpenAI publicly announced that it had nothing to do with these tokens and had never approved any transfer of its equity.

Anthropic went even further in May 2026, and declared that any sale of its stick not approved by its board is void. PreStocks had been running a tokenised Anthropic market for months, and when the announcement came, the price of its tokens dropped 27% within hours.

It goes to show that exposure to a stock’s price movement and ownership of the stock are two completely different things. One gives you price upside, but the other gives you rights, a vote, and a legal claim that a court would recognise.

The most extreme case was SpaceX, where all crypto exchanges promised and sold tokenised access to its anticipated IPO, collecting over a billion dollars in orders. The hype was off the charts because retail investors who had never had a way to buy into SpaceX before could do so now, and it felt like crypto was finally delivering on its original promise of access. But then the main provider, xStocks, could not deliver the underlying shares, and every exchange had to cancel and refund the orders.

There was literally nothing to tokenise because the entire product is built on the claim of an asset that no one in the chain could source.

There are also exceptions to this that show how it can work, as well as a third model to do it. Superstate is an SEC-registered transfer agent that records legal ownership directly on Solana. The tokens represent direct shares with no custodian between you and the company. This is what dematerialisation was supposed to be fifty years ago, and the only time when “ownership” actually means what people think it means. Prathik will cover this side in more depth on Friday, so watch out!

Kraken also ran its tokenised stocks through its own broker-dealer, and it held where everyone else’s collapsed. In fact, Singapore’s Central Depository already gives every retail investor direct legal right to their shares, with direct voting and no nominees in between.

The regulatory path is already available. In May 2025, the SEC confirmed that a registered transfer agent can use a blockchain as its official shareholder register, with no off-chain duplication required. Superstate already does this on Solana, where you hold the token, and your name is on the transfer agent’s master file as the registered owner. Securitize runs on the same model and powers BlackRock’s BUIDL fund, which holds over $4 billion in tokenised assets, and the NYSE chose them in March 2026 to build its tokenised securities platform.

Countries like Switzerland, Germany, and Liechtenstein have also passed laws recognising on-chain records as the legal instrument of ownership. But the problem is that the intermediary chain itself is a $20 billion-a-year business in the US alone. Broadridge pulls in roughly $3.4 billion annually from processing proxy materials and investor communications alone. When the DTC itself ran a tokenisation pilot in late 2025, it still chose a model in which Cede & Co. remains the registered owner, preserving every intermediary in the chain.

To be fair, there is also a case where being a wrapper could be valuable. For someone in Lagos or Jakarta who has never had a way to buy Apple or Nvidia stock, a tokenised claim, even through an SPV, is still access they did not have before. But when OpenAI can reject your tokens within hours and Anthropic can void them with a single board statement, that access will only be as strong as its weakest link. And as we saw with SpaceX, when nobody in the chain can even source the underlying shares, the access doesn’t even matter. The point is that you should not have to choose between global reach and actual ownership.

And that’s the whole story, most of what the industry calls tokenised ownership today is the same IOU from 1973 on a new database. The tools exist, and a few teams are actually using them, but the rest of the industry chose to become the new middleman because that is where the real money is for them.

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.

And why wouldn’t it be?

Cigna built an algorithm that lets doctors reject claims without opening the file. UnitedHealth cut off nursing care payments when an algorithm’s clock ran out, regardless of what the treating physician said. The business model has always been to collect your capital, pocket a cut, build walls between you and a payout.

Now your bank deposit is covered by FDIC, but only up to $250,000, a number set in 1934 and barely moved since. Your brokerage account has SIPC, capped at $500,000, useless the moment your account is worth more than that. The protection is smaller than most people assume, and the insurer decided on the ceiling.

DeFi insurance almost fixed it. It stepped up to remove the corporate middlemen. If the code’s conditions are met, the payout triggers automatically. It is insurance stripped of human malice.

But nobody buys it because the insurance premium eats so deeply into your yield that the yield no longer justifies the risk you’re taking to earn it in the first place.

Today I want to show you exactly how that plays out, and why the industry couldn’t fix it even if everyone wanted it to.

Nexus Mutual, the largest DeFi insurance provider that exists, has paid out just over $18 million in claims across its entire history since 2019.

In April 2026, Kelp DAO got exploited for $292 million.

This single hack is 16x the total claims an insurer has settled in seven years.

The contrast with the wall of denials above is antithetical. Traditional insurers collect big premiums and fight hard to avoid paying out. DeFi insurers barely collect any premiums at all, because very few of the crowd is buying the product.

Traditional Insurance works well because risks are uncorrelated. The burning of one’s house has no relation to the burning of another’s house. Because of this, insurers can sell policies to a million people and pay out claims from one house burning. They can maintain usefulness because of the uncorrelated risks. DeFi doesn’t have that property. An oracle failure or a bridge exploit, cascade through every vault and lending market built on top of the asset that broke. When USDC depegged in March of 2023, it impacted every protocol that had USDC as collateral, all on the same day. For a DeFi insurer, it is priced as a correlated bet that the failure is small enough that the pool can sustain the failure.

When Euler Finance got exploited for $197 million in March 2023, the damage didn’t stop at Euler. Angle Protocol lost $17 million because it held Euler LP tokens. Yield Protocol paused its protocol. Inverse Finance and several others had direct exposure.

When something breaks, it tends to break a lot of things simultaneously, which means a single bad day can wipe out not one protocol’s claims but the entire pool’s capacity at once.

I pulled current premium rates from Nexus Mutual and InsurAce against the live yields on the protocols they’re meant to protect. Aave V3 pays about 3.14% APY on USDC. Cover runs 1.5 to 2.5%. Net yield after insurance: 0.6 to 1.6%. You’re working for a return that barely clears a savings account, in exchange for protection you think you wouldn’t need.

If you look at the image, Morpho, Compound, and Spark land in a similar place. Yields around 3.5 to 4%, premiums eat roughly a third to half of it, which is annoying, but survivable.

But when you look at Maple Finance, it pays 4.77 to 4.90% on its institutional lending vault. Premium is 3 to 6%. Net yield after cover ranges from negative 1.1% to positive 1.9%. Ethena’s stake yields 3.6 to 4%. Insurance costs 3 to 6%. Net yield is negative 2.4% to positive 1%. Buy insurance on these, and you can end up paying to lose money.

Sky, formerly MakerDAO, wins here. Its savings product yields 3.6%, and covers costs as little as 0.11%, because the market has priced Sky as close to the safest thing in DeFi. Net yield after insurance: 2.8 to 3.5%. Almost the entire return survives.

Read: Sky’s New Sovereign

Insurance is priced right for the risk, but on new platforms, the cover costs so much it completely eats the high yield you went there for anyway.

When crypto investors choose to skip insurance, you don’t call them lazy or reckless. They know that buying insurance means making zero money, most times. Even if every depositor in DeFi decided tomorrow that they wanted full coverage, the industry could not deliver it. Nexus Mutual’s total capital pool sits around $81.56 million. Total active cover across the entire sector is a few hundred million dollars at most, against hundreds of billions in TVL across the protocols that need protecting.

A hack the size of Kelp hitting a fully insured protocol could wipe out a meaningful share of the industry’s entire capital in one claim.

Eighteen million dollars in claims is a demonstration that the capital pool has not yet faced the one failure that would show how vulnerable the capital pool really is.

When users submit a claim to Nexus Mutual, it gets voted on by all the other Nexus members. The people who voted in favour of a claim stand to lose financially if the claim is not successfully paid out. This inherently creates a bias in favour of not paying the member claim. Traditional insurers have underwriters and claims adjusters for this reason. DeFi insurance combines these elements back into one system, by design.

Before 2008, the people pricing financial risk assumed that because a nationwide housing collapse hadn’t happened in their lifetimes, it was impossible. Insurance giant, AIG, sold promises of protection on a massive scale. Promises it couldn’t keep when the market finally broke.

Before the US government created bank insurance (the FDIC), regular people had no safety nets whatsoever. Then the Great Depression forced the government to step in and make insurance mandatory. The government made insurance a non-negotiable cost of operating a bank.

When it comes to DeFi, nobody can force Aave or Morpho to buy insurance, because nobody can force a smart contract deployer to do anything. All of this is permissionless, and that prevents anyone from mandating the protection that might keep it solvent through the next shock.

The three biggest claims on record for Nexus Mutual are the FTX disaster for around $7.3 million in two claim rounds, the TribeDAO hack for $5 million, and the Euler Finance exploit for $3.4 million. If you add those three up, that comes close to being the total amount of the $18.6 million paid out by the mutual to date in seven years of operations.

Mutual is now moving into prevention mode.

It does this through Bug Bounty Cover in tandem with Immunefi and other security teams, including Cantina and Sherlock, with protocol teams paying just 20% of a critical bug bounty while Nexus pays the rest, thereby allocating capital towards exploiting an exploit before it takes place. In addition, Nexus Mutual is exploring regulated insurance cells in a bid to channel crypto risks into reinsurance pools which can afford a lot more capital.

Cantina went further in March 2025 and launched a separate product, Native Protocol Cover, which pays out after an actual exploit, even if no bounty hunter caught the bug first.

Both of these moves directly admit that on-chain capital isn’t enough to insure on-chain risk. The pool is too small, the risk is too correlated, and the people judging the claims are the people funding them.

Nexus Mutual’s $81.56 million in DeFiLlama-tracked TVL is 85% of the entire DeFi insurance category combined.

Other projects have not been able to expand in the same way. A few years back, InsurAce peaked at $150 million, but now it is only at $132,000, having paid only one notable claim after the UST de-pegging in 2022. In only one year, Sherlock’s pool dwindled from $60 Million to $505,000. Unslashed Finance has a couple of million dollars stuck in obsolete code, which has not been updated since late 2024. All of the others are either abandoned or have pivoted focus.

A lighthouse, paid for or not, warns all ships of the rocks. A passing ship cannot be billed for the light used to avoid a wreck. Thus, it is assumed that lighthouses are never built voluntarily, as the benefits are to others who never paid for them.

DeFi Insurance cover forestalls the aftermath of a cascading event of crumbling positions. Everything in crypto is connected to everything else. Because of this, the value of Insurance is that your neighbour bought it too. Everything stays balanced.

Ghosts guard nothing.

—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.

That’s how this dismantles any romanticised vision of mafia honour, like The Godfather, and shows the world the ugly, cold, and parasitic nature of organised crime.

Let me do you the honour of applying the same (or similar) logic to big tech.

Capture the margin, and you capture the value. To accomplish this, you need neither a protocol nor a project. It’s a war for the margins, and there are no rules.

But we can’t blame Coinbase, or Stripe, or Kraken for choosing that.

What they did, in the most fundamental, basic sense, was make a good real estate move. They got to distribution first. Now, they can look down and say, “Who has the leverage here?”

Coinbase built a blockchain, and Stripe bought infrastructure for $1.1 billion that it could have rented. Kraken built a derivatives trading platform for $1.5 billion. Apple built an App Store. You let someone else build the market, take the early risk, and you swallow the infrastructure once it becomes too profitable to avoid.

This is a piece about what happens when distribution no longer has value.

Coinbase has 110 million verified users. For years, the lending product it offered those users ran through Morpho, an open protocol, which kept the protocol fees.

Then, Coinbase built its Layer 2 blockchain, Base. Morpho deployed on Base because Coinbase’s users meant volume. Every transaction processed on Base now generates a sequencer fee that goes to Coinbase, instead of Morpho.

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

Base generated $76 million in net sequencer revenue in 2024 and $74 million in 2025. Up until February 2026, they had to hand a chunk of that over to Optimism under a licensing deal. But Coinbase eventually cut the cord, moved to their own stack, and now they pocket the full $64 million. Meanwhile, Morpho is still right there on Base, doing great and managing $2.5 billion in TVL. But a cut of every single transaction Morpho processes goes directly into Coinbase’s pockets.

Coinbase built its $300 million bitcoin-backed loan product directly on Morpho infrastructure. cbBTC, Coinbase’s own wrapped bitcoin product, is the single largest collateral type on Morpho, representing 38% of all assets locked in the protocol. This makes Coinbase simultaneously Morpho’s biggest customer and the company that takes a toll on everything Morpho does. Morpho has leverage over Coinbase’s product. Coinbase has leverage over Morpho’s revenue. Neither can easily walk away.

In Stripe’s case, in early 2025, it paid $1.1 billion for Bridge. Before that, Stripe routed its stablecoin operations through Circle’s infrastructure. But with that, Circle controls the stablecoin issuance and captures the reserve float on the backing assets. When Stripe moved through Circle, it was a billion dollars generating yield for Circle. Bridge redirects that. It has USDB, its own stablecoin backed by BlackRock money market funds. By switching to USDB, Stripe keeps that massive reserve yield entirely inside its own ecosystem. When you process $1.4 trillion in annual payment volume as Stripe does, renting the yield engine from a competitor is worth hundreds of millions of dollars a year.

Patrick Collison called stablecoins “room-temperature superconductors for financial services.” Spending $1.1 billion was the price of owning that superconductor outright, rather than paying someone else’s toll to use it.

As a basic spot exchange, Kraken has a natural ceiling where users can only trade a few hundred tokens. However, the institutions and serious retail traders Kraken wants to attract run their books on futures and cleared derivatives. Operating that kind of business requires CFTC registration, NFA membership, broker-dealer credentials, and compliance frameworks that take years to establish. Even if you try to build everything from scratch, the government can still deny your application for reasons entirely outside your control.

This is why Kraken looked at NinjaTrader. The $1.5 billion acquisition in January 2025 bought them 1.7 million funded accounts, but more importantly, it secured the broker-dealer licenses Kraken couldn’t build fast enough organically.

By purchasing those readymade permission slips, Kraken eliminated its dependency on outside partners. They no longer have to rely on anyone else or wait years for government approval because they now own all the technology and licenses themselves.

So big company eats small protocol. Big deal, what’s new?

Morpho’s total value locked stands at $6.4 billion. Of this, $3.308 billion is on Ethereum, and $2.488 billion is on Base. If Coinbase decided to remove Morpho from Base in favour of their own lending protocol, Morpho would lose 39% of its TVL, but would maintain 52% on Ethereum and, along with its growing positions on Hyperliquid L1, Monad, Arbitrum, and many other chains, would continue to thrive.

Aerodrome on Base shows the effects when a chain operator sponsors one of their own competitors. Aerodrome launched Base’s native DEX and was optimised for Base’s architecture. Coinbase Ventures has approximately $20 million in AERO, its largest liquid token investment, and votes with locked AERO to direct liquidity to Coinbase’s products, including the cbBTC liquidity pools. Aerodrome processed approximately 51% of the Total DEX volume on Base, peaking at 77% in September, 2024. Uniswap, which runs on 44 chains, is the second DEX on Base, processing 30% of Base DEX volume. Uniswap has not ‘died’, as it lost dominance on one chain, while processing Base’s $212 billion in 2025 volume and estimating $73 billion monthly across all chains.

In this case, being multi chain is actually protection. A protocol that focuses on a single chain is at the mercy of the single chain operator if that chain’s operator then favours a competing protocol. A protocol that engages with multiple chains loses only one market segment, but retains the rest. Morpho watched Aerodrome capture Base’s DEX volume from Uniswap and quickly deployed across multiple chains. The massive distribution companies can go downward. The systems can extend horizontally.

If you depend on an infrastructure layer you don’t own, you don’t own your business. The entity controlling that layer has massive leverage over your profit margins. They dictate your product experience. Ultimately, they control your operational stability. At the scale these companies operate, that dependency costs real money every single day. This is not an insight unique to crypto. Amazon did this with AWS, Apple saw how Intel’s chip roadmap dictated the direction of every Mac and spent years developing its own custom silicon.

Anyone can track exactly how much Coinbase earns from Base sequencer fees or view Morpho’s total value locked across different chains. This value extraction is visible in real-time, in a way that Amazon’s internal infrastructure margins never are.

If the future is a few massive companies like Coinbase, Stripe, Kraken, and a couple of banks own everything.

They have all layers from protocol to card, with open source protocols only filling the gaps that the companies have not occupied. This is a legitimate potential FinTech scenario. Open-source tech won’t be this grand, free frontier anymore. It’s just going to be the duct tape used to fill the tiny cracks that the mega corporations haven’t figured out how to monetise yet. ‘Oh, look at that lovely little open-source protocol... let’s build a Starbucks over it.’

My optimistic side view is that the outcome is less likely than it looks from the current deals. The protocol layer is hard to own, the way distribution is owned. Morpho can deploy on a new chain in weeks. The switching costs for replacing an embedded, battle-tested lending protocol are high, even when they’re invisible from the outside. Coinbase’s $300 million bitcoin loan product runs on Morpho because rebuilding Morpho’s security model from scratch would take years and introduce risk Coinbase doesn’t want to carry.

The protocols poised to survive this wave are those that achieved multi-chain ubiquity before the distribution giants started building their own networks, embedding themselves so deeply into corporate backends that replacing them becomes economically prohibitive. Even a distribution powerhouse like Robinhood chose to integrate Lighter for its trading infrastructure. Lighter is a third-party zk-rollup perp exchange, and started routing users there. Robinhood Ventures backed Lighter’s $68 million raise. Vlad Tenev sits close enough to the project to call it personally.

If distribution alone were the only moat that mattered, Robinhood would have built in-house the way Coinbase did. It didn’t, because verifiable zk-proven matching at CEX speed is a hard, narrow problem that took Lighter’s team over a year to solve, and Robinhood decided buying access to that expertise was cheaper than rebuilding it.

Morpho currently occupies this dual strongpoint, and Uniswap was its pioneer. Velocity of institutional encroachment versus the speed of protocol proliferation. The result of that can tell us what is what.

The layer below each of them big names (Stripe and Coinbase) is still open source. For now, that’s enough. Ask again in two years.

I will come back in two days, until then…

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.