In a world of ledgers, who holds the memory? The past seven days have been telling for those watching the tokenized treasury market. We have surpassed $160 billion in tokenized U.S. Treasury funds, yet the question that keeps me up at night is not how large this market can grow, but how much of it is actually doing anything.
Aave Horizon has crossed $250 million in total value locked. Figure PRIME has grown by over $200 million this year alone. These numbers signal a phase shift. We are moving from the era of issuance to the era of utility. The tokenization narrative is no longer about creating digital representations of bonds for the sake of digital representation. It is about putting those assets to work in the most demanding environment on earth: DeFi lending. This transition is not a technical upgrade. It is a philosophical test of whether our industry can build infrastructure that respects both the speed of code and the gravity of traditional finance.
For years, the tokenization thesis was simple: bring real-world assets on-chain, and they will create liquidity and efficiency. The tokenized funds were built for distribution, with a focus on access and transfer. This was the onboarding phase, a necessary first step. But as I have argued in my essays, distribution is a beginning, not an end. An asset that can be held but not actively leveraged is a passive instrument, a curated artifact. The next phase of tokenization is utility, and the best test of utility is whether these assets can function as collateral in DeFi lending markets.
This shift brings the industry to the core of its promise. We are not moving money; we are moving belief. The belief that traditional assets can be used with the same efficiency as native crypto collateral.
The emerging infrastructure layer is being built by players who understand that utility comes with friction. Midas has launched mWIN, a tokenized fund that is natively issued on-chain, rather than wrapping an existing fund. It invests in investment-grade CLOs and other asset-backed credit and currently yields around 6.9%. The fund is managed by Wellington Management and held by Northern Trust, and it uses T+1 issuance and redemption cycles. This is a deliberate design choice, optimizing for the use case of collateralization, not for the ease of traditional settlement.
We are also seeing the entrance of major players. Aave has launched Horizon, a protocol specifically designed to allow institutions to borrow stablecoins against their tokenized assets. Morpho has seen a growing number of markets curated around tokenized credit. Sentora, acting as a market planner, has been setting parameters on Morpho based on historical NAV, stress events, liquidity, and redemption mechanisms. These are not small initiatives. These are major DeFi protocols moving to integrate a new asset class into their core machinery.
The core of this design is the utility thesis. A holder of a tokenized fund can deposit that fund into a lending market and borrow a stablecoin, like PYUSD, without needing to sell the underlying asset. The user retains their credit exposure and their yield while gaining access to liquidity. This is a dual-yield structure: the yield from the underlying asset plus the utility of borrowing a stablecoin to deploy elsewhere. The economic appeal is undeniable.
My audits have shown me that this is where the system is designed to shine, but also where its fundamental tensions lie. The deepest challenge is the liquidation time mismatch. DeFi protocols liquidate in minutes; traditional credit markets settle in days. Tokenization does not erase this gap. For crypto-native assets, like ETH, there is a 24/7 market and liquidations are immediate. For a tokenized credit fund, the underlying asset trades only during traditional market hours, the NAV is calculated periodically, and the redemption is not instant. This is the structural bottleneck that the entire sector is trying to design around.
If a borrower defaults, the protocol cannot simply dump the collateral. The protocol must have a special liquidation path that considers the low liquidity of the underlying asset and the settlement delay. This is where the risk profile changes. The protocol is not just a simple code execution; it is a manager of a complex workflow. The custody and pricing are dependent on a multi-layered trust chain, not the pure, trustless collateral of ETH. We are increasing the number of trust assumptions, and I have always believed that we must be able to audit these assumptions.
This is why the design is moving towards native issuance. An asset built for distribution is structured differently from an asset built for collateral. The former has a focus on legal ownership and a clear transfer mechanism. The latter has a focus on frequent, reliable, oracle-readable pricing, fast redemption, and executable liquidation. mWIN is a prime example of a native on-chain issuance. It has multiple competitive liquidity sources rather than relying on secondary market depth. This is a sophisticated attempt to solve the liquidity problem at the source.
We must also consider the shift in market metrics. I believe we should stop measuring the value of tokenized assets by the amount issued and start measuring it by the amount of active use. How many tokenized assets are securing loans? How much stablecoin liquidity is being borrowed against them? This shift from issuance to usage is the true signal of maturity. When a token is just sitting in a wallet, it is an expensive line item in a treasury. When it is deployed as collateral, it is a productive asset, a belief in the system.
But as we celebrate this growth, I must take a contrarian angle. The current infrastructure is building on a fragile foundation. The oracle risk is often overlooked. The pricing of the NAV depends on a central source, a source that can be a single point of failure. If the oracle is compromised, the liquidation mechanics fail. If the fund manager freezes, the collateral is frozen. In a world of ledgers, who holds the memory? The process is binary, but the meaning is fluid.
There is also a governance risk. The protocol is neutral, but the user is human. The governance of these assets is a dual-track model. On the one hand, you have the decentralized protocol that sets parameters on-chain. On the other, you have the centralized asset management that decides what to buy and when to redeem. This dual governance is a critical flaw. It creates a conflict of interest. The asset manager is not bound by the protocol, and the protocol is not bound by the asset manager. This is not a technical issue; it is a governance issue.
We are entering a phase where we are trying to create a new system of trust. In this system, we need to have the courage to audit our own biases. We code the trust, but we must audit the soul. The protocols are not neutral; they are a reflection of the people who built them. If we are not careful, we will build a system that is just as centralized and just as fragile as the system we are trying to replace. The proof is binary, but the meaning is fluid.
The path forward is not just to make better infrastructure, but to build better governance. We need to ask the hard questions about who is in control, who is accountable, and who is the final custodian of the trust. In a world of ledgers, who holds the memory? The answer must be a resilient system, not a single point of failure. We are not moving money; we are moving belief. And belief must be built on a foundation that can withstand the test of time, not just the test of the market.
