Author: Castle Labs Research Team

Translation: Deep Tide TechFlow

Deep Tide Introduction: The cryptocurrency industry has just experienced the heaviest security blow in history, with Binance suffering a staggering theft of over $6.3 billion. However, the market did not spiral into panic; instead, it exhibited increasingly strong resilience and capital absorption. Under the shadow of this macro black swan event, various native protocols that focus on the real economy and institutional finance are seeing a significant breakthrough in scale.

This research report focuses on four major cutting-edge trends: the first is the on-chain reinsurance protocol Re, based on the trillion-level market of real-world assets (RWA), which saw its total deposits surge from $300 million to over $400 million in just 40 days, outperforming most traditional assets with a low correlation real yield of 7.14% to 12.56%; the second is the lending protocol USDai targeted at the mid-sized market for distributed AI computing power, which aims to become the universal clearing dollar for numerous computing cloud vendors through a model of "on-chain warehouse bridging and bundled resale to institutions."

Additionally, the report delves into the underlying adapters connecting trillion-dollar banking clearing networks with blockchain, Quant ($QNT), the new predictive market Trueo ($TRUE) that has settled in DeFiLlama's traffic entry, and the ingenious financial engineering design that tokenizes US stocks into day (SESSION) and overnight (OVERNIGHT) exposure tokens (IETs), drawing inspiration from Pendle’s separation of principal and interest.

Last week, the entire cryptocurrency industry faced one of the heaviest security blows ever, with hackers stealing over $6.3 billion in assets from the Binance platform.

However, this rare major black swan event did not automatically plunge the overall market into a cliff-like decline, clearly reflecting the current crypto capital's increasingly mature resilience against systemic extreme events.

While closely reviewing the complete picture of the Binance theft incident, this report will also focus on the following core issues:

The reinsurance protocol Re introduces RWA real yields with low correlation to the crypto market through on-chain reinsurance business, with its reUSD and reUSDe total deposit scale now officially surpassing $400 million.

An exclusive interview with David Choi, co-founder and CEO of USDai, deeply analyzes why this protocol aims to become the universal clearing dollar for the global AI computing power industry.

Market momentum asset review: Focus on the underlying architecture service provider $QNT, which connects multinational banks and blockchain networks, and the emerging predictive market power $TRUE, which has just accessed the DeFiLlama front-end distribution traffic.

Reinsurance Protocol Re Deposits Surpass $400 Million

Figure: On-chain reinsurance protocol Re total deposits surpass $400 million

Re is an innovative on-chain reinsurance protocol that allows liquidity providers to deposit stablecoin assets, and these funds are directly used to provide capital backing for regulated traditional reinsurance policies. The global traditional reinsurance market is a massive mature financial branch with an annual scale exceeding $760 billion.

Figure: The clearing and underwriting architecture of on-chain capital connecting regulated reinsurance policies

The substantial stock scale of this industry, the urgent demand from institutions for on-chain unrelated insurance yields, and the extreme refinement of the protocol’s DeFi composability collectively propelled the exponential surge in Re's scale.

This week, the total deposit scale of Re's two earning stablecoins, reUSD and reUSDe, has officially surpassed $400 million, with annualized net yields currently reaching 7.14% and 12.56% respectively. This level of return significantly outperforms many traditional strategies such as short-term US Treasury bonds, institutional unsecured lending, and basis arbitrage, continuously attracting substantial new funds into the market.

Figure: Comparison of current annualized yields of reUSD and reUSDe and their historical stable performance

A more critical signal is that the growth momentum of stablecoin capital's accumulation is accelerating sharply. Previously, Re's deposit scale grew from $200 million to $300 million over 186 days, while crossing from $300 million to $400 million took only 40 days, fully showcasing the market's strong appetite for allocating on-chain real reinsurance yields.

This absorbed liquidity has been fully invested in real production operations:

Since the protocol went live, the cumulative effective reinsurance premiums signed have reached $510.5 million.

Just in October, the protocol injected $65 million into Cover Re to fully support its business expansion in 2027.

Since its establishment, it has allocated over $13 million in cash returns to holders of reUSD and reUSDe.

DeFi composability has formed the best distribution channel for driving Re's rapid expansion, and currently, reUSD has been widely deployed across as many as 14 different public chain network ecosystems.

Recent ecological integration has further broadened its liquidity leverage application: from leveraged circular lending of reUSD on Solana's Jupiter Lend to PT-reUSD circular strategies on Monad. Additionally, PT-reUSD is one of the first three lending markets established on Pendle's high-yield USDC vault on the Morpho platform, significantly enhancing its liquidity potential as quality collateral.

This deep interlocking with DeFi forms a powerful growth flywheel: reinsurance is a business that can bring strong, certain cash flows to fund providers, with performance yields surpassing most on-chain assets; when this underlying yield-generating asset can be further embedded into complex leverage and lending LEGO blocks, it will greatly unleash asset potential and stimulate doubled allocation demand.

We firmly believe that reinsurance is a strategically valuable on-chain blue ocean track, as it introduces a pure native external yield source that is almost zero correlated with the crypto secondary market, whose cash flow depends on policy incidence rather than price fluctuations, thus powerfully helping on-chain stablecoin capital continuously earn healthy returns and remain in the Web3 environment during extreme market volatility.

AI Computing Power Settlement Dollar: Exclusive Interview with USDai Co-founder and CEO David Choi

USDai is a decentralized lending protocol focused on providing GPU mortgages for AI computing infrastructure companies, with all debts denominated in its native stablecoin $USDai. The protocol recently executed a buyback of governance token $CHIP worth $1 million using real business income and officially established a programmatic buyback mechanism.

To deeply explore the team's strategic ambitions and how they foresee the evolution of the industry in the coming years, we conducted an in-depth interview with David Choi, co-founder and CEO of USDai.

As a lending protocol, USDai's business flywheel depends on the true strength of underlying borrowing demand, and this demand for computing power lending is experiencing profound structural changes. Currently, the largest borrowing demand is concentrated in what is called the "Compute Middle Market." As David analyzed: “The power metric has become an absolute rigid constraint, so the scale of construction of data centers is trending smaller: the previously common 250 MW giant campuses are now being dispersed into modular small clusters of 2 to 5 MW. This means a large number of computing operators emerge needing medium-sized loans of $5 million to $100 million, rather than just a few super giants needing billions.”

Despite the extremely strong borrowing demand in the middle market, USDai's core moat differentiating it from traditional Wall Street lending institutions lies in its ability to economically handle small to medium scale orders.

This is the unique dimensionality reduction capability released by on-chain crypto capital: USDai can “quickly complete a standard loan's rapid disbursement in a very short time utilizing smart contracts, hold this asset on-chain to build a good credit repayment record, and then bundle and resell it to traditional financial institutions that were initially reluctant to spend costs on due diligence from scratch.”

USDai has named this unique business model “Season-and-Sell.” The reason for adopting this model is that on-chain crypto capital “views immediate liquidity as the primary demand, while Wall Street's specialized credit funds are relatively blunt about the cost duration of funds.” The team has successfully completed its first model trade: successfully reselling a $106 million GPU financing asset package to leading private credit institutions Silver Point and Victory Park.

Through this architecture, USDai fundamentally avoids heavy asset competition with traditional financial giants, and cleverly plays the role of “high-quality credit asset deep processing manufacturer.”

Regarding the recently announced protocol value alignment and programmatic buyback, David explained the deeper considerations for choosing to start the buyback at this moment: “We chose to start now because the entire business closed loop has been thoroughly run through. We successfully initiated loans on-chain, completed asset incubation, and smoothly resold claims to traditional buyers. This milestone marks a complete break from the model of merely consuming its own asset-liability balance sheet, transforming imperiously into a light asset generation engine relying on high turnover to earn fees.”

Additionally, as the scale of the lending network continues to grow, the source structure of buyback funds will also undergo a qualitative leap. Currently, buybacks mainly rely on the net interest margin income from the collateral certificate sUSDai; and with the upcoming explosion of the total outstanding USDai circulation across the network, CHIP token holders will be able to directly initiate the underlying extraction network for USDai stablecoin itself as a settlement currency, which is the North Star vision of our entire protocol.

At the close of the interview, when asked about USDai's ultimate form, the team firmly stated, “To become the universal settlement dollar for the global AI computing power industry.”

If USDai can provide financial liquidity for 10,000 emerging computing cloud vendors (Neoclouds) in the future, “what they hold on their books will all be the same dollar token, and their computing power scheduling and leasing will also be fully settled in this dollar.”

Aftermath of Binance's Historic Theft Case: Hacker Fund Flow and Market Resilience

On October 29, the world's largest cryptocurrency exchange Binance fell victim to the most significant attack in cryptocurrency history, with total losses amounting to about $6.36 billion, surpassing the total losses from the previous seven worst hacking thefts in history.

Figure: Direct comparison of Binance theft amounts with the top seven hacking cases in cryptocurrency history

However, during this crisis, a positive micro development was that stablecoin issuers Tether and Circle reacted extremely quickly after the event, decisively freezing a significant proportion of the stolen funds (about $3 billion), while Binance quickly covered the remaining exposure with its strong own reserves, restoring normal withdrawal services in a very short time.

Another intriguing detail of this incident is that core members of the North Korean government unusually came forward to publicly deny any connection to the robbery targeting Binance. In the past, North Korean officials typically remained silent or slow to respond to hacking accusations, but this time, a high-ranking official made a swift denial, which is truly unprecedented.

This, of course, cannot legally absolve them of suspicion, but their eagerness to cut ties, to some extent, reflects the subtle geopolitical avoidance, especially considering that relevant U.S. law enforcement and national security agencies have clearly stated they will thoroughly investigate the attack.

Momentum Asset Deep Dive: The Institutional Breakthrough Paths of Quant and Trueo

First, let’s examine $QNT.

Quant Network (QNT) has recently garnered significant attention within the institutional finance circle, hailed as the “universal protocol adapter” between major commercial banks and underlying blockchains. Any traditional commercial bank only needs to complete a single system connection to freely access data crossing different heterogeneous blockchains and traditional bank internal clearing ledgers, entirely eliminating the substantial technical burden of building underlying gateways for each new public chain.

The real turning point of this project began when the Clearing House Group, backed by dozens of major banks across the US, publicly confirmed in September that it is conducting tokenized deposit settlement tests using Quant Network’s underlying infrastructure.

From a business positioning standpoint, QNT does not need to compete with other mainstream public chain tokens for ecosystem liquidity; its role is to empower traditional licensed banks to smoothly embrace blockchain as an infrastructure provider. Therefore, the more surging the global financial asset tokenization wave becomes, the greater certainty there is regarding Quant Network’s institutional adoption rate.

Figure: Quant Overledger architecture connecting traditional bank clearing and distributed public chains

The existing RTP and CHIPS traditional clearing channels operated by the Clearing House Group have daily fund transfers exceeding $2 trillion. Even if a very tiny percentage of that could circulate through tokenized deposits coordinated by Quant, the corresponding fund scale would be enough to trigger infinite market speculation.

However, from a secondary market valuation perspective, its circulating market value has reached approximately $2.8 billion, and the core focus will be on how the massive enterprise-level software adoption will substantively feed back to the QNT token itself.

Another momentum asset worth noting is $TRUE.

This project captures the fastest-growing predictive market track in the current network (Prediction Markets). Currently, a majority of this track's trading volume is firmly held by the two oligopolies, Polymarket and Kalshi.

Figure: Explosive growth trend of total predictive market trading scale in recent months

After announcing a deep strategic partnership with the largest data platform in the network DeFiLlama (DL), Trueo seamlessly integrated its prediction contract module into DeFiLlama's main site front-end interface, with all prediction topics directly selected and presented by DL based on its vast on-chain data, granting Trueo a full endorsement from a top vertical traffic entry.

Figure: Trueo's predictive market component seamlessly embedded in the DeFiLlama homepage front-end product display

Both tokens recorded impressive gains in the recent secondary market battle, accurately capturing fund momentum in the segmented sector.

Quant has completed the sedimentation of its underlying infrastructure and began providing a technical base for large institutions; whereas Trueo attempts to tear open a gap in the entrenched dual oligopoly pattern of Polymarket and Kalshi; its path is undeniably fraught with peril, but its strategy of leveraging industry-leading traffic nodes for a high-speed cold start displays an extraordinary level of operational skill.

Frontline Radar: The Overnight and Daytime Exposure Split Mechanism for Tokenized US Stocks

Recently, Beans has innovatively launched “Isolated Exposure Tokens (IETs).” This mechanism deeply draws on the financial engineering concept of leading interest rate swaps Pendle, which separates principal tokens (PT) from yield tokens (YT). Through a pledge and redemption mechanism, tokenized US stocks are precisely divided into two independent non-interfering time exposure tokens: the “Overnight Token” (OVERNIGHT) representing the market conditions from the close of the US stock market to the next open, and the “Daytime Token” (SESSION) representing the market conditions during normal daytime trading hours from open to close.

Traders can freely buy and sell these two independent time exposures on-chain, making it possible for advanced long and short arbitrage strategies targeting specific time windows such as US stock earnings report nights and unexpected non-farm payroll data to achieve a high degree of programmability on blockchain for the first time.

Figure: Beans' structural design for cutting tokenized stocks into overnight and daytime exposures on the time dimension

Moreover, the Castle Labs team will launch a deep interview podcast with Re co-founder Karn Saroya this week, analyzing in full the technological breakthroughs and future development blueprint of the on-chain reinsurance track, please stay tuned.