In the third quarter of 2026, Wall Street almost simultaneously highlighted the same keyword during their earnings call — tokenization. Goldman Sachs, Wells Fargo, and Morgan Stanley all described it as a business growth opportunity for the next few years, emphasizing that it can open new spaces in product design and settlement efficiency while deliberately avoiding specific revenue expectations. Meanwhile, on-chain data has shown that traditional assets are "going overseas" through new channels: according to Chainalysis data, Italy, Portugal, and Singapore collectively hold 66.9% of geo-located tokenized U.S. Treasury bonds, while the U.S. only accounts for 10.5%. Tokenized U.S. Treasury bonds have attracted more overseas funds into this asset class by lowering the minimum holding threshold. However, outside the narrative of the "new growth engine," risk volatility has been accumulating resonance signals during the same period — according to AiCoin data, on October 18, 2026, from 05:22 to 05:32 Beijing time, U.S. stock AI concept perpetual contracts experienced a sharp drop followed by a swift recovery; Western Digital and Nebius contracts saw a drop of nearly 7% at one point, while Bloomberg reported that CoreWeave's high-yield bond yield rose to about 15% this week. The market is beginning to reassess the feasibility of AI companies continually financing through high-yield bonds. Tokenization has been promoted to the forefront by Wall Street as a new growth engine, but the extreme volatility of on-chain derivatives and AI assets within the same time window reminds us that this engine has been accompanied by noticeable risk resonance from the start.

Goldman Sachs and other banks view tokenization as a growth engine

During the third quarter earnings call of 2026, Goldman Sachs, Wells Fargo, and Morgan Stanley all pushed tokenization to the forefront of their growth narrative using almost the same language, but deliberately avoided specific revenue forecasts. Goldman Sachs' CEO directly linked tokenization with DeFi and prediction markets, emphasizing “more products, more trading activity, and more new clients,” rather than a single business line's profit model. This statement positions tokenization as a technological layer that can overlay existing investment banking, asset management, and trading businesses: as long as traditional assets can be mapped on-chain and connected to broader off-chain and on-chain trading demands, Goldman Sachs believes it has the ability to capture this incremental growth into its existing system.

Wells Fargo has gone a step further by already using tokenized deposits in cross-border payments, allowing funds to reach the recipient's account in about two minutes. This pilot itself serves as a model for the path from concept to production environment. Based on this, Wells Fargo has begun exploring tokenized securities, RWA, and crypto trading services for wealth management clients, with the logic of progressively expanding verified efficiency advantages to higher-margin assets that have stricter compliance requirements. Morgan Stanley has put it even more simply and directly: tokenization can accelerate and improve financial transactions and settlement processes, serving as a tool to enhance existing business turnover rather than being an "enemy" to snatch business. When viewed alongside Goldman Sachs' active exploration to expand the use of DTCC tokenization services to reduce costs, automate trading, and improve settlement efficiency, these banks share a common goal of rewriting back-office settlement and product packaging with tokenization, rather than the front-office risk architecture, which is why they dared to label tokenization as a growth engine during their calls while remaining cautious towards revenue streams.

Italy and Portugal emerge as major holders of tokenized U.S. Treasury bonds

If Wall Street is still focused on the "efficiency story" during the earnings calls, on-chain data has begun to provide more geopolitically colored footnotes. Chainalysis data from 2026 shows that Italy, Portugal, and Singapore together hold 66.9% of geo-locatable tokenized U.S. Treasury bond assets, while U.S. domestic addresses account for only 10.5%. In other words, this type of on-chain asset linked to U.S. credit has already formed a "offshore holding alliance" dominated by Southern European marginal countries and Asian financial hubs, with the real large players appearing first in overseas addresses rather than Wall Street itself.

This structure is no coincidence. Tokenized U.S. Treasury bonds have lowered the minimum holding threshold, allowing smaller amounts of capital to directly enter bond rate assets on-chain. According to the Chainalysis report, this is significantly expanding overseas markets. From the perspective of asset and policy games, the low proportion of domestic holdings in the U.S. means that the U.S. can let more overseas funds take on its credit products without significantly squeezing the domestic risk budget. On the other hand, it objectively outsources the liquidity and pricing nerves of tokenized U.S. Treasury bonds to regions concentrated in Italy, Portugal, and Singapore. This mismatch where the issuer is in the U.S. but the on-chain holding focus is overseas is quietly changing the future distribution of regulatory and market discourse surrounding tokenized U.S. Treasury bonds.

pump.fun utilizes Hyperliquid for perpetual contracts

As tokenization pushes more traditional assets onto the blockchain, the derivative infrastructure responsible for "amplifying risks" is also quietly rewriting its business model. Recently, pump.fun announced the introduction of perpetual contract trading services based on Hyperliquid's Builder Code mechanism for external project access, effectively embedding its front-end directly onto Hyperliquid's perpetual layer. From a single source, this service charges a maximum of 0.1% Builder Fee per order, becoming a new layer of cost added on top of Hyperliquid's systemic rates. Currently, the specific asset list supported by pump.fun's perpetual has not been made public, nor has the launch time or transaction volume been disclosed, which feels more like a preemptive layout, with follow-up on-chain performance set to validate its chosen bet.

The design of the Builder Fee determines whether this model can stand or not. For pump.fun, charging a maximum of 0.1% per order naturally drives it to pursue higher user trading frequencies, converting each click into measurable protocol income; for traders, whether they are willing to pay this additional fee will directly reflect in their routing choices and order habits — they will need to reassess convenience versus cost, deciding whether to stay on the Hyperliquid native interface or to access the same perpetual contracts through pump.fun's "tokenized front end." Currently, due to the lack of a public asset list and transaction volume, outsiders cannot yet determine the actual impact of pump.fun's perpetual on the existing on-chain derivatives landscape. It can only be seen as a signal: in the ongoing extension of the tokenization narrative, who can master the user entry points and who can lock in their revenue stream through the Builder Fee is becoming a key variable in the next stage of competition in the perpetual track.

Sharp drop in AI concept stock perpetuals exposes emotional fragility

According to AiCoin data, on October 18, 2026, from 05:22 to 05:32 Beijing time, U.S. stock AI concept perpetual contracts suddenly exhibited a nearly vertical downward trend. Western Digital's perpetual contract plunged 7.03% in under ten minutes, dipping to as low as $298.70, before quickly returning to around $318.65 to narrow the losses; almost at the same moment, Nebius' perpetual contract saw a maximum drop of approximately 6.97%. This set of synchronous sharp drops and quick rebounds appears to be a concentrated trigger of high-leverage positions during a low liquidity period, rather than a trend driven by fundamental news, with the price range and rebound speed squeezed tightly within a single 5-minute candlestick.

The glaring aspect is that this emotional stampede occurred during a time window that overlaps significantly with Goldman Sachs, Wells Fargo, and Morgan Stanley’s optimistic rhetoric emphasizing that "tokenization will bring product innovation and settlement efficiency improvements." In macro narrative terms, tokenization is seen as Wall Street's new growth lever, while on-chain perpetuals are packaged as the "new infrastructure" supporting AI concept stocks; however, any near 7% flash crash during low liquidity periods followed by a rapid recovery is enough to expose this track's high sensitivity to short-term volatility. As the tokenization story continues to be told smoothly, any violent oscillation of perpetual contracts related to AI assets will become a real-time stress test for investors to gauge whether this narrative can withstand emotional pullbacks.

CoreWeave's 15% bond yield indicates AI risks

This week, Bloomberg reported that CoreWeave's high-yield bond yield has risen to about 15%, almost pushing this representative AI company directly into the highest risk compensation bracket. The surge in yield reflects market doubts about AI companies' ability to continue relying on high-yield bond financing and signals tightening overall risk preferences among high-yield bond investors. CoreWeave's co-founder publicly stated that the company is exploring entry into the investment-grade debt market and seeking funding from insurance institutions, which indicates that management intends to lower financing costs while also confirming that current financing pressures are becoming significant. However, without more complete financial data and details of the debt scale, external assessments of its real risk exposure remain limited. The rising tokenization narrative on Wall Street coincides with the high volatility of AI assets in both equities and bonds and rising financing costs; this combination of "15% high-yield bond yield + violent on-chain perpetual oscillation" is becoming a litmus test to examine whether the dual stories of tokenization and AI can endure emotional pullbacks. What is worth closely tracking next is whether AI companies can indeed access the investment-grade funding pool, whether insurance funds genuinely enter the market for allocation, and whether large banks begin disclosing more specific tokenization revenue and related risk exposures in future earnings reports.

Join our community to discuss and grow stronger together!
AiCoin exclusive Hyperliquid benefits: https://app.hyperliquid.xyz/join/AICOIN88
AiCoin exclusive Aster benefits: https://www.asterdex.com/zh-CN/referral/9C50e2
On-chain Telegram community: https://t.me/AiCoinWhaleData
On-chain community: https://www.aicoin.com/link/chat?cid=N6OVMor5g
AiCoin on-chain Twitter: https://x.com/aicoinwhaledata

免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到[email protected]，本平台相关工作人员将会进行核查。