BTC
ETH
HTX
SOL
BNB
View Market
简中
繁中
English
日本語
한국어
ภาษาไทย
Tiếng Việt

2008 Again? CDS Returns to the Market Spotlight

星球君的朋友们
Odaily资深作者
This article is about 2621 words, reading the full article takes about 4 minutes
An AI debt bubble is quietly emerging, and Wall Street is replaying the credit derivatives frenzy at the fastest pace in history.
AI Summary
Expand
  • Core Viewpoint: The heating up of the AI debt bubble is driving an explosive surge in credit default swap (CDS) trading volume, with the market shifting from index hedging to single-name CDS—a trend that reminds some participants of the credit derivatives market on the eve of the 2007-2008 financial crisis.
  • Key Elements:
    1. Weekly average trading volume of AI-related CDS on Oracle, SpaceX, Alphabet and others has soared from less than $2 billion in July to more than $9 billion in September, a increase of over 4.5 times; Goldman Sachs data shows that AI single-name CDS trading volume grew 400% year-over-year.
    2. So far in 2026, AI-related companies have issued approximately $466 billion in bonds, up more than 3.6 times from $101 billion in the same period of 2025, making AI companies the largest group of borrowers in the bond market.
    3. Price pressure is significant: Meta's 5-year CDS rose to a record high of 108 basis points, Oracle surged from 144 basis points at the start of the year to 322 basis points, and SpaceX doubled from 110 basis points to 226 basis points.
    4. The market is turning to CDS rather than stock put options because the latter are "extremely difficult and costly" for hedging risks of hyperscale cloud computing companies; chipmakers such as Nvidia and AMD are viewed as borrowers after providing vendor financing, and their related CDS trading volume has surged about 10 times.
    5. CoreWeave has become the most vulnerable link: its 5-year CDS is quoted at about 1,090 basis points, implying a probability of default within five years of about 63%, and the market is already pricing in a default scenario rather than merely hedging risk premiums.
    6. The transmission chain extends to financial institutions: CDS in the financial and insurance sectors is also in high demand, with the market worried about institutions' investment concentration risk amid an AI trade pullback.
    7. Goldman Sachs predicts that direct bond issuance by hyperscale cloud computing companies will reach another $420 billion in 2027, and each new bond issuance will generate new hedging demand.

Original Author: Zhao Ying

Original Source: Wallstreetcn

As the AI debt bubble heats up, credit default swap trading volume has exploded, and Wall Street is replaying a familiar hedging frenzy.

According to the latest research data from Morgan Stanley, in September of this year, the average weekly trading volume of credit default swaps (CDS) related to Oracle, SpaceX, Alphabet, Amazon, Meta, Broadcom and Nvidia exceeded $9 billion, surging more than 4.5 times from less than $2 billion in July.

Meanwhile, data from Goldman Sachs' Fixed Income, Currencies and Commodities (FICC) division shows that AI single-name CDS trading volume grew 400% year-over-year, with the number of liquid names expanding from 4 to 10-12. The pace of this expansion has reminded some market participants of the credit derivatives market on the eve of the 2007-2008 financial crisis.

Avishek Biswas, a credit strategist at Goldman Sachs, noted in his early-October credit market strategy report that the weekly notional trading volume per name for AI-related CDS has jumped from $50 million to $100 million last year to the current level of hundreds of millions of dollars per day. Manish Kabra, head of US equity strategy at Societe Generale, said CDS has become a "key barometer" for gauging the financing conditions of hyperscalers and overall AI trade sentiment, and listed it as one of the bank's core trades in its Q4 2026 US equity outlook, saying outright that investors should "watch CDS, not EPS."

AI Debt Scale Balloons, CDS Market Explodes in Turn

Behind the AI boom lies an unprecedented debt expansion. According to Morgan Stanley data, year-to-date in 2026, AI-related companies have issued approximately $466 billion in bonds, up more than 3.6 times from $101 billion in the same period of 2025. As AI companies leap to become the largest group of borrowers in the bond market, demand for default protection around these names has grown exponentially.

On the pricing side, the market has already sent clear signals of stress. Five-year Meta CDS rose to a record high of 108 basis points this month, Nvidia reached 101 basis points, and protection costs for Alphabet and Amazon also hit all-time highs. Oracle CDS quotes surged from 144 basis points at the start of the year to 322 basis points, while SpaceX doubled from 110 basis points at the end of June to a record high of 226 basis points.

Alisha Pasi, a cross-asset credit specialist at Goldman Sachs, disclosed in her latest weekly report that the bank has recently received a record number of inquiries from non-traditional accounts regarding CoreWeave debt and Oracle CDS. Demand was so strong that Goldman Sachs set up a dedicated AI cross-capital-structure investor call, with participants spanning multiple desks including tech, media and telecom equities, convertible bonds, investment-grade cash and "distressed/AI debt trading"—the latter having become an independent functional line.

Why CDS Instead of Equity Put Options

A structural shift is underway in the choice of hedging tools. Societe Generale's Kabra said outright that using equity put options to hedge hyperscaler risk is "extremely difficult and extremely costly." This judgment explains why a growing number of equity accounts are crossing asset classes to seek protection in the credit derivatives market.

Morgan Stanley, in its October 1 report "The State of AI Financing: A Pause, Not a Retreat," likewise noted that buying AI-related CDS protection "remains our preferred way to participate in the AI theme in the credit derivatives market," and expects "non-economic hedging demand" to continue rising as banks need to manage counterparty exposure.

Miles McCarthy, an IG trader at Goldman Sachs, previously noted that CDS buyers come from diverse sources: some accounts engage in directional trades based on leveraged or unfunded credit views, some exploit relative-value dislocations between CDS and bonds, and still others—equity accounts—use CDS to hedge public-market and private AI-related equity portfolios. Notably, as chipmakers such as Nvidia, AMD and Broadcom begin providing vendor financing and guarantees to buyers, the market has started treating them like borrowers, and related CDS trading volume has surged roughly 10-fold.

In addition, Goldman Sachs credit sales strategy data shows that last week the CDX IG index widened only 4.4 basis points to 62.5 basis points, while core AI names could move 30 basis points in a single day. This divergence is precisely the fundamental reason the market has shifted from index hedging to single-name CDS—the index can no longer precisely capture the risk exposure investors actually hold.

CoreWeave: The Weakest Link

Among all AI-related names, CoreWeave is the stress point drawing the most concentrated market attention. According to the Financial Times, CoreWeave's 5-year CDS is currently quoted at around 1,090 basis points; buying $10 million of protection requires an upfront payment of $1.82 million, followed by $125,000 per quarter—the first-year cost alone accounts for roughly 23% of the notional principal.

Gavan Nolan, a credit analyst at S&P Global, noted that "once a name's CDS trades above 1,000 basis points, the pricing logic begins to shift toward expected recovery rates," meaning the market is already pricing in a default scenario rather than merely hedging risk premium.

Goldman Sachs had previously upgraded CoreWeave, but HSBC followed with a $39 price target and a warning that "the more they invest, the more value is destroyed." On October 14, CoreWeave CDS broke through from 1,000 basis points to 1,050 basis points intraday and further rose to 1,117 basis points at the close, implying a roughly 63% probability of default within five years. Meanwhile, Elice, a data center operator behind South Korea's largest IPO, plunged 32% within seconds of listing, further corroborating market doubts about returns on AI infrastructure investment.

Robert Dafforn of Polus Capital told the Financial Times that the circularity in the AI financing chain means "whichever link is weakest will drag down the entire system first."

From Hedging Tool to Systemic Risk Signal

Does the current CDS boom foreshadow deeper systemic risk?

Structurally, the demand is self-reinforcing. Goldman Sachs forecasts that hyperscalers will issue another $420 billion in direct debt in 2027, plus about $300 billion in project financing; each new bond issuance will spawn new hedging demand, and each newly established special purpose vehicle (SPV), lease guarantee or vendor financing arrangement will create new counterparty risk exposure.

Pasi also noted in her report that CDS on financial and insurance sectors were similarly sought after this week, "with the market worried about these institutions' concentration risk in the AI trade correction"—meaning the transmission chain has extended from AI borrowers to the balance sheets of financial institutions holding their bonds.

Societe Generale data shows that hyperscaler 5-year CDS has more than tripled since the start of 2025, while forward P/E valuations have compressed in tandem, and the divergence between the two is narrowing.

The last time real-money accounts returned to the single-name CDS market on a large scale after more than 15 years of dormancy was in 2007. They had equally good reasons to do so then.

finance
invest
AI
Welcome to Join Odaily Official Community