Field report SE-80 / Q1 · sheet /macro
Boardroom MacroFiled from the territory
CoreWeave bond yield tops 16% amid $570 billion in projected AI debt
AI funding fell from $239.1 billion in Q1 to $103.3 billion in Q3 as CoreWeave’s 2031 bond yield topped 16% and Nvidia lost nearly $2 trillion in value.
- Grid
- SE-80 / Q1
- Surveyed
- November 19, 2026
- Reading
- 4 min read
- Correspondent
- Priya Raman

Field notes
- Global AI funding fell from $239.1 billion in Q1 2026 to $153.1 billion in Q2 and $103.3 billion in Q3.
- Morgan Stanley projects nearly $570 billion of AI-related debt issuance in 2026, more than double 2025; nearly $236 billion was issued by May’s end.
- Nvidia lost nearly $2 trillion—more than 30% of market value—after its October 6 peak, and Jensen Huang resigned on November 12 after 33 years.
- CoreWeave’s 2031 bond yield passed 16% in October; Reuters puts its five-year default probability at nearly 63%.
- Nvidia guaranteed up to $105 billion in lease obligations tied to an Ohio data-center campus built for OpenAI.
By Priya Raman
AI’s financing boom is entering its first serious credit stress test, not yet a full debt crisis, as Morgan Stanley projects nearly $570 billion of global AI-related debt issuance in 2026, more than double last year. Nearly $236 billion was issued by the end of May. The wager now rests on demand that remains partly circular.
What does the funding slowdown mean?
AI Funding Tracker reports global AI funding of $239.1 billion in Q1 2026, $153.1 billion in Q2 and $103.3 billion in Q3. The figures do not establish whether Q3 was the bottom, but they remove the old assumption that capital will keep arriving without limit.
Technology will keep advancing. But the gold rush has ended: easy financing is gone, and unquestioned valuations have lost their exemption. The analysis is explicit: “Having great technology does not justify every valuation. The people behind the tech must be able to responsibly steward a company and its mission.”
What has Nvidia’s repricing exposed?
CNBC reports that Nvidia has lost nearly $2 trillion in market value since its October 6 record, more than 30% of its peak value, in under six weeks. At that peak, Nvidia was the most valuable company on earth.
On November 12, 2026, Jensen Huang resigned after 33 years. Nvidia’s CFO became interim CEO, and Congress called Huang to testify. The company now carries succession risk and national-security exposure at the same time. Recent earnings reports may have softened the sting, but they leave core questions unanswered.
How does circular financing distort demand?
Bloomberg’s chart of circular AI financing places Nvidia at the center of a global web spanning Microsoft, Google, Amazon, OpenAI, Anthropic, Oracle and CoreWeave. Billions move among them through equity stakes, chip purchases, cloud contracts and guarantees.
Bloomberg’s chart shows the sequence: chipmakers invest in AI companies that buy their chips; cloud providers invest in developers that commit billions back to cloud services; lenders then finance data centers against those commitments. The unresolved question is how much demand is organic and how much the companies that benefit from it finance.
CNBC reports that Nvidia has guaranteed up to $105 billion in lease obligations tied to an Ohio data-center campus built for OpenAI, Nvidia’s own customer.
Capital circulates and valuations rise, a loop that can create growth but also make dependent demand look independent. When suppliers finance customers and both valuations depend on the same revenue, one default can travel through the circle. The analysis rejects both easy labels: calling the system a Ponzi scheme would be irresponsible, while calling it safe would be worse.
What has the credit market priced in?
In October, the yield on CoreWeave’s 2031 bonds passed 16%, a level traditionally associated with distress; yield means the income investors demand from a bond. Reuters calculates a nearly 63% chance of default, or failure to repay, within five years, even though Nvidia has an agreement to buy CoreWeave’s unsold capacity through 2032.
Lenders now demand higher yields and tighter covenants, meaning stricter loan repayment conditions. Credit investors have begun to separate financial strength from AI enthusiasm. Their question is blunt: who absorbs losses when a customer, supplier and creditor occupy the same circle?
Which projects can survive the repricing?
Nvidia remains highly profitable, and Microsoft, Google and Amazon retain deep financial resources. Debt-heavy neoclouds—meaning debt-laden cloud computing providers—and data-center developers face power shortages, construction delays, rising borrowing costs and local opposition.
Billions have been committed to AI campuses before demand and profitability are proven. Data centers will persist, but speculative projects face trouble. The analysis expects a wave of distressed and underused projects from 2028 through 2031; survivors will have cheap power and creditworthy customers on long contracts.
The analysis states the test plainly: “A gigawatt announced is not a dollar earned.” A gigawatt measures power capacity, not revenue.
How will regulation decide who scales?
The EU AI Act is already reshaping compliance. Individual US states are drafting and passing different legislation, and after the recent midterm elections, Washington will press harder on safety, energy costs, competition and national security. The era of building first and answering questions later has ended; regulation will determine who can scale and who cannot.
Is AI repeating the fiber build-out?
The fiber build-out of the late 1990s produced genuine technological progress but destroyed vast amounts of investor capital. The internet survived; many companies that financed it did not. AI is real, and demand for compute remains large. Technological inevitability does not guarantee financial sustainability, however.
The analysis remains bullish on AI’s long-term contribution to economies and industries, but far less bullish on the financial engineering around it. For now, this falls short of a debt crisis. Investors face the first serious stress test of the financial system behind AI.
The central risk is that financing assumptions break before AI delivers its economic returns. The next phase’s winners will generate sustainable returns on invested capital, rather than merely post GPU (graphics processing unit) counts and gigawatt announcements, which will matter less. For the most interconnected companies, their biggest customers and biggest creditors may sit in the same circle.
via fortune.com/feed/ (Source)
More from Priya Raman
Show full bio
Senior reporter covering technology and product news in corporate strategy and chief-executive leadership at CEO Frontier.
459 articles
Adjacent sheets
- Oxford Economics Warns Boom Is Sowing Seeds of a Deeper Recession
- Nvidia's 9.3% Drop Meets Governance Strain and Broken Supports
- Microsoft Winds Down Oracle AI Deal as Chip Age Sparks Investor Alarm
- AI selloff exposes concentration and debt risks after 5.5% Nasdaq fall
- S&P 500 to Sit 17% Off Peak by End-2027, Capital Economics Warns