
ECB names AI and sovereign debt as leading financial risks
ECB Vice President Boris Vujcic says AI funding and sovereign debt now pose the most prominent new risks to euro zone stability, with major tech firms’ capex projected at $5.5 trillion.
- By
- Grace Kim
- Filed
- Channel
- AI & Compute
- Read
- 3 min read
By Grace Kim
Major U.S. technology firms’ cumulative capital spending is projected to reach $5.5 trillion by 2030, ECB Vice President Boris Vujcic said Monday, Nov. 16, at Euro Finance Week in Frankfurt.
He spoke two days before the ECB publishes its Financial Stability Review. Vujcic identified AI and sovereign debt markets as the two risks standing out most compared with a year earlier.
What changed in the ECB’s risk assessment?
Vujcic said: “While most risks identified in the FSR have been on our radar for quite some time, two stand out very prominently compared with the situation a year ago.”
A year earlier, the ECB tied AI risk mainly to stretched stock valuations and a possible shift in market sentiment. By Nov. 16, Vujcic said the risk of a market repricing had increased further. He also cited AI funding moving from equity into debt markets, concentrated exposures, faster AI adoption and potential cyberattacks.
How is AI infrastructure being financed?
“The scale of firms’ financing needs across the AI supply chain is unprecedented,” Vujcic said. Large cloud and AI infrastructure providers have turned to debt markets, with issuance at record highs this year.
He said borrowing is likely to rise further because major U.S. technology firms’ cumulative capital spending is projected to reach $5.5 trillion by 2030. The figure is a projection cited by Vujcic, not a confirmed spending total.
Companies increasingly rely on opaque off-balance-sheet financing structures and circular financial arrangements, he said. In one example, chipmakers and cloud providers invest in AI labs that then commit to buy their hardware and services.
Equity issuance is also surging. Vujcic said IPO volumes are on track to exceed the dot-com bubble high relative to U.S. GDP. Such issuance surges have often coincided with market peaks.
He characterized the market correction after the U.S. midterm elections earlier that month as contained. He nevertheless urged caution about AI valuations. The warning precedes Nvidia’s Tuesday results and Anthropic’s planned Thursday Nasdaq debut.
The Federal Reserve’s financial stability report the previous week had identified AI as the top concern among market contacts.
Could AI debt compete with sovereign borrowing?
Vujcic said sovereign debt fragilities have increased further on both sides of the Atlantic. He listed those fragilities among factors pushing up long-term yields and called the notable rise in debt issuance by hyperscalers a compounding risk.
He asked whether AI-related borrowing could compete with governments for capital or crowd out other investment. Large Japanese investors may reduce euro-area government bond holdings further if Japanese yields rise, he said.
Where could a valuation reversal become a liquidity problem?
Private credit and private equity funds now play a growing role in financing AI infrastructure, he said. Euro-area investment funds have significantly increased their AI exposure.
A reassessment of the AI investment narrative could trigger redemptions and forced asset sales, Vujcic warned, linking concentrated technology exposure to the liquidity demands facing funds.
What safeguards did Vujcic propose?
Vujcic addressed three funding risks: hedge-fund leverage, money-market-fund liquidity, and investment-fund redemption terms that do not match the liquidity of the underlying assets they hold.
He called for:
- Leverage limits, minimum margins or haircut requirements for hedge funds.
- Larger and more usable liquidity buffers at money market funds.
- Redemption terms at investment funds that match asset liquidity more closely.
What did he say about cyberattacks and inflation?
Increasingly capable AI systems magnify the scale and speed of cyberattacks and may, at least in the near term, shift the balance toward attackers, Vujcic said.
Euro-area inflation rose to 4% in October from 3.8% in September, mainly because of energy prices. He said recent energy-price moves may push inflation higher.
If projected technology spending generates additional debt issuance, borrowing is likely to rise further; his proposed fund rules target the redemptions and forced sales that could follow a reassessment of AI valuations.
Source: mlq.ai
More from Grace Kim
Show full bio
Market editor covering people, launches and day-to-day practice in Semiconductors, chip manufacturing, hardware, AI compute at Chip Dispatch.
1,035 articles
Related articles
87-of-fed-survey-contacts-now-call-ai-a-top-financial-risk
87% of Fed Survey Contacts Now Call AI a Top Financial Risk
ai-debt-may-exceed-800-billion-as-bowman-flags-equity-risk
AI debt may exceed $800 billion as Bowman flags equity risk
kkr-warns-investors-that-ai-debt-is-piling-up-faster-than-returns-90b2cf4f
KKR Warns Investors That AI Debt Is Piling Up Faster Than Returns
ai-boom-drives-surge-in-debt-financing-across-chip-sector-0ec9d7f6
AI Boom Drives Surge in Debt Financing Across Chip Sector



