JupiteX Get the app
Markets & Finance22 Nov 2026 · about 7 min

Boris Vujčić: Risks and resilience in the euro area financial system

The brief

AI-related vulnerabilities and sovereign debt fragilities were the two issues Vujčić said had become especially prominent compared with a year earlier. They matter because shocks in either area can unsettle markets and create risks beyond one company or country. For AI, the concern has widened beyond stretched share prices and a sudden shift in market mood. The speech also points to growing borrowing, concentrated exposures, faster adoption, cyber risks and reliance on a small number of critical technology providers. For sovereign debt, elevated public debt and weak fiscal fundamentals in some euro area countries raise concerns about governments’ ability to manage their debt. The speech says sovereign debt fragilities have increased on both sides of the Atlantic and are among the factors pushing up long-term yields. It also warns that global spillover risks have risen. For AI, the ECB’s concerns now span financing, valuations, cyber capabilities and the concentration of essential services, making it a broader stability issue than a year ago.

01

Which two risks did the ECB Vice-President say have become especially prominent for the euro area financial system?

AI-related vulnerabilities and sovereign debt fragilities were the two issues Vujčić said had become especially prominent compared with a year earlier. They matter because shocks in either area can unsettle markets and create risks beyond one company or country.

For AI, the concern has widened beyond stretched share prices and a sudden shift in market mood. The speech also points to growing borrowing, concentrated exposures, faster adoption, cyber risks and reliance on a small number of critical technology providers. For sovereign debt, elevated public debt and weak fiscal fundamentals in some euro area countries raise concerns about governments’ ability to manage their debt.

The speech says sovereign debt fragilities have increased on both sides of the Atlantic and are among the factors pushing up long-term yields. It also warns that global spillover risks have risen. For AI, the ECB’s concerns now span financing, valuations, cyber capabilities and the concentration of essential services, making it a broader stability issue than a year ago.

02

What does it mean for financial stability when banks and investors have concentrated exposures to AI companies?

Concentrated exposures mean that a large share of banks’ or investors’ financial risk is tied to a small number of companies or activities. If those companies run into trouble, many institutions may face losses at once. That makes the system less able to absorb a shock.

In this case, the speech points to growing debt and equity exposures to AI-linked firms. It warns that market concentration, overly optimistic earnings expectations and potential credit risks are concerns. If funding or investor confidence weakens for a heavily relied-on AI firm, the consequences could affect those holding its shares or debt.

The speech does not name particular banks or quantify their exposures. It does identify concentration as a system-wide vulnerability, alongside record AI-related debt issuance and opaque financing structures. In general, a broad mix of exposures can help limit the impact of one company’s difficulties; concentration can make losses more closely linked. That is why the ECB’s concern is not only whether an individual AI firm succeeds, but how much of the financial system depends on similar firms.

03

How large is the expected investment in AI infrastructure by major US technology firms by 2030?

The speech gives a striking figure: cumulative capital expenditure by major US technology firms is projected to reach USD 5.5 trillion by 2030. It presents this as a sign of the unprecedented financing needs across the AI supply chain. The figure is a projection, not a report of spending already completed.

The speech says large cloud and AI infrastructure providers are increasingly turning to debt markets, with issuance volumes reaching record highs this year. It also notes that firms are using opaque off-balance-sheet structures and circular financial arrangements. These examples show that AI investment is being financed through multiple channels, not just straightforward equity funding.

Borrowing is likely to increase further, according to the speech, as firms continue to build AI infrastructure. More debt can create financial exposure for firms and their funders, while opaque arrangements can make risks harder to assess. The speech does not provide a breakdown of the projected USD 5.5 trillion by company or type of investment, so it should be read as an overall projection for major US technology firms’ cumulative capital expenditure.

04

How can circular deals and off-balance-sheet borrowing make AI companies’ financial risks harder to see?

Off-balance-sheet financing is borrowing or funding arranged in a way that may not appear clearly in a firm’s main balance-sheet figures. Circular arrangements can also make funding relationships harder to follow. Together, these structures can make it difficult to see the scale and connections of financial risks.

The speech gives an example: chipmakers and cloud providers may invest in AI labs, which then commit to buying their hardware and services. The providers supply investment, while the labs become customers for the providers’ products. This links the firms’ funding and expected sales, rather than showing two fully separate relationships.

The speech describes these arrangements as opaque and says firms increasingly rely on them to fund AI infrastructure investment. It does not quantify the amount involved or say that a particular firm has hidden losses. The concern is that less transparent financing can make it harder for investors and authorities to judge how much debt exists, how dependent firms are on one another, and what might happen if expected purchases or funding do not materialise.

05

How could more capable AI systems make cyberattacks on banks and financial firms more dangerous?

The speech does not describe particular attack methods, but it directly warns that rapid advances in AI models’ cyber capabilities raise risks with systemic implications. In general, more capable tools could help cyber attackers carry out some activities faster or more effectively. That could make it harder for financial firms to prevent or respond to attacks.

The speech also highlights the high concentration of third-party providers of critical services, especially information and communications technology systems. Euro area banks and financial intermediaries rely on these providers. If a widely used provider were disrupted, many firms could be affected at the same time; this is a potential channel for a local incident to become a broader problem.

These are risks, not claims that a specific attack has occurred. The ECB’s early-release Financial Stability Review special feature published that day explores the issue further. The combination of stronger AI cyber capabilities and concentrated reliance on essential service providers means the potential consequences may extend beyond one institution. The speech therefore treats cyber resilience as part of financial stability, not only an individual firm’s technical concern.

06

Why can concerns about governments’ ability to repay debt push up long-term interest rates and affect other countries?

Sovereign debt sustainability concerns arise when investors doubt whether a government can manage its debt over time. In bond markets, greater perceived risk can lead investors to demand higher returns, which means higher yields. The speech identifies sovereign debt fragilities as one of the factors driving up long-term yields.

The speech links these concerns to elevated public debt and persistently weak fiscal fundamentals in some euro area countries. It says fragilities have increased on both sides of the Atlantic. It also lists factors compounding the risks, including growing government financing needs, a more price-sensitive investor base, inflation concerns and policy and geopolitical uncertainty.

Higher yields can make new borrowing more costly for governments, while shifts in investor demand can affect markets beyond the country where concerns first arise. The speech explicitly warns that global spillover risks have increased, although it does not detail a particular route or event. Its broader point is that sovereign debt risks are shaped by shared market conditions and can have consequences across borders.

07

What is market-based finance, and why can the financial system need stronger safeguards as it grows?

Market-based finance generally means companies and governments raise money from investors through financial markets, for example by issuing shares or bonds. This differs from relying only on loans from banks. The speech’s central message is that Europe needs more market-based finance, giving it a larger role in funding economic activity.

The speech describes large cloud and AI infrastructure providers increasingly turning to debt markets, with issuance volumes reaching record highs this year. It also notes growing equity issuance and the rise of non-bank finance. These examples show how firms can seek funding from markets and investors, rather than only through bank loans.

A larger non-bank role can create more connections across firms, investors and markets. In general, those links can allow financial stress to spread, so safeguards need to look beyond individual institutions. Vujčić says the expansion of non-banks must go hand in hand with stronger system-wide safeguards. The speech does not list specific safeguards, but it points to the need to address shared vulnerabilities such as concentrated exposures, opaque financing and reliance on critical service providers.

This brief was written by AI from the original reporting and checked by other models. Names, figures and quotes come from the source; read it for full context.

Read more in the JupiteX app

Pulse is free. New stories every 4 hours, each one broken into the questions that explain it.

Or read more news on the web