Vujčić calls for safeguards as AI and debt risks could interact
Vujčić warned that risks tied to artificial intelligence and sovereign debt could interact with vulnerabilities in non-bank finance and amplify financial instability. In a speech prepared for Frankfurt’s Euro Finance Week, he called for stronger safeguards alongside a greater role for market-based financing in Europe.
Key takeaways · 4
- 01
Vujčić said major US technology companies were projected to spend a cumulative $5.5 trillion on capital expenditure through 2030.
- 02
He warned that opaque financing and supplier investments tied to customer purchases can make AI-sector exposures harder to assess.
- 03
His proposals include matching investment-fund redemption terms more closely to asset liquidity and increasing usable liquidity buffers at money market funds.
- 04
He said existing regulatory capital buffers did not appear to constrain lending and should be preserved.
Risks extend beyond AI share prices
Vujčić said AI-related risks now extend beyond expensive equities to debt financing, concentrated exposures, wider technology adoption and cyberattacks.[1] Companies across the AI supply chain face what he described as unprecedented financing needs, while major cloud and infrastructure providers are increasingly borrowing in debt markets.[1] He said those providers are likely to borrow more, and cited a projection that major US technology companies will spend a cumulative $5.5 trillion on capital expenditure through 2030.[1] He also pointed to opaque off-balance-sheet financing and arrangements where suppliers invest in customers that commit to buy their products.[1]
Debt and market exposure could compound
Vujčić said sovereign-debt fragilities had increased on both sides of the Atlantic and were among the factors pushing up long-term yields.[1] He warned that elevated debt, higher financing costs and structural growth constraints could strain fiscal positions and revive concerns about highly indebted countries.[1] Investors’ capacity to absorb additional government-bond issuance will be crucial to orderly markets, he said.[1] He also warned that higher Japanese yields could lead large Japanese investors to reduce euro-area sovereign-bond holdings.[1] Separately, technology-related debt issuance could raise financing costs or crowd out other investment.[1]
Non-bank vulnerabilities are a focus
Vujčić described hedge funds as an important but procyclical source of liquidity.[1] He said investment funds’ growing AI exposures leave them vulnerable to valuation losses and redemptions that could force asset sales; concentration and leverage could magnify that stress.[1] He warned that risks occurring together could transmit losses and liquidity pressures across markets.[1] His proposed responses include stronger EU-level supervision and measures addressing leverage, liquidity mismatches and connections with banks.[1] He also proposed considering leverage limits or minimum margin and haircut requirements for hedge funds.[1]
Safeguards alongside market-based finance
Vujčić called for aligning investment-fund redemption terms more closely with the liquidity of their assets and increasing usable liquidity buffers at money market funds.[1] He said existing regulatory capital buffers did not appear to constrain lending and should be preserved.[1] He also said the euro area had remained resilient despite multiple challenges, while third-quarter growth was less dynamic than in the previous quarter.[1] Underlying inflation remained broadly stable, and wages had not shown a material response at that stage, he said.[1]
The speech links AI investment, public borrowing and non-bank finance as connected sources of potential market stress. Professionals assessing financing, investment exposures or liquidity planning can use the proposals to identify where risk controls may need closer review.
Why it matters
AI training for your team
kju.ai is built for teams: daily sessions, team analytics, and rollout support for your organization.
How this developed
16 November 2026
Vujčić calls for safeguards as AI and debt risks could interact
