Key Takeaways
- • CoreWeave's debt valuations deteriorated sharply, with the cost of insuring its debt against default rising to roughly 1,200 basis points and some bond yields reaching 15%-16%, according to Reuters.
- • The AI cloud provider carries approximately $50bn of debt per LSEG data, making it one of the largest borrowers tied to private credit financing for data centre and computing infrastructure.
- • Blackstone, Magnetar, Carlyle, and BlackRock have anchored CoreWeave's GPU-backed facilities, including an $8.5bn facility raised in March 2026 led by Blackstone Credit & Insurance.
- • Carlyle estimates private credit may need to finance about $1tn of the AI computing buildout, exceeding half of the industry's current AUM, raising concentration-risk concerns.
- • Direct-lending fundraising hit $129.15bn in the first nine months of 2026 (up 35% year-on-year per Preqin), while Q3 direct-lending volumes fell 28% to $49.77bn per PitchBook/LCD.
Summary
A sharp deterioration in CoreWeave's debt valuations has raised concerns over private credit's growing exposure to AI infrastructure financing, with the company's borrowing costs climbing to distressed levels, according to Reuters. The AI cloud infrastructure provider has approximately $50bn in debt according to LSEG data, making it one of the largest borrowers linked to private credit financing for data centre and computing infrastructure. Last week the cost of insuring its debt against default rose to roughly 1,200 basis points, while yields on some of its bonds reached 15%-16%, reflecting heightened credit-risk concerns.
Private capital managers have played a central role in funding CoreWeave's expansion. Magnetar Capital and Blackstone Tactical Opportunities led a $2.3bn GPU-backed financing facility in 2023, followed by a $7.5bn facility in 2024 involving Blackstone, Magnetar, Carlyle, BlackRock and other investors. An additional $8.5bn facility raised in March 2026 was anchored by Blackstone Credit & Insurance. Some CoreWeave facilities are reportedly secured against GPUs and long-term customer contracts involving Microsoft, OpenAI and Meta. While contracted revenues strengthen lender security, concentrated exposure to a few technology companies creates risk if contracts are not renewed or demand weakens.
The pressures come as private credit managers expand their exposure to AI infrastructure. Carlyle recently estimated that private credit could be required to finance approximately $1tn of the AI computing buildout, more than half of the industry's current assets under management. The firm reportedly warned that rapid capital deployment could create concentration risks similar to those seen in software lending, particularly where financing depends on a limited number of customers or fast-evolving technologies. The repricing of AI infrastructure debt could prompt lenders to demand wider spreads, lower leverage, stronger covenants and tighter concentration limits on future transactions.
The concerns coincide with record private credit fundraising. Direct-lending funds raised $129.15bn in the first nine months of 2026, up 35% year-on-year, according to Preqin. However, estimated third-quarter direct-lending volumes fell 28% to $49.77bn, according to PitchBook/LCD, as competition from syndicated lenders intensified. Approximately $16.39bn of direct-lending loans were refinanced in the broadly syndicated market during the quarter, compared with just $2.60bn moving in the opposite direction. This divergence between fundraising and deployment highlights growing competition for private credit investments, even as demand for AI infrastructure financing continues to expand.
Private capital managers have played a central role in funding CoreWeave's expansion. Magnetar Capital and Blackstone Tactical Opportunities led a $2.3bn GPU-backed financing facility in 2023, followed by a $7.5bn facility in 2024 involving Blackstone, Magnetar, Carlyle, BlackRock and other investors. An additional $8.5bn facility raised in March 2026 was anchored by Blackstone Credit & Insurance. Some CoreWeave facilities are reportedly secured against GPUs and long-term customer contracts involving Microsoft, OpenAI and Meta. While contracted revenues strengthen lender security, concentrated exposure to a few technology companies creates risk if contracts are not renewed or demand weakens.
The pressures come as private credit managers expand their exposure to AI infrastructure. Carlyle recently estimated that private credit could be required to finance approximately $1tn of the AI computing buildout, more than half of the industry's current assets under management. The firm reportedly warned that rapid capital deployment could create concentration risks similar to those seen in software lending, particularly where financing depends on a limited number of customers or fast-evolving technologies. The repricing of AI infrastructure debt could prompt lenders to demand wider spreads, lower leverage, stronger covenants and tighter concentration limits on future transactions.
The concerns coincide with record private credit fundraising. Direct-lending funds raised $129.15bn in the first nine months of 2026, up 35% year-on-year, according to Preqin. However, estimated third-quarter direct-lending volumes fell 28% to $49.77bn, according to PitchBook/LCD, as competition from syndicated lenders intensified. Approximately $16.39bn of direct-lending loans were refinanced in the broadly syndicated market during the quarter, compared with just $2.60bn moving in the opposite direction. This divergence between fundraising and deployment highlights growing competition for private credit investments, even as demand for AI infrastructure financing continues to expand.
Go deeper than the headline
Research every firm in this story, track the ones that matter and get the daily briefing — free for 7 days.