Key Takeaways
- • Bloomberg's headline argues that private credit's exposure to software is larger than it appears, but the full article could not be retrieved, so specific figures are unavailable from the primary source.
- • Available commentary frames AI data-center exposure as a large, systemic credit risk that someone in the financial system must ultimately absorb.
- • Credit default swap pricing on Oracle, cited as one rough gauge of AI data-center risk, was about 0.33% of principal per year five years ago, illustrating how cheaply such risk was once insured.
Summary
Bloomberg's piece, titled "Private Credit's Software Bet Is Even Bigger Than It Appears," examines the scale of private credit lending tied to software companies. The original article could not be retrieved, so its specific data points, lender names and exposure figures cannot be confirmed here.
A related Bloomberg commentary in the supplemental results takes a broader view of AI infrastructure financing. It argues that the world holds a very large amount of systematic data-center credit risk and that someone must own it. The author notes that five years ago, few investors held meaningful AI data-center risk, and that protection was inexpensive. As an example, five-year credit default swaps on Oracle Corp., cited as one rough measure of AI data-center risk, cost about 0.33% of principal per year. At the time, an Oracle default was viewed as an unlikely, idiosyncratic event.
For institutional allocators, the combination of these themes points to concentration risk in private credit books that have grown alongside software and AI-linked borrowers. The supplemental material does not quantify current exposure or pricing, so readers should consult the full Bloomberg article for detailed figures.
A related Bloomberg commentary in the supplemental results takes a broader view of AI infrastructure financing. It argues that the world holds a very large amount of systematic data-center credit risk and that someone must own it. The author notes that five years ago, few investors held meaningful AI data-center risk, and that protection was inexpensive. As an example, five-year credit default swaps on Oracle Corp., cited as one rough measure of AI data-center risk, cost about 0.33% of principal per year. At the time, an Oracle default was viewed as an unlikely, idiosyncratic event.
For institutional allocators, the combination of these themes points to concentration risk in private credit books that have grown alongside software and AI-linked borrowers. The supplemental material does not quantify current exposure or pricing, so readers should consult the full Bloomberg article for detailed figures.
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.