Private Credit's Software Panic Will Look Overdone Within 2 Years

What's New

The private credit pullback over software exposure is an overreaction that will look overdone within 1 to 2 years. Stephen Schwarzman, co-founder and chief executive of Blackstone, argues this in an interview on Leaders with Francine Lacqua. He says these funds still generate roughly 9% returns. Defaults have risen only a little. He traces the selloff to one small hedge fund report predicting mass software bankruptcy. Allocators should underwrite software borrowers one by one and avoid exiting the category wholesale.

Why It Matters

Redeeming investors are on the other side of this argument. Blackstone limited withdrawals from BCRED at around 5% after requests rose. Other managers imposed similar limits. Critics read those gates as proof the asset class carries more risk than investors understood. Schwarzman challenges the view that AI makes software credit uninvestable. As the manager of BCRED, he benefits directly if redemption requests ease.

Between The Lines

• His main evidence comes from the head of one of the 3 big AI companies. That executive sells AI and tokens to the same software firms, so he has a stake in their survival.

• Blackstone is the largest owner of data centers in the world and works for all 6 hyperscalers. Schwarzman presents this as an early read on the AI economy. It also ties Blackstone's returns to continued AI spending.

• He concedes that some share of software companies will be mispositioned for AI. He never sizes that share.

• He treats the BCRED gate as a tool to prevent fire sales and protect value. He makes no link between the gate and credit quality.

Memorable Quotes

• “Those pieces that don't fit tell you the future.” His method is to look for the data point that breaks the pattern, which he applies here to the software panic.

• “Once you panic people, it's tough to calm them down.” He frames the redemption wave as a confidence problem driven by sentiment.

• “These funds are still generating roughly 9% returns and defaults have gone up a little bit.” He admits defaults are rising, which gives readers a number to track against his claim.

• “It'll take a year or two to play out that this was way overdone.” He puts a clock on his own call.

The Wrap

Schwarzman's case stands if defaults stay close to current levels while AI tools improve most software borrowers' businesses. It breaks if the mispositioned share proves large enough to push defaults well above today's rate. It also breaks if redemption requests keep exceeding preset limits. By his own timeline, the answer arrives by late 2028. BCRED's withdrawal pattern and software default data will settle it.