Posted on 03/21/2026 7:57:56 PM PDT by SeekAndFind
It’s not just the stocks of software companies that are taking hits from worries about artificial intelligence. In the private credit market, where loans to software firms had become a favored sector over the last five years, sentiment has soured.
The concern in both markets is that AI threatens the profit margins and underlying business models of many software companies by reducing the barriers to entry and enabling customers to build their own software. Those worries strike at the heart of the reasons many lenders had for issuing the loans: fat profit margins for software companies, stable customer bases, and reliable revenue from license subscriptions.
Unlike in stocks, the degree to which investors are marking down prices on software loans is difficult to assess, in part because there is no active secondary market for trading private credit loans.
Private credit lending accelerated in 2020, when uncertainty about covid-19 and the resulting extreme market volatility closed off other sources of financing to borrowers, such as that from the syndicated loan market. The direct lending segment of private credit is now estimated at $1.7 trillion.
A significant amount of private credit goes to software companies.
This made sense because most software firms are startups and too small to sell corporate bonds. They don't have the track record or financial statements that banks require.
Private credit became their lifeline. Need $10 million to expand? A private credit fund would provide it at 12% interest instead of the 6% a bank might charge (if the bank would lend at all).
The borrowers got capital. The lenders got higher returns. Everyone seemed happy.
Beyond the pandemic, software was thought to have sticky customer bases, due to high switching costs. Lenders liked the reliable revenue streams provided by subscription-based enterprise accounts, as opposed to more fickle consumer customers. Ultimately, software borrowers drove the growth of the private credit market, with record private credit loans issued to software companies.
This month, the launch of Anthropic’s Claude Cowork has fueled a global selloff of publicly traded software and IT companies, on expectations that new AI tools mean people will create bespoke software solutions, eliminating the need for costly subscriptions.
Here's where things get complicated.
Many software companies exist to build tools, apps, and platforms that businesses need. But artificial intelligence is rapidly changing that equation.
It raised an unconventional question. If AI can develop software with just a prompt, do we need as many software companies? Can’t AI just replace the products these firms sell?
These questions aren't theoretical anymore, and they're affecting valuations right now. Investors started worrying that software companies might struggle to repay loans if AI makes their products obsolete. Share prices for private credit managers dropped as this concern spread.
That drop itself isn't catastrophic. But it might be an early warning signal.
“Credit investors should be most concerned about software concerns with high technical debt, fragmented data silos preventing effective AI training, no clear path from copilot to agent, and business models vulnerable to incumbent platform vendors rolling out comparable AI features at lower incremental cost,” PitchBook analyst Derek Hernandez wrote in a January report. “Software companies with these traits face existential margin compression and potential obsolescence as the market bifurcates between AI-native and AI-embedded category leaders and legacy laggards unable to transform their architectures fast enough.”
The impact of these concerns could be seen as private credit providers—and the private equity firms that own them—have begun talking about the sector in far greater detail than before in response to investors’ questions.
For example, on a Feb. 4 shareholder call by Ares Capital, a private credit provider that has lent to software, CEO Kort Schnabel said that AI would disrupt many software companies with single-function software products that produce content or analyze and visualize data. (Schnabel said Ares was “not seeing weakness” in its portfolio of software loans “at all.”)
Private credit doesn't exist in isolation. It's connected to the broader financial system in ways that create potential problems.
Big banks, the ones considered too big to fail, often lend money to private credit managers. Those managers then use that borrowed money to make riskier loans to software companies and other borrowers.
Think of it like a chain: Banks lend to private credit funds. Private credit funds lend to risky companies. If those risky companies can't repay, the private credit funds take losses. If the funds take big enough losses, they can't repay the banks.
This creates contagion risk. A problem in one part of the chain can spread to others.
Some limited transparency can come from the business development companies that make the loans, which are part of credit funds and publicly traded entities of asset managers. In BDC portfolios, asset managers provide a “fair value” of private credit loans. These valuations may come from an independent third party. But fair values often vary widely between lenders holding the same loan, so they are open to interpretation.
Recently, investors have focused on how BDCs classify private credit loans by sector, with some BDCs categorizing them by end markets in healthcare or food, rather than software.
However, private credit lenders consider them buy-and-hold investments, and typically plan to retain them until maturity or repayment. As a result, many private credit loans continue to be marked in BDC loan portfolios at the price they were originally issued (generally at 100, known as par), indicating that a lender still expects full repayment of principal. Not only that, but any enhanced level of diligence is nearly impossible for private credit borrower companies, which do not publicly disclose earnings.
In peak credit market conditions, when lenders actively competed for the chance to lend to software and tech companies, private credit lenders agreed to loan structures and terms that were not necessarily available in the syndicated loan market.
For instance, private credit lenders underwrote loans based on annual recurring revenue streams, rather than earnings, allowing a borrower more financial flexibility. Loans would be structured so that after a time—say, three years—covenants would change from revenue to EBITDA metrics.
Private equity firms argued that the recurring revenue structure is particularly useful for borrowers in a growth phase, and many of the borrower companies have performed well. Many software companies subsequently flipped to EBITDA-based covenants.
The willingness of lenders to underwrite loans based on these “alternative” financial metrics led to erosion of the syndicated loan market in favor of private debt in 2020-23. Since then, many private credit loan borrowers have refinanced debt in the syndicated loan market, securing even lower spreads.
Private credit lenders have also been willing to allow interest to be paid in kind, like by adding principal to a loan. Some borrowers were allowed to pay PIK interest from day one, and some were able to negotiate with lenders the addition of PIK interest later to help a borrower conserve cash.
Dear FRiends,
We need your continuing support to keep FR funded. Your donations are our sole source of funding. No sugar daddies, no advertisers, no paid memberships, no commercial sales, no gimmicks, no tax subsidies. No spam, no pop-ups, no ad trackers.
If you enjoy using FR and agree it's a worthwhile endeavor, please consider making a contribution today:
Click here: to donate by Credit Card
Or here: to donate by PayPal
Or by mail to: Free Republic, LLC - PO Box 9771 - Fresno, CA 93794
Thank you very much and God bless you,
Jim
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.