The Price of Opacity: Private Credit, AI Lending, and the Risk Nobody Sees Until It Arrives

The Financial Times reports distressed loans at the 20 largest public private credit funds have reached their highest level since 2017. Fitch recorded a record level of private credit defaults in July. The absolute numbers remain low. The direction they are moving in, and the structural opacity that hides them from view, are the actual concern.

The private credit market has grown to an estimated two to three trillion dollars in assets. A large portion of that capital flowed into loans to AI and software companies during the boom of 2024 and 2025, when revenues were rising rapidly enough to service elevated debt loads and lenders were competing fiercely to put money to work. That was the benign phase. The numbers now arriving tell a different story.

The Financial Times reported in August 2026 that distressed loans at the 20 largest publicly listed private credit funds had reached their highest level since 2017. Fitch Ratings found that private credit defaults hit a record level in July. One large fund disclosed that approximately 7 percent of its entire credit book was in difficulty. The co-head of that fund told its own investors, in their characterisation, that the denial phase was over.

The appropriate response to those numbers is not panic. In absolute terms, default rates across the private credit sector remain low. The reason to pay attention is not where the numbers are today. It is the direction they are moving, and the structural feature that ensures nobody sees the full picture until it is too late.

The Business Model of Opacity

Private credit is not a hidden asset class by accident. It is structured around opacity as a feature, not a flaw. Loans originated outside the public bond markets are not marked to market daily. They are valued by the lenders themselves, typically quarterly, using internal models rather than observable transactions. The funds are lightly regulated, with disclosure requirements far below those of publicly listed debt vehicles.

This structure serves a purpose: investors are compensated for accepting illiquidity and limited visibility. The illiquidity premium, the additional yield available for lending in a market where capital cannot easily exit, has historically been real and persistent. The business model works as long as the loans perform.

What the model does not handle gracefully is the transition from performing to non-performing. Because prices are not observable, stress accumulates quietly. One fund knows its own book is deteriorating; it does not know what is happening across the rest of the market. The other lenders know the same about their own portfolios. Aggregate stress is invisible until someone discloses it, which creates the precise conditions in which everyone discovers a problem at the same time.

Patrick Boyle, whose analysis of the private credit sector in August 2026 drew on the Fitch and FT data, noted the structural parallel to 2008: hidden risk in an opaque, lightly-regulated corner of the financial system, concentrated in a single thematic bet (AI capex lending, as subprime mortgages were the theme then), with no mechanism for the market to price that risk in real time. The parallel is structural, not predictive. The scale and instruments are different. The mechanism is the same.

What Liquidity Actually Provides

The case for accepting illiquidity is well understood: you receive extra yield in exchange for giving up the option to sell. What is less often discussed is what that option to sell actually contains.

Daily liquidity in publicly traded instruments means daily price discovery. When conditions deteriorate, observable prices decline. That decline is painful, but it is also information: it tells every participant simultaneously where risk is, how much there is, and who holds it. The pain is distributed in real time rather than compressed into a single moment of disclosure.

An allocator evaluating a liquid, daily-marked strategy against a private credit fund is not simply choosing between a higher and a lower yield. They are choosing between two different mechanisms for how risk gets revealed. One produces continuous, observable price signals. The other produces quiet accumulation until someone discloses.

Systematic futures strategies, which trade listed instruments across equity indices, interest rates, currencies, and commodities, sit structurally at the opposite end of this spectrum from private credit. There is no illiquidity premium because there is no illiquidity. Every position is observable, priced continuously, and can be exited within the trading session. The return does not come from hiding risk inside a private structure until a quarterly valuation cycle forces it to the surface. It comes from capturing price trends and dislocations across markets that are transparent by construction.

That is not a claim that systematic strategies are superior in every market environment. It is a claim that they offer something structurally distinct: return that comes from price discovery rather than from withholding it.

The Allocator's Question

For an allocator with private credit exposure, the relevant question is not whether to panic about the FT and Fitch data. The absolute numbers do not justify panic. The relevant question is what role that exposure is playing in the portfolio, and whether the yield it provides is adequate compensation for the combination of credit risk, illiquidity, and the structural opacity that means the true risk level is unknowable until it materialises.

A 7 percent distressed rate in one large fund, announced by that fund's own management as the end of a denial phase, is not a signal to sell everything. It is a signal to think carefully about what disclosure has not yet arrived from the funds that have not yet disclosed.

The private credit market will not stop functioning because default rates are rising. Some loans will be restructured, some written off, some sold at distress prices to more patient capital. That is how credit markets have always worked. What makes the current situation worth examining is not the defaults themselves. It is the opaque structure that delays their visibility, concentrates their arrival, and leaves every allocator uncertain about how much of the problem sits in the parts of the market they cannot see.

Sources: Financial Times (August 2026); Fitch Ratings (August 2026); Patrick Boyle On Finance podcast, "How Much Would an AI Crash Destroy?" (August 2026). This publication is issued by Black Flower Capital GmbH (Zug, Switzerland) and Black Flower Capital Management SARL (Luxembourg). It is intended for professional and institutional investors only, is provided for information purposes only, and does not constitute investment advice, an offer, or a solicitation to buy or sell any financial instrument. Past performance is not indicative of future results.