Private-credit funds manage exposure to borrowers affected by AI by assessing how AI could change each company’s revenue, costs, competitive position and ability to refinance, then monitoring those risks alongside loan protections and portfolio concentrations. “AI-dependent” is not a standardized borrower category: AI may threaten a company’s product, strengthen it, or affect it through reliance on third-party models or cloud infrastructure. Exposure is a reason to analyze a loan, not proof that the borrower has already been impaired.
What makes a borrower exposed to AI?
The relevant question is not simply whether a company is a software business. A lender needs to understand what the borrower sells and how AI might change the work its product performs, the alternatives available to customers, or the company’s own costs and capabilities. A software company could face new competition from AI tools while also using AI to make its product more useful or less expensive to deliver.
For credit analysis, the risk becomes important when those changes could weaken the cash flow available to repay debt, reduce the value of collateral, or make refinancing harder. The Bank for International Settlements (BIS) and J.P. Morgan Asset Management describe possible paths including revenue erosion, margin pressure, lower valuations and impaired refinancing access. Those are potential transmission channels, not evidence that every exposed borrower is experiencing losses.
How do funds assess an individual loan?
Managers connect the borrower’s business model to its capacity to pay. The following questions form a practical analytical checklist, not a universal regulator-mandated scorecard. The appropriate emphasis depends on the loan and the fund’s mandate.
Free tools Windows power users keep installed
One-click scans. No signup required.
Test the product’s resilience
- What does the product do? Identify the customer task it supports and whether AI could make that task easier to automate, replace or perform with a cheaper alternative.
- How difficult is it to switch? Consider customer retention, renewal patterns, implementation needs and the time or cost involved in adopting another product or building an alternative.
- Does AI weaken or strengthen the offering? Assess whether the borrower can incorporate AI into its product, and whether competitors could use AI to replicate its most valuable features.
- Can the company sustain its pricing? Examine how it charges customers and whether customers might reduce usage, negotiate lower prices or move to a different provider.
- What does the company depend on? Consider reliance on third-party models, cloud infrastructure or other shared technology providers, as well as the possibility that those dependencies create added costs or operational risk.
Translate business risk into repayment capacity
A weaker competitive position matters to a lender when it affects cash flow or the terms on which the borrower can borrow again. Relevant indicators include recurring revenue, customer retention and concentration, gross margins, operating costs, cash generation, leverage, interest coverage, covenant headroom, collateral value, sponsor support, debt maturity and access to refinancing. The Federal Reserve’s discussion of leverage and floating-rate borrowing is relevant because high debt burdens and changing interest costs can make a borrower less able to absorb an operating shock.
Managers also need to consider timing. A company can remain current on its loan while its valuation or refinancing prospects worsen. That gap matters most when the loan is approaching maturity before the borrower has time to adapt its business or restore cash generation.
Compare disruption and maturity timelines
J.P. Morgan Asset Management argues that the nature of a software exposure can matter more than its headline amount and that conventional fundamental metrics may not capture every disruption risk. In practice, a manager can compare the time needed for AI to affect the borrower’s business with the time left until debt maturity and the time likely needed to secure replacement financing. A longer runway may give a company more room to respond; it does not, by itself, establish that the loan is safe.
Oaktree Strategic Credit Fund’s March 31, 2026 shareholder update offers one example of a manager’s approach: it described a business-resilience framework combining operating KPIs, financial metrics and AI-related considerations. The update said pressure had been concentrated in older, pre-2022 vintages and ARR loans facing 2027–2028 maturities. That is a report about one manager’s portfolio and method, not evidence of a common industry standard.
What do funds monitor after making a loan?
Monitoring tests whether the assumptions behind a loan still hold. Managers can track operating performance and borrower developments alongside payment and loan data, then assess whether emerging changes affect liquidity, covenant compliance or refinancing prospects.
- Operating results: recurring revenue, retention, customer concentration, margins, costs and cash generation.
- Credit signals: liquidity, payment behavior, covenant tests and headroom, waivers, leverage and interest coverage.
- Value and refinancing: valuation changes, collateral value, debt maturity and the borrower’s ability to obtain financing.
- Business developments: changes in the product, customer demand, competitive alternatives and dependence on shared technology providers.
Funds can also aggregate exposure by sector, product type, sponsor, loan vintage, maturity, borrower and shared dependencies. This helps a manager see whether separate loans could be affected by the same technology shift or financing condition. BIS found that some large business development companies (BDCs) had exposure to a shared pool of borrowers. The Financial Stability Board (FSB) has warned that technology-sector concentration, interconnected financing, valuation opacity and limited loan-level information make it difficult to assess system-wide exposures.
How do loan protections help—and what can’t they do?
Seniority, collateral, covenant terms, reporting requirements and limits on additional debt can affect a lender’s recovery or its ability to respond when performance weakens. The protections are part of the analysis, not a substitute for evaluating whether the borrower can keep generating cash. No single covenant removes the risk that AI could change a company’s business model.
The Federal Reserve has cautioned that competition and pressure to deploy capital can weaken underwriting standards or encourage more covenant-lite lending. It has also highlighted the vulnerability that high leverage and floating-rate borrowing can create. These are credit risks that may compound an operating shock; they do not establish that AI has caused borrower distress.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
How big is the exposure, and what do the figures show?
Available figures describe different things: lending exposure, market-price movements and survey responses. They should not be treated as interchangeable measures of AI-caused credit losses.
| Measure | Reported figure | What it describes |
|---|---|---|
| BDC lending to software firms | About $115 billion | BIS reported this as about one fifth of BDC lending and more than 80% of BDC technology portfolios in 2026. It measures exposure, not borrower defaults. |
| Private-credit loans to SaaS firms | More than $500 billion at end-2025 | BIS reported this as 19% of total direct loans and said one third of private-credit funds had extended loans to the SaaS sector. It is a sector exposure estimate, not a measure of AI-driven impairment. |
| Software-company stock prices | Almost 30% decline from October 2025 to February 2026 | BIS reported a market-price movement. Over the same period, BDC stocks fell about 10% on average; BDCs with high software exposure underperformed those with low exposure by around 5 percentage points. These figures are not private-loan default rates. |
| Credit lines to private-credit funds | Around $220 billion drawn and undrawn | FSB’s 2026 available data across its member jurisdictions captured this amount. Some commercial estimates ranged from $270 billion to $500 billion. The figures indicate interconnection with banks, not direct AI-specific borrower exposure. |
In a July 2026 report, BIS said uncertainty about AI-related revenue had not yet affected the BDC software loans it studied or changed how BDCs and their equity investors priced those exposures. That finding sits alongside the software-market repricing and weaker share-price performance described above: market concern is not the same as realized loan impairment. The cited sources do not provide a statistic measuring how many private-credit borrowers have already suffered a deterioration in repayment capacity specifically because of AI.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Can AI tools help managers monitor borrowers?
AI tools can assist with repetitive, checkable tasks such as extracting terms from credit agreements, summarizing data rooms, comparing covenant definitions and flagging missing reports or exceptions. These uses can help organize information, but they do not replace judgment about whether a borrower’s business can adapt or repay its debt.
PwC’s 2026 survey page reported that 53% of respondents were more frequently implementing technology in private-credit investment processes, 54% were most likely to use AI in underwriting, and 16% viewed AI-enabled portfolio management as a current priority. These are survey responses, not adoption rates for the entire private-credit market. PwC also emphasized data quality, integrated workflows and governance, and said final economic judgment remains human.
Best Value
A CRISIL vendor-authored case study describes a U.S. fund using an LLM-based tool to review loan agreements and covenant data across about 100 active deals, identify exceptions and support borrower engagement. It illustrates a possible workflow; it is not independent proof of the tool’s performance or evidence that all funds use comparable systems.
What makes portfolio-wide exposure difficult to measure?
Private-credit loan data are less transparent than public-market data. The FSB has identified limited fund- and loan-level information, inconsistent definitions, valuation opacity and concentration as obstacles to assessing exposures and transmission channels. Public disclosures from an individual manager can show how that firm describes its own portfolio, but they should not be generalized to every private-credit fund.
For a comparison between loans or manager approaches, focus on the borrower’s exposure and ability to adapt; revenue and customer resilience; margin and cash-flow sensitivity; leverage and covenant headroom; seniority, collateral and other protections; time to maturity relative to plausible disruption and refinancing timelines; and concentration in related borrowers or shared dependencies. J.P. Morgan Asset Management’s loan-level disruption framework and the FSB’s discussion of concentration support those comparison axes, but the sources do not establish one validated industry-wide AI-disruption score.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
Do these 3 things before closing this tab:
1Scan for outdated or missing drivers - takes under a minute2Clear out junk files and repair common Windows errors3Fix the driver behind crashes, sound loss and screen glitches
