Private credit investing in the age of AI

ARTIFICIAL intelligence has moved from concept to commercial reality at remarkable speed. For investors, the question is no longer whether AI will reshape industries, but how to position portfolios for its effects – and where the likely winners and losers may emerge.

In equity markets, much of the attention has focused on identifying companies that can harness AI to accelerate growth.

For credit investors – and for wealth clients allocating to private credit – the perspective is necessarily different. The primary concern is not who grows fastest, but which business models can sustain cash flows if competitive dynamics shift. When lending capital, resilience matters more than excitement.

AI is already prompting a reassessment of risk across sectors. The debate extends well beyond traditional software developers. Legal technology, consulting, insurance brokerage, real estate services and online comparison platforms are all being evaluated through a new lens.

In some cases, market volatility reflects uncertainty rather than structural decline. Even so, assumptions that once appeared stable now require a closer look.

Private credit sits at the centre of this shift. Over the past decade, technology-enabled and software businesses attracted significant private-equity investment. Strong growth, high share of recurring revenues, scalable platforms and high margins supported elevated valuation multiples.

Many transactions were structured with leverage of five to seven times Ebitda (earnings before interest, tax, depreciation and amortisation), and occasionally higher when revenue-based metrics such as annual recurring revenue were used to support additional borrowing.

That model was sustainable in a period of abundant liquidity and steady growth. AI introduces a fresh variable.

The risk is not that software demand disappears. Rather, if new technologies lower barriers to entry, compress pricing or shorten product life cycles, revenue growth and margins in certain niches may soften. When leverage is high and equity valuations are full, even modest changes in outlook can reduce equity cushions and complicate refinancing.

SEE ALSO

Distressed-debt funds target private credit downturn as ‘greatest opportunity’ since 2008

Insurers’ search for yield must not compromise policyholder protection, says MAS assistant MD

Nvidia’s price-to-earnings ratio sinks to 7-year low as war and AI angst weigh

Not all disruption, however, is equal. One of the most important distinctions for credit investors lies between software that is deeply embedded in mission-critical operations and products that perform more standardised tasks with limited differentiation.

Enterprise resource planning systems, billing platforms, financial management tools and public-sector administration software are often tightly integrated into daily workflows.

They connect multiple datasets, support regulatory compliance and underpin reporting functions. Replacing them can be costly, time-consuming and operationally disruptive. In these areas, AI is more likely to enhance existing capabilities than to trigger wholesale displacement.

By contrast, software primarily designed to automate routine human tasks, such as basic document processing, commoditised marketing tools or certain customer-service applications, may face more sustained competitive pressure.

As AI improves baseline functionality and reduces development costs, pricing power can erode. For lenders, understanding where a borrower sits along this spectrum is central to assessing long-term resilience.

The economics of AI itself adds another layer of complexity. Developing and operating sophisticated models remains capital intensive. Many providers are still refining monetisation strategies, and profit pools are evolving.

Rapid innovation can shorten product cycles and introduce uncertainty around long-term returns. For credit investors, that uncertainty must be factored into cash-flow projections and leverage tolerance.

The ripple effects extend beyond operating companies. AI’s expansion has driven heavy investment in data centres and specialised computing infrastructure. While such assets are critical to the ecosystem, technological advances in chip design and computing architecture may accelerate obsolescence in ways not typically associated with traditional infrastructure sectors. For lenders with long-dated exposures, asset durability cannot be assumed.

Against this backdrop, underwriting discipline becomes even more important. For private credit allocations within wealth portfolios, leverage should be anchored to realistic and sustainable cash-flow expectations rather than ambitious growth projections. Structures built on optimistic assumptions leave little margin for error if competitive dynamics shift.

Equity buffers matter as well. Meaningful sponsor capital at risk can provide protection during periods of volatility and support alignment through refinancing cycles. Diversification also plays a role: Concentrated exposure to a single subsector undergoing rapid technological change can amplify downside risk.

Qualitative judgment is equally critical. How central is the product to customers’ daily operations? What are the switching costs – financial and operational? Is management actively integrating AI tools to strengthen its offering, or is the business vulnerable to displacement? These questions increasingly sit at the heart of credit analysis.

It is also important to recognise that AI can strengthen credit profiles. Many companies are deploying AI to streamline internal processes, improve efficiency and enhance customer engagement. When productivity gains translate into stronger margins and more predictable free cash flow, debt-servicing capacity can improve. The challenge lies in distinguishing structural benefit from short-term enthusiasm.

For wealth investors, private credit continues to offer attractive income and diversification characteristics. Yet the asset class does not exist outside broader technological shifts. As AI reshapes competitive landscapes, familiar indicators of stability – recurring revenues, historical growth trends and strong margins – require deeper examination.

For credit investors, the objective is not to forecast technological winners, but to understand how innovation interacts with leverage, capital structure and cash-flow durability. In periods of rapid change, careful structuring, granular and diversified portfolios, moderate leverage and disciplined analysis provide the most reliable foundation for consistent returns.

Gianpaolo Pellegrini is co-head of parallel lending; Andrew Tan is Apac chief executive officer and head of Apac private credit, Muzinich & Co