Monroe Capital's Ted Koenig says AI will not lift every software company equally. The real winners will be the firms using rich data and automation to improve specific, high-value workflows.
Ted Koenig, chairman and CEO of Monroe Capital, told Bloomberg that AI will make a select group of software businesses meaningfully better, but not every category will benefit in the same way. That distinction matters because Monroe, a major private credit investor, is paid to look past the excitement and underwrite the durability of future cash flows.
The argument cuts against the broad AI story that has pushed valuations higher across software. As Bloomberg's recent coverage of Koenig's comments made clear, the opportunity is not simply adding an AI label to an existing product. The stronger case is in software that owns useful data, sits inside an essential workflow, and can automate decisions customers already need to make.
That is a narrower market than the current AI boom sometimes suggests. CRM platforms with deep customer histories, legal tools that can process large bodies of case material, healthcare software connected to patient records, and insurance systems built around claims data all have a clearer path to value. In those areas, AI can reduce manual work, improve speed, and make a product harder to replace.
Monroe's view is pragmatic because lending leaves less room for narrative. A credit investor is not mainly betting on a splashy demo or the next funding round. It is looking for recurring revenue, customer retention, margins, and enough resilience to survive a tougher refinancing market. That makes AI useful only when it strengthens the economics of the business.
The weaker version of the trend is easy to spot. A chatbot added to a dashboard, a generic search box, or a thin summarization feature may improve the user experience, but it does not automatically create a moat. If rivals can copy the feature quickly, the value flows to customers and large platform providers rather than to the software company selling the tool.
That is why Koenig's point lands beyond private credit. Investors across growth equity, venture capital, and leveraged finance are sorting software companies into two groups: those using AI to replace or improve expensive workflows, and those using it as a sales message. The first group may earn better retention and pricing power. The second will face harder questions.
Implications for Founders
For founders, raising capital on a broad AI narrative is getting harder. Lenders and equity investors want evidence that AI is improving the business in measurable ways, whether through higher revenue per customer, lower churn, faster implementation, better gross margins, or stronger product usage. Claims are cheap. Operating proof is not.
That favors vertical SaaS companies with data advantages. Legal tech, healthcare software, financial compliance, insurance, logistics, and supply chain tools all handle specialized information that customers cannot easily move or recreate. When AI is trained on that kind of proprietary context and embedded into daily work, the product becomes more valuable than a generic assistant.
Horizontal tools have a tougher road unless they already control a large platform or unique dataset. GitHub can improve coding workflows because it sits close to the developer and the codebase. A smaller company offering a general AI helper has to show why customers will keep paying once similar features appear inside Microsoft, Google, Salesforce, or other incumbent platforms.
The credit market also changes the timeline. Private lenders often live with a company for years, so they care less about the first wave of excitement and more about what happens after the feature becomes normal. If AI compresses prices, weakens switching costs, or reduces demand for seats, the same technology that helped sell the deal can become a refinancing risk.
That does not mean investors are turning away from AI. It means the bar is moving from exposure to execution. Software companies that can show real workflow lock-in, defensible data, and clear customer outcomes will still attract capital. Those relying on vague automation promises will find the market less patient.
Koenig's message is a useful signal for builders because private credit usually sounds cautious before public markets do. The next phase of AI software will be less about who can mention the technology most loudly and more about who can prove it changes the economics of a specific job. Narrow focus may end up being the strongest advantage.
Also read: GitHub Copilot moves to token billing in June and developers face a reckoning • Comedian poisons AI datasets with adversarial jokes to corrupt model humor • Asimov opensources v1 humanoid to kickstart robot dev ecosystem
