Five years ago, after acquiring insurance technology, a private equity firm’s first move would be adapting it to cloud servers; yesteryear’s frontier advancement in business technology. That day-one-priority often dictates acquisition strategy and business outcomes.
Today it's almost entirely about artificial intelligence: 95.2% of all global insurtech funding in the first quarter of 2026 went to AI-focused companies, according to Gallagher Re's Q1 2026 Global InsurTech Report. That shift is reshaping which insurtech categories actually command a premium, according to Donald Light, principal at Donald Light Insurance Technology Advisory and former director in Celent's North America property/casualty insurance practice.
"What are (insurtech’s) doing today with AI, and how am I going to make them faster, smarter, better using AI going forward?" That question, he said, now dominates the due diligence private equity firms run before acquiring an insurance technology vendor – which matters just as much to insurers licensing that vendor's product, since its answer will become their roadmap by default.
Insurtech on offer to private equity breaks into a few recurring core systems, said Light. Policy administration, billing and rating tend to command the most value. Point solutions like claims fraud detection; or distribution solutions for MGA platforms and insurance marketplaces, are valuable largely for the market access they hand an acquirer.
Among these, the presence of AI is nearly ubiquitous and no longer distinguishes a target, said Light. Rather, its functionality is separated into three distinct forms buyers and licensers should ask about specifically: predictive AI, used mainly in underwriting and pricing to estimate outcomes and loss experience; generative AI, used for sorting and presenting written information; and agentic AI, the newest and most heavily promoted of the three. What separates a real acquisition candidate from an overhyped one, he said, is demand growth and proof.
"Is there a convincing case that can be made that I, as the buyer or the licensee insurance company, am going to see one of those levers – revenue, cost, experience – change for the better?" he said, framing the test as whether a technology can plausibly move a licencing carrier's pricing, underwriting, claims or service functions on one of those three axes for both staff and customers.
Buyers will press vendors for hard metrics: whether existing customers have shown substantially increased revenue or measurable cost savings, not just an impressive screen. Light also flagged the tiering question underwriters and dealmakers use to size a market: insurers are commonly grouped from tier one, above $5 billion in premium, down through tier five, covering the smallest carriers. A vendor selling well to tier two carriers, in the $1 billion to $5 billion range, still has to answer whether it can move down to smaller insurers, or up into tier one, before its growth story holds up as an acquisition case.
Light pointed to a feedback loop that explains why private equity firms move as a herd on technology trends. To raise the next fund, a firm has to demonstrate that its prior acquisitions succeeded – and "succeeded," in practice, means the acquired technology saw real uptake.
Cloud was the proof-point five years ago, when a portfolio company’s shift to cloud-based delivery translated into a visible uptick in adoption and sales that a firm could point to when courting its next round of institutional capital. “One of the largest agenda items for a private equity firm that acquired a technology provider - if it's not cloud-based to start out, it does things to make it more cloud-friendly,” said Light.
AI is filling that role now. Light cited arrangements between Vista Equity Partners and OpenAI: firms strike enterprise deals with AI providers, not just to improve one portfolio company, but to generate a repeatable proof point across the entire portfolio that can be shown to partners.
On timing, Light said insurance companies that do get involved as buyers (itself unusual, since most insurers are customers rather than acquirers of insurtech) tend to want visibility into a private equity firm's modernization plan before, not after, the deal closes. "There are no secrets about it," Light said. "If we acquire you, here's how you're going to look different in order to be more successful at the end of year one, year two, year three and so on." A plan alone isn't enough: "It's one thing to say we're going to use a lot more AI in underwriting. That's wonderful. But then they have to execute." Underperformance happens: "Could over-promising happen? Yes. Could less-than-projected quality outcomes happen? Yes," Light said.
For carriers evaluating a vendor – whether as a potential acquirer, a partner in a private equity deal, or simply a customer choosing between competing platforms – Light's framework offers a way to separate real commercial traction from AI-branded marketing: verified revenue and cost outcomes from existing licensees, evidence of tier mobility, a specific category (predictive, generative or agentic) rather than a vague AI label, and an executed roadmap rather than a slide describing one.
