Insurance has already solved the problem it spent the last decade obsessing over: proving that new technology can work. The harder problem, according to senior leaders at Lloyd's and The Hartford, is scaling it - turning a strong pilot into something embedded across an entire market. That shift, from proof of concept to proof of adoption, is what's actually driving the next phase of insurtech.
Insurers are increasingly looking beyond traditional risk transfer as new technology gives carriers a greater ability to identify hazards earlier and help policyholders prevent losses. The shift is being driven by growing catastrophe exposure, widening protection gaps, and advances in data analytics, artificial intelligence, and environmental monitoring. But insurers have also become far more selective about which technologies they adopt, and that selectivity - not the pace of invention - is now the defining constraint on the industry's innovation cycle.
Matt Scott, The Hartford's head of property and casualty innovation and risk services, said risk mitigation represents one of the clearest opportunities for insurers to deliver additional value to customers. "Today we have technologies, data sources and analytics where we can identify risks earlier, provide insights sooner and work with them to prevent or reduce losses before they occur," Scott said. "That's a better outcome for our customers." Businesses are increasingly looking for insurance partners that can help them strengthen resilience, he added, making prevention a more meaningful part of the carrier-policyholder relationship.
That prevention-first approach is particularly relevant as insurers confront complex and interconnected risks including wildfire, flood, cyberattacks, and infrastructure deterioration. New technologies are giving underwriters access to more detailed, localised, and potentially real-time information, supplementing the broader view traditionally provided by catastrophe models - not replacing it.
Dawn Miller, CEO of Lloyd's Americas and chief commercial officer at Lloyd's, said the growing use of more granular information should not be viewed as a rejection of established catastrophe modelling. "I do not believe it's a shift away from catastrophe modelling, but a recognition that insurers - in the face of an increasingly complex risk environment - need more granular insights alongside traditional models," Miller said. This is particularly relevant in the US, she said, where "the frequency and severity of natural catastrophes are driving demand for more actionable, real-time insights at a local level." "Greater insight equals more impactful outcomes to address protection gaps," she added.
That layering approach - adding granular data where it earns its place, rather than discarding what already works - illustrates just how deliberate the industry's technology adoption has become. It is precisely this discipline that raises the bar for what any new insurtech solution now has to clear.
Scott said The Hartford assesses insurtech solutions according to whether they produce measurable improvements in loss prevention, safety, and resilience. Structured pilots allow the insurer to test products in real-world settings before considering wider implementation. "Just as important, the technology must fit into the workflows of customers, brokers and carriers," Scott said. "Innovation only creates value when it's practical enough to be adopted at scale."
That focus on integration is becoming increasingly important as insurers test AI across claims, underwriting, delegated authority oversight, and operational functions. The technology may accelerate processes and improve accuracy, but it must still address an identifiable business problem and work within existing systems - the same adoption filter Scott applies to any pilot.
The Hartford's collaboration with Lloyd's Lab has reinforced the importance of combining insurance knowledge with specialist technology expertise, Scott said. The Hartford has supported Lloyd's Lab through Syndicate 1221 since 2020, providing 11 mentors to 15 accelerator teams.
Lloyd's Lab has also built substantial links with the US technology sector. Forty-seven US startups have participated in its accelerator programme, collectively raising more than $600 million to tackle exposures including hurricanes, wildfires, cyber threats, and AI-related risks. Miller said startups across the latest cohort are developing tools to assess flood exposure, monitor infrastructure deterioration, and identify environmental hazards, while cyber-focused companies are helping insurers detect vulnerabilities earlier and others are developing tools intended to make complex placements more efficient.
That volume of capital and activity might read as evidence the innovation problem is solved. Miller's own assessment says otherwise. "For many of these companies, the challenge will shift to scaling adoption," she said. "Several are already gaining commercial traction, and the next step is helping those solutions become embedded more widely across the market." That will require continued collaboration between technology companies, insurers, brokers, capacity providers, and investors, along with proof that these products can deliver consistent results beyond a controlled pilot environment.
"Ultimately, innovation only creates impact when it is tested against real market needs and adopted at scale," Miller said - the line that, more than any other in this piece, defines what the next chapter of insurtech actually has to answer.
