AI isn’t transforming reinsurance through a single revolutionary tool. It’s reshaping it through a series of small, practical shifts - especially in how insurers collect, process and act on data.
“AI is already changing the customer experience,” said Veronica Judice (pictured), global head of specialist claims solutions for Aon. “It’s increasing efficiencies, it’s offering quicker insights into claim trends, claim strengths and that also helps improve risk assessment and modeling.”
For Judice, claims are not the tail end of the process - they are the product.
“From a business perspective, claims is the product of insurance. It's what clients buy. They don't buy the policy. They buy the likelihood of a claim,” she said. “They do that for various reasons. It could be for daily cash flow protections or cat event response.”
That shift in focus - from policy to performance - is shaping how AI is applied. Generative tools may be dominating headlines, but the most effective AI applications in insurance are less about novelty and more about utility.
“There seems to be this perception that there will be, at some point, this single AI tool that magically will change everything... improves claims handling quality and makes underwriting perfect,” Judice said. “The truth is very different.”
The real progress, Judice argued, is coming from integrating a range of tools into existing processes - especially tools that help structure data, uncover patterns and support faster decisions across the value chain.
“There’s not going to be one tool that will bring revolution to organizations,” she said. “But there are going to be various tools... that will need to be incorporated in a process and ensure appropriate data extraction, appropriate data built, and then data extraction and automation.”
Most of this work is happening well before claims are filed. AI is being embedded across risk modeling, underwriting, and early-stage decision-making - areas where unstructured data has historically slowed things down.
Judice said the benefit of AI is its ability to “drive correlations from unstructured data” and “accelerate the way we process data.” But these gains aren’t just technical - they’re also cultural.
“AI is everything from algorithms, basic algorithmic functions, all the way through machine learning and very recently, deep learning,” she said. “Generative AI adoption in the industry is in its infancy. I think that’s fair to say.”
Technology is advancing quickly, but the bottleneck is no longer legacy systems - it’s people and mindset.
“They don’t always realize it, but firms often already have the tools and people in place to gain many AI efficiencies, and just need some guidance around potential AI use-cases and implementation,” Judice said. “However, in some areas a significant advancement in technology is required, and the key to integration is having a talent pool and a mindset that evolves with it.”
For AI to succeed more broadly, insurers must build cross-disciplinary teams fluent in both insurance and technology.
“You will need people with technical skills. You will need data scientists and technology and software engineers,” she said. “You will need people able to understand and respond to emergent risks. And you will need them to maintain high levels of client empathy in an ever more digitalized world.”
The pressure is on organizations to be intentional in how they develop and retain talent across underwriting, claims, operations and IT.
“You're going to have to have people that understand the business, that understand property insurance, for example, that understand the overall business of strategic retention and transfer of risk, but that they're also technology literate,” she said.
Judice pushed back against the idea that AI introduces opacity into insurance decisions. Instead, she argued, the right implementation of AI supports transparency and regulatory compliance.
“Transparency is actually the underlying factor and the main motivator for digital transformation in insurance,” she said.
“AI is enhancing decision making thanks to the ability to draw from bigger, more sophisticated data. But it is data that exists,” she said. “It allows to reorganize it, to keep it in a more structured manner, which also allows to retrieve it more easily and to ensure the transparency that is key.”
That transparency, she added, is essential in a highly regulated industry where technology must align with solvency and consumer protection mandates.
Judice also pointed to the need for proper oversight and governance - not just automation for efficiency’s sake. “There are - checks and balances that are kept in place, and they are preserved by talent and the appropriate talent pools,” she said.
“Technology is only as good as the people operating it,” Judice said. “The computer itself is not making the decision. The computer generates a decision based on the data that it has access to and based on the instructions to process that data that's been given.”
She stressed that the organizations best positioned for the future will be those that recognize the need for a “holistic approach” to tech adoption - one where business strategy, client expectations, and operational capabilities evolve together.
“An intentional approach to recognize the challenge and to ensure that technology advancement and talent pool evolve hand in hand,” she said.
Judice pointed to the work of Aon’s Strategy and Technology Group, which is supporting clients as they build for future volatility and long-term resilience. But the message was broader: the advantage isn’t in the tools themselves, but in how they’re used.
“You need to match your level of innovation to the level of understanding and expectation of your clients,” she said. “There’s no shortcut for that.”
