Group benefits carriers are spending heavily on artificial intelligence, but the technology alone will not determine who comes out ahead. The competitive gap is forming around something less visible: the quality of the member-level data each carrier is able to put to work.
That argument has been made before, but it is gaining traction as AI tools become standardized across the market. When every carrier can deploy similar platforms, the question shifts from who has the technology to who has the data infrastructure to make it produce reliable outputs.
Group benefits carriers accumulate vast operational records across the full member lifecycle, including quotations, underwriting decisions, enrollments, benefit changes, claims, and member interactions. In principle, that information represents a proprietary asset no external data provider can replicate. In practice, much of it sits in legacy, policy-based administration systems that were not built to track individual members’ cover or experience over time, making it structurally difficult to use for AI-driven analysis.
The problem is not specific to one market. An AM Best survey of more than 150 rated insurers and managing general agents, published in April 2026, found that data readiness ranked as the single largest barrier to AI deployment in insurance, above security and privacy concerns and legacy system integration challenges. Sridhar Manyem, senior director of industry research and analytics at AM Best, noted in the report that AI systems produce unreliable outputs when data is fragmented across legacy systems, insufficiently governed, or lacking appropriate context. Carriers that had already invested in modernizing their data infrastructure, he added, were finding it measurably easier to integrate AI across their workflows.
That finding holds in the benefits space specifically. Research into how AI is reshaping insurance underwriting found that access to more information does not automatically produce better decisions, not unless the data flowing into an AI system is clean, connected, and contextualized. A July 2024 Bain & Company brief, Insurers Can Parlay Technology into a Competitive Edge, found that only 5 to 10 percent of carriers consistently capture value from their data and technology investments, a finding the AM Best survey suggests has not materially changed as AI adoption has accelerated.
“Over time, AI is likely to become increasingly available across the market and is unlikely to remain a core source of competitive advantage in its own right,” said Tim Stoves, group protection lead at Lumera. “The real differentiator will be the quality and completeness of providers’ proprietary member-level data, and how effectively they use it to create value, deliver better outcomes for customers and reinforce trust.”
Lumera is a Sweden-based insurtech that provides digital administration platforms for the life and pensions industry. The argument aligns with what the company sells, but it tracks independently with what carriers across the market have been finding as AI deployments have matured. The underlying point, however, is consistent with what carriers are finding independently as AI adoption matures. Faster processing only improves outcomes if the data feeding the decision is accurate. A carrier running a sophisticated AI platform on fragmented member records will produce faster outputs, rather than better ones.
The downstream effects reach plan sponsors and those advising them directly. Carriers with well-organized, longitudinal member data can build more precise risk assessments, identify claims trends earlier, and price group products with greater accuracy. Those running on siloed systems face a ceiling at each of those points, regardless of which AI platform they adopt. Data privacy and AI governance are drawing growing scrutiny from plan sponsors too, as AI becomes more deeply embedded in group benefits administration, giving the data infrastructure question a compliance dimension alongside the operational one.
The prescription Stoves outlined centers on how carriers collect, consolidate, and connect member information over time. “By connecting information on members’ cover, claims and interactions over time, insurers can build a richer picture of their customers and unlock greater value from AI,” he said.
Modern administration platforms can help consolidate records that have historically been scattered across disconnected systems. But the more fundamental shift is organizational: treating member data as a strategic asset rather than an operational byproduct. The challenges carriers face in extracting value from legacy technology are well documented, and solving them requires decisions about data architecture that precede any AI deployment.
The regulatory dimension adds pressure. In a closely regulated environment, AI-driven decisions in underwriting and claims need to be auditable and explainable. Carriers on fragmented data architectures face not only weaker AI outputs but growing governance risk as regulators pay closer attention to how automated decisions are made and on what basis.
“Without the right data foundations, even the most sophisticated AI capabilities will be limited in what they can deliver,” Stoves said.
That gap is increasingly visible to advisors evaluating carrier proposals. Pricing accuracy, claims turnaround, and product responsiveness all flow from the same upstream question - how well a carrier knows its members. As AI tools standardize across the market, those differences will become harder to attribute to technology and easier to trace back to data.
