Every insurer I speak with this year has an AI initiative underway. Almost none of them can tell me, without checking, how long it actually takes a piece of data to move cleanly from one of their own systems to the next. That gap, not the AI itself, is what will decide who is still standing in this market a decade from now.
The Decade Will Not Be Decided by Who Adopts AI First
The current AI race in insurance is being fought on the wrong terrain. Every carrier wants to be seen adopting it. Almost none are being asked whether their core actually supports it. A model that produces a recommendation in half a second is not faster than the underwriter reading it, if that underwriter is still waiting on three other systems to agree on the same client’s details first.
This is the mistake the market is making at scale. Treating AI as the differentiator, when the differentiator was always whether the data underneath it moves cleanly. Two insurers can announce the same AI capability this year. Only one of them has a core built to actually carry it.
Where Underwriters and Brokers Actually Lose the Time
The scale of this is already documented. Aon’s Strategy and Technology Group found that underwriters spend more than two thirds of their time on data transformation before they ever reach a decision. Not analysis. Not judgment. Moving, reconciling, and correcting data that should have arrived usable in the first place.
An AI layer placed on top of that does not remove the two thirds. It just makes the remaining third faster, while the real bottleneck sits exactly where it always did.
This is the part of the market that adoption announcements never mention, because it is not a feature anyone can put in a press release. It is infrastructure, and infrastructure does not photograph well.
Outdated Technology Is Already the Deciding Vote
London Market brokers were asked directly what the single biggest impediment to modernisation actually is. Reliance on outdated technology ranked highest, ahead of every other barrier offered, according to Guidewire’s 2026 London Market Tech Barometer. The same survey found that 78 percent of brokers say insurer technology plays a decisive or highly significant role in where they choose to place risk.
That is not a satisfaction survey. That is a placement decision already happening. Brokers are not waiting for the next decade to arrive before they start selecting on integration quality.
The firms treating this as a future consideration are already a step behind the firms it is currently costing business.
The Two-Tier Market Is Already Forming
This is where the shape of the next decade becomes visible, not speculative. Reinsurance coverage of the market has begun describing the same split directly, pointing to data and automation, not underwriting appetite alone, as what will define the next era of the industry. The signal is consistent across both sides of the desk. Brokers naming integration as their real concern, and market commentary naming data infrastructure as the deciding factor for what comes next.
Two tiers are forming inside the same market. One tier where data moves once, correctly, and reaches the point of decision usable. One tier still paying someone, quietly, to make it look like it did. The AI headline looks identical from the outside. The operating cost behind it does not.
Most of the AI built for this industry was designed by technology teams first and applied to insurance second, which is precisely why it sits comfortably on top of a fragmented core instead of replacing it. Agiliux was built the other way round, by people who worked inside broking and reinsurance before writing a line of the platform, which is why the architecture assumes integration from the outset rather than retrofitting it once the cracks are already visible.
Key Takeaways
| Five things to retain from this article |
|---|
| 01 Underwriters spend more than two thirds of their time on data transformation before reaching a decision, not on judgment itself (Aon Strategy and Technology Group) |
| 02 Outdated technology ranks as the single biggest impediment to modernisation named by London Market brokers, ahead of every other barrier offered (Guidewire, 2026) |
| 03 78 percent of brokers say insurer technology is a decisive or highly significant factor in where they place risk, meaning integration already shapes placement decisions today (Guidewire, 2026) |
| 04 Market coverage of reinsurance is already describing data and automation, not underwriting appetite alone, as the deciding factor in the next era of the industry (Send Technology, Reinsurance News) |
| 05 A two-tier market is forming based on whether data moves once correctly or is reconciled by hand, and that split is visible in broker placement behaviour now, not a future prediction |
Frequently asked questions
Not on its own. Market data shows brokers already weighing insurer technology heavily in placement decisions, but the deciding factor named most often is integration quality and outdated core systems, not how quickly a firm announced an AI initiative.
Aon’s Strategy and Technology Group found that underwriters spend more than two thirds of their time on data transformation, reconciling and correcting information moving between systems that were not built to share it, before analysis or judgment ever comes into play.
It is already visible. London Market brokers currently name outdated technology as their single biggest modernisation barrier, and a majority say insurer technology already shapes where they place risk, which means the split is showing up in placement behaviour now.
A platform designed by people with direct broking and reinsurance experience builds integration into the architecture from the outset, rather than adding an AI layer on top of a legacy core and asking brokers or underwriters to reconcile the gap it leaves behind.
Glossary
| Key terms used in this article |
|---|
| Data transformation The manual or semi-manual work of converting, cleaning, and reconciling data as it moves between systems that were not built to share it |
| Core System The primary platform an insurer or broker runs policy administration, rating, or claims through |
| Bolt-on architecture Technology added on top of an existing core system rather than designed into it |
| Algorithmic underwriting Underwriting decisions made or assisted by automated rules or models rather than manual review alone |
| Two-speed market A market splitting into firms whose core systems support real-time data integration and firms still reconciling data by hand between systems |
| Practitioner-built Software designed by people with direct operational experience in the industry it serves, rather than generic technology adapted to fit it |
Conclusion
None of this is a prediction about which AI vendor wins. It is a statement about which firms will still be reconciling the same client data by hand in ten years, and which ones stopped needing to. The market is already telling us the answer through where brokers place risk today, not through what carriers say in their AI roadmaps.
The firms that define the next decade of broking will not be the ones with the most AI features announced in 2026. They will be the ones whose core stopped asking a broker or an underwriter to close the gap the system itself created.
Is your architecture built for the decade that is starting, or the one that is ending?
Agiliux was built by people who worked inside broking and reinsurance before writing a line of the platform, which is why integration sits in the foundation rather than being added once the cracks show.
Want to see what a platform built for that shift actually looks like for your book of business?
Sources cited
- Statistic: Aon Strategy and Technology Group, reinsurance data and analytics study, finding that underwriters spend more than two-thirds of their time on data transformation before making decisions. Aon Strategy and Technology Group
- Statistic: Guidewire Software, “London Market Tech Barometer 2026” (survey of 251 insurance brokers conducted by Censuswide, 19-27 November 2025), published via Business Wire, 19 February 2026. London Market Tech Barometer
- Finding: Send Technology, “Reinsurance 2.0,” Reinsurance News, January 2026. Send Technology
