Balancing talent and technology in London Market claims

How claims organisations can balance talent and technology for long-term success

In October 2025, we hosted a discussion on balancing talent and technology in London market claims.

At the time, the framing still felt like a live question. Would technology reduce the need for claims talent? Could AI and automation take work away from claims professionals? How should firms balance investment in people with investment in systems?

Less than a year later, the conversation has already moved on.

The more useful question is no longer whether it is talent or technology. It is how claims teams combine both to improve client service, reduce friction and prepare for a more data-driven market.

One participant put it simply: “This is a false choice.”

That feels even more true now.

Claims transformation is not a technology problem

The most interesting point from the discussion was that the biggest barrier to transformation is rarely the technology itself.

Legacy systems, non-linear processes, inconsistent data and lack of APIs all matter. They slow things down. They create workarounds. They make consistency difficult across classes, systems and market participants.

But the harder challenge is people, ownership and culture.

Claims professionals need time, skills and support to make change work. Your experienced claims experts need to be close to transformation projects because the detail that matters is often hidden in the small decisions, exceptions and judgement calls that never appear in a process map.

As another participant said: “Technology transformation isn’t just about technology, it’s about ownership.”

That is the point many transformation programmes still miss. You can buy a tool, appoint a project team and build a roadmap. But unless the claims function owns the business problem and the change behind it, the risk is that technology becomes another layer of process rather than a better way of working.

Digital literacy is becoming a claims skill

 One of the clearest shifts is that digital capability can no longer sit in the IT team alone.

Claims professionals do not need to become coders. But they do need to understand how data, workflow, AI outputs and automation affect their decisions.

The phrase that stood out from the discussion was that claims professionals need to become “data translators”.

That matters because claims data is only useful when someone understands the context behind it. A system may classify, extract or summarise information. But in specialty claims, the difference between a plausible answer and the right answer often depends on technical judgement.

This is particularly important as AI is used to analyse unstructured data, classify causes of loss and extract insight from reports, messages, documents and claims histories.

The promise is significant. Claims information that was previously buried in files can start to inform underwriting, broking, risk mitigation and client conversations.

But AI does not understand the market unless the market teaches it.

The future is augmented claims, not automated claims

The strongest examples from the discussion were not about replacing claims handlers. They were about giving claims teams better tools.

One example focused on smarter segmentation and triage. Work that had taken people around 20 minutes could be reduced dramatically, freeing claims handlers to focus on judgement, service and problem solving.

Another example described a virtual assistant approach: pulling together key information from different systems, applying red, amber and green indicators and giving claims professionals a clearer view of the file.

That is a powerful direction of travel. Not AI as a decision-maker. AI as support.

The phrase from the room that captured it best was that AI should not be “autopilot for the claims handler, but the sat nav.”

That is exactly the distinction the market needs.

Autopilot removes control. Sat nav provides guidance, warnings and options. The human still decides whether the route makes sense.

For London market claims, that difference is critical. Complex claims involve judgement, negotiation, policy interpretation, client communication, empathy and commercial context. Those are not peripheral parts of the job. They are the job.

Senior expertise is needed more, not less

One of the more important lessons from the discussion was that experienced claims professionals are essential to making AI work.

In one example, AI was used to classify causes of loss from unstructured claims material. The model produced outputs that looked plausible. Less experienced users could have accepted them. But senior claims specialists were able to identify which outputs were right, which were wrong and which needed more investigation.

That is where the risk sits.

AI can produce answers that look convincing. But claims teams need enough expertise in the loop to challenge those answers and refine the model.

This changes the talent conversation.

It is not just about attracting the next generation. It is also about capturing the judgement of experienced claims professionals before it leaves the market.

One speaker described this as a “call to arms”: the current generation has to teach the machines. That may sound dramatic, but the point is serious. If claims expertise is not built into AI systems now, the market risks automating shallow understanding.

Time to think is not wasted time

Perhaps the most practical insight was about capacity.

Claims teams are often resourced around file loads, transactions and processing. That makes sense if the role is simply to keep work moving. It makes less sense if claims professionals are also expected to improve processes, support clients, work with data teams, train AI models, develop insight and contribute to underwriting strategy.

If claims teams are fully loaded with process work, they do not have the space to improve the system.

The discussion explored the idea of deliberately creating capacity for learning, innovation, client conversations and transformation work. That matters for junior team members as much as senior specialists. It creates space for shadowing, questioning, exposure and confidence building.

This is an uncomfortable issue because capacity costs money. But the alternative is also expensive: transformation programmes that rely on already stretched teams, senior experts who cannot spare time to validate outputs and junior colleagues who are expected to adapt without proper training.

 Culture change is harder than technology change

The conversation kept coming back to trust.

Traditional claims professionals may fear losing control. They may worry that technology will take the claim away from them. They may not have had consistent training on AI, Copilot or other digital tools.

That cannot be solved by telling people to adopt the technology.

People need to understand how tools work, where they help, where they are limited and how decisions remain accountable. Senior claims handlers need to be involved early so they can challenge the design and build confidence across the wider team.

One line from the discussion sums it up: “Culture change is so much harder than technology change.”

That is why the best technology projects are not really technology projects. They are business change projects.

---

What has changed since October 2025?

In a very short period, the market has moved from asking how to balance talent and technology to asking how to redesign claims roles around both.

AI adoption is accelerating. Digital workflows are becoming more realistic. Data extraction from unstructured material is becoming more practical. Claims insight is moving closer to underwriting, broking and client strategy.

But the core lesson from last year’s discussion still holds.

Technology will not transform claims on its own. People will.

The claims organisations that make the most progress will be those that give their people the skills, time and confidence to shape the tools they use. They will automate the work that creates friction, but protect the judgement that creates trust.

That is the real balance.

Not tech versus talent.

Talent, supported by technology, focused on better claims outcomes.