What Insurtech Insights 2026 Revealed About the Real State of AI in Insurance

Insurtech Insights 2026 closed its doors on 19 March, but the conversations it sparked have continued long after. Several Camelot Consulting Experts were in London for the two‑day event, and rather than rush out a summary, we took the time to compare reflections, challenge one another’s impressions, and let the key ideas settle.

In these field notes, we explore what was said, what surfaced repeatedly, and what it means for an industry standing at a genuine inflection point.

A Conference of Scale, Defined by Substance

Held at the InterContinental O2, the conference brought together thousands of people across insurance, reinsurance, technology and beyond. More than 200 speakers featured across the agenda: from AI governance and the energy transition to quantum computing, talent, sustainability, geopolitics, and the future of distribution.

The tone was set early. Ericson Chan, Group Chief Information and Digital Officer at Zurich Insurance, opened with a line that framed almost every discussion that followed:

“AI is no longer a tech wave. It is becoming the Operating System of insurance.”

Keynotes came from digital and technology leaders at Swiss Re and Zurich, underwriting leaders from Munich Re Specialty and Vig Re, and even an astronaut from the European Space Agency. The scale was impressive. But scale wasn’t what defined the event.

What stood out was the honesty, with speakers repeatedly naming the obstacles, trade-offs and realities that technology alone cannot solve.

Two themes, in particular, cut through the noise.

Agentic AI Moves From Theory to Practice

Agentic AI wasn’t presented as a future concept. It was treated as something already operating inside a small number of insurers, and reshaping expectations in real time.

Simon Torrance’s session on the purple stage became one of the most talked-about of the conference. He described an AI maturity model that captures the state of play with uncomfortable clarity: most insurers sit at Level 1, working with co‑pilots, chatbots, and tools that provide answers but do not act. But Level 4, defined by agentic teams that collaborate in unstructured environments and solve complex problems autonomously, is already being built.

And the results are no longer marginal:

• Allianz Partners: straight‑through processing increased from 10% to 90%

• Qover: on track to automate 60% of claims decisions

  
  

Torrance’s most important point was not about automation or efficiency. It was about Intelligence Capital, the idea that when machine‑readable reasoning underpins decisions, organisational knowledge no longer disappears when people leave. It compounds. It strengthens. It becomes a competitive advantage that cannot be copied.

The message from the panel was unequivocal: Agentic AI is real, it is running, and waiting is now the most dangerous strategy available.

Scaling AI: Technology Is the Easy Part

If agentic AI was the conference’s momentum story, the second theme was the reality check. The industry talked openly about why so many AI initiatives stall. Across stages, languages and job titles, the explanation was consistent:

Scaling AI is not a technology problem. It is a people, process, and culture problem.

Pravina Ladva, Group Chief Digital and Technology Officer at Swiss Re, put it directly: the technology is the easy part. The difficulty lies in unpicking processes that have been operating for decades and asking teams to work in different ways.

Dr Fabian Winter of Munich Re reinforced that AI only succeeds when it is built in lockstep with the business. Machine learning on its own is insufficient; it must be fused with deep subject‑matter expertise to produce something that actually works.

Steven Zuanella, former Group Chief Digital Officer at Generali, offered perhaps the clearest signal of where strategies falter: when digital transformation remains technology‑driven rather than business‑driven, when data remains fragmented, and when AI objectives are not tied to leadership performance metrics.

As he observed:

"Piloting is easy. You can pilot without management commitment. Scaling is entirely different."

That commitment, or lack of it, will determine which insurers move ahead over the next three to five years.

A Necessary Conversation, Not a Comfortable One

Insurtech Insights 2026 was worth attending not because it delivered simple answers, but because the right people were willing to ask difficult, necessary questions. Honesty, not hype, defined this year’s discourse.

The industry has clarity on where it needs to go. The harder work (changing behaviours, reshaping processes, and committing at scale) is still ahead.

At Camelot Consulting, we recognise that gap between intention and impact. It is where we spend our time: navigating complexity, aligning ambition with delivery, and helping the industry move from experimentation to execution.