As exclusively revealed to Insurtech Insights, Munich Re Ventures, the venture arm of global
reinsurer Munich Re, has invested in Mechanical Orchard, the company with a unique approach
to legacy modernization.

Ben Bergsma, Director at Munich Re Ventures and Rob Mee, CEO, Mechanical Orchard, weigh
in together on the state of play.

Can you explain the problem of legacy IT in the insurance sector?

Ben: Legacy IT systems are the elephant in the room for a lot of large insurers, with the industry
facing modernization challenges due to its six-decade entrenchment with mainframe
technology. 70% of insurers’ IT budgets go towards maintaining legacy IT infrastructure.

Any disruption risks affecting customer operations, with significant regulatory consequences.
For insurers that handle sensitive data across complex processes, the modernization stakes are
exceptionally high. And maintaining stability, consistency, and compliance remains paramount.

Rob: It’s estimated 71% of Fortune 500 companies still house mission-critical applications on
this aging mainframe infrastructure. With such an impressive tenure, you can’t doubt the fact
that they’ve proved sturdy and reliable. But sturdy and reliable aren’t enough in today’s world.

Why is it so risky to keep legacy IT systems in place?

Ben: Like most industries, insurers are facing pressure to innovate in an AI-driven environment.
Mainframe reliance hinders adoption of open-source technology and generative AI, which are
rapidly becoming competitive necessities.

Research shows modern IT platforms can improve customer experience and accelerate product
launches by 3-5x. As market expectations shift toward more personalized and sophisticated
insurance products, providers must modernize core systems to remain competitive.

It might be an uncomfortable boardroom conversation, but the truth is many future, tech-enabled
visions will remain unrealized if we carry on as we are.

What’s the problem with other modernization approaches?

Rob: Most insurers know they need to migrate to more agile cloud-based environments. But
modernization has always been a high-stakes game. For insurers, any disruption to customer
operations or problems while handling sensitive data can have significant regulatory consequences. When you hear statistics like “79% of all modernization projects fail,” it’s easy to see why there’s hesitancy.

The most common approaches over the past two decades—emulation, “lift and shift,” and code
translation—have resulted in systems that, while technically migrated to the cloud, often remain
cumbersome and difficult to maintain. As a result, these systems struggle to capitalize on key
cloud benefits such as greater agility, reduced cost of change, and elastic pricing.

Now we’re seeing a rush of options out there promising that AI can “solve” for the difficulty,
complexity, and slow pace of modernization. But they fundamentally still focus on code first,
then address the integrations and context and regression testing later, creating huge levels of
uncertainty. The long pole in the tent isn’t getting to modern, understandable code, it’s getting to
modern, understandable code that doesn’t break what’s already working, that’s ready to go into
production as soon as it’s generated.

Tell us about this new approach

Rob: Our approach looks at this from a more holistic, systems-thinking perspective. We built a
platform called Imogen that looks at the inputs and outputs of a particular mainframe system
component. These data flows form the specification, the model, of how the rewritten code needs
to behave. Imogen then uses this model as a test harness, using generative AI to converge on
the precise code that creates exactly the same behavior.

Then we put the new code for this component into production, orchestrating it with the rest of
the application that’s still running on the legacy system. This means you know it’s working, and
you can rewrite the rest of the system, piece by piece, without disrupting operations. And once
you have something in modern code, changing and maintaining it becomes far easier and less
expensive.

What does this unlock for insurers?

Ben: By moving applications from the mainframe to the cloud while maintaining operational
continuity, large and complex incumbent organizations can accelerate digital transformation
without business disruption. This provides an improved foundation for security, innovation, and
competitiveness, creating high-quality, maintainable code that supports ongoing innovation.

As investors representing the strategic interests of the insurance industry, we believe this can’t
come soon enough.