As an industry, insurance exists to mitigate risk. And the risks currently faced by insurers are unprecedented. From more frequent natural catastrophes to overlapping regulations and AI-enabled fraud, insurers are facing challenges that their models, business processes, and systems were never designed to address.
Insurers have a quiet history of innovation. In some parts of the industry, sophisticated AI helps determine and refine pricing. Auto insurers use telematics to track willing drivers’ behaviours and customise premiums. Embedded insurance is expected to become a $70 billion market by 2030, compared with about $13 billion today. Parametric insurance is also growing quickly - especially in Asia, where it’s finding promise as a buffer against the financial consequences of extreme weather events.
The next wave of insurance innovation will be catalysed by agentic AI – a paradigm shift in the way insurers do business. Agents can generate contextual content and work intelligently across multiple complex, variable workflows to dynamically learn and improve outcomes. Agentic AI brings insurers the chance to become more nimble, enabling the faster rollout of new products and greatly improved customer experiences. It holds the potential for great efficiencies and more creative use of an organisation’s human capital. And it can speed the data integration that makes all this possible. Together, these and other capabilities will create the AI-native insurer of the future – one that is hyper-efficient, insight-driven, and agentic by design.
Of course, no insurer becomes AI-native overnight. It’s a journey that begins with these five strategies:
1. Start with the end in mind Define AI-native for your organisation, and map a journey toward that future
The journey to becoming AI-native begins with an outcome, not a technology. Agree on the business outcome you’d like to reach, and make sure it’s in line with your organisation’s North Star vision.
Then, build a pathway to that outcome. Once you’ve mapped it out, look for the initiatives that will bring maximum value with minimum time and risk. Be willing to consider automation, traditional AI, and agents, and understand that some tasks may need to be accomplished with technology you’ve yet to co-create. Get aligned with your governance and security teams, and start proving out the value bit by bit.
2. Transformation must become business as usual Focus on fast, incremental innovation that delivers quick value
Transformation projects no longer need to be drawn-out, multi-year marathons. Instead, select use cases that allow you to demonstrate value quickly, building credibility, ROI, and permission to continue forward.
A good early use case simplifies a complex process. If the use case is too small, it may qualify as a success without driving significant value. If it’s overly big and complex, save it for later, after your organisation has built up the expertise and culture to make it work.
3. Modernise your existing technology ecosystem with AI Use AI as an orchestration layer to harmonise your current IT environment
Agentic AI provides a fast and efficient option for modernising technology environments. Use agentic AI to read code written in older languages and then create APIs that serve as a bridge between the old application and a modern interface.
Agents also allow enterprises to be more agnostic about where their data sits. If agents know what data is available and where it’s stored, and if those sources are trusted, then the data can be extracted and exposed to agents in a modern and scalable way.
In the case of determining Scope 3 emissions, for example, agentic enablement can greatly streamline, accelerate and automate the data collection process. It can bring end-to-end visibility of Scope 3 impacts throughout the value chain, and provide the insight necessary to amplify what’s working and make changes where necessary.
4. Build transformational resilience Ensure innovations are introduced from a foundation of strong regulatory, compliance, and security capabilities
Agentic innovation places new demands on regulatory, compliance, and security functions.
For regulatory and compliance teams, the first impulse may be to introduce a slew of controls specifically for agents. But that doesn’t scale when insurers are operating in multiple jurisdictions around the globe. Instead, organisations need to make a shift in mindset. It’s not about rules. It’s about providing secure guardrails, and giving agents and people the freedom to work and innovate within those boundaries.
Security needs to be front-and-centre in any agentic innovation, especially as quantum computing moves closer to reality. In the case of a data breach, you may find that some of your agents stop performing. More worrying, and extremely difficult to detect: A subtle interference with your data that manipulates your agents, changing the trajectory of how the business makes key decisions.
5. Adopt AI-fluent leadership
Rethink your operating model to support AI-native ways of working, including upskilling and evolving your team’s roles
At many enterprises, the cultural hurdles to becoming AI-native are just as high as the technical ones. If your teams aren’t using the AI tools already available to them – perhaps Co-Pilot or Gemini or Chat GPT – they may not be eager to adopt the more targeted technologies you’re planning to roll out.
Leaders need to change too, because AI will affect their business more profoundly than they may expect. They’ll need to learn to use AI themselves, of course. They’ll also have to get comfortable being a bit more in the weeds, without lapsing into micromanagement. To reimagine how the business operates, leaders will need to understand their business at a finer level of detail: which areas are working well, which are stuck, and who’s best-suited to make fixes.
Agentic AI is evolving quickly, and for most insurers, the journey to becoming AI-native will take some time. The good news is that it’s possible to make significant progress quite quickly, showing real value and creating the momentum to continue.
We thank Kyndryl for sharing practical ways insurers can start their AI‑native journey to build resilience. As a proud Platinum Sponsor of our 2026 Annual Conference, we’re delighted to be partnering with them. Join us on 3 February at the QEII centre and connect with the Kyndryl team who will be sharing actionable strategies and real-world examples at Stand 7.