By Dr Lollie Mancey, Head of Innovation at Instech.ie
Insurance has always existed to help society navigate uncertainty. Today, that uncertainty is changing faster than ever.
Climate change, cyber threats, global disruption, and emerging technologies are creating risks that challenge traditional approaches to prediction and decision making. According to the Swiss Re Institute, insured losses from natural catastrophes exceeded US $100 billion in 2025, while the Allianz Risk Barometer says cyber risk ranks as the top concern for businesses globally this year.
The insurance industry has always adapted to new forms of risk. The challenge now is whether organisations can build the capability to adapt continuously.
The uncomfortable truth is that AI will not transform insurance if it simply automates existing processes. It will only transform insurance if it changes how organisations think, make decisions, and learn.
Too many AI strategies begin with technology and end with efficiency. But insurance has never competed on processing speed alone. It competes on judgement, trust, and resilience. Unless we rethink how decisions are made, how expertise is developed, and how people collaborate with intelligent systems, we will simply become more efficient at doing yesterday’s work.
The organisations that succeed will not necessarily have the most AI. They will have the cultures that adapt the fastest.
The industry is rightly exploring how AI can improve efficiency, enhance customer experiences, and transform operations. McKinsey & Company estimates that generative AI could generate USD $50-70 billion in additional insurance industry revenue, with the greatest impact expected across customer operations, software engineering, and marketing and sales. However, those gains will depend on organisations redesigning how work is done, not simply adding new technology on top of existing processes.
Every claim arrives at a deeply human moment: a flood, a diagnosis, a cyber attack, a business interruption, the loss of a home, or a loved one. Insurance is built on trust during moments of vulnerability.
As an anthropologist, I see insurance not simply as a financial service but as one of society’s great trust infrastructures. AI therefore is not just changing operations. It is changing one of the social institutions people depend on when life becomes uncertain.
Many organisations are asking: “Where can we deploy AI?”
The better question is: “Where does AI genuinely improve human decision-making?”
AI is not another digital transformation project. It changes organisational behaviour: how expertise develops, how knowledge flows, how teams collaborate, and ultimately what leadership looks like. The limiting factor is not the algorithms; it is organisational confidence.
Leaders now have to guide organisations through uncertainty while accepting that they do not yet have all the answers. The future belongs to leaders who create environments where people can experiment, learn, and build confidence alongside capability.
Technology adoption follows trust, not the other way around. Historically, we trusted experienced professionals because we understood their expertise. Increasingly, we will be asked to trust systems whose reasoning we may never fully understand.
That means insurers need to be transparent about where AI is used, how decisions are governed, and where humans remain accountable. Trust does not disappear; it changes shape. The more capable AI becomes, the more valuable human trust becomes.
AI also changes our understanding of expertise. For decades, expertise meant knowing more than everyone else. AI changes that equation. Knowledge is becoming abundant, so the scarce capability is judgement.
Human judgement in an AI enabled insurer is not reviewing every AI recommendation. It is knowing when not to follow it.
The future role of professionals is moving from producing every answer to exercising judgement over increasingly intelligent systems: asking better questions, recognising context, balancing competing values, and accepting accountability when decisions affect people’s lives.
Judgement becomes less about calculation and more about wisdom. This is why the biggest human risk AI creates is not job displacement, but judgement displacement. If we become overly dependent on AI for thinking, deciding, and questioning, we risk losing the capabilities that make organisations resilient.
Anthropology teaches us that humans do not just use tools. Tools reshape us.
The greatest risk is not AI replacing people. It is people gradually surrendering their curiosity, critical thinking, and responsibility because the machine appears more certain than they feel.
There is, however, a significant reason for optimism.
Insurance has always been remarkably good at adapting to new forms of risk. Climate change, cyber security, global pandemics, and autonomous vehicles have all challenged traditional models, yet the industry continues to evolve.
What gives me hope today is the willingness I see across the Irish insurance ecosystem to collaborate rather than compete on foundational capability.
At InsTech.ie, we have increasingly focused on developing innovation capability rather than simply introducing innovation tools. Our innovation bootcamps help organisations identify meaningful opportunities before technology is selected. Our Digital Sandbox allows teams to test ideas safely before scaling, while our communities of practice ensure learning continues beyond individual initiatives.
Because innovation is not an event. It is a system.
Over the next five years, the industry’s focus should be less on deploying AI and more on becoming organisations capable of adapting continuously. That means investing in leadership, experimentation, governance, and learning just as seriously as we invest in technology.
The insurers that thrive will not necessarily build the smartest models. They will build the strongest innovation systems.
The future of insurance will not be defined by who adopts AI first. It will be defined by who learns fastest, builds trust most effectively, and develops the human judgement to use AI wisely.
