The insurance industry has always lived at the intersection of risk and uncertainty. Its very purpose is to measure the unmeasurable, price the unpredictable, and provide a safety net in moments of loss. That makes insurance uniquely sensitive to change in customer expectations, societal shifts, or technological disruptions.
Artificial intelligence brings both promise and peril to this equation. On one hand, AI offers insurers unprecedented opportunities: the ability to harness vast datasets, spot emerging risks in real time, personalize coverage, and transform the speed and accuracy of decision-making. On the other hand, it introduces new questions around fairness, explainability, and trust. If AI becomes a black box that customers or regulators cannot understand, it risks undermining the very confidence on which insurance is built.
The AI opportunity is hard to ignore
Even for the naturally risk-averse insurers and underwriters, the AI opportunity is hard to ignore. Innovators are already reaping the benefits, harnessing data through predictive analytics, AI, and other cutting-edge technologies to make confident decisions and enhance risk assessments. Over the past decade, a wave of InsurTech companies has redefined what’s possible in insurance. Unlike traditional carriers, these firms were born digital, leveraging AI, machine learning, telematics, IoT, and seamless digital experiences from the start.
This doesn’t mean that traditional insurers are obsolete. In fact, many have partnered with or acquired InsurTech firms to accelerate innovation and modernize their platforms. But what InsurTechs have proven is that AI-driven insights and customer-centric design are becoming essential for improving risk modeling and ensuring transparency. It takes only a fraction of the time for insurers to truly personalize their offerings using customer intelligence and to demonstrate fair outcomes to regulators with confidence.
One aspect of AI that will be particularly interesting to follow is how roles and responsibilities evolve, not only within insurance companies but also in the regulatory sphere, as authorities adapt to new technologies. 81% of underwriting executives expect that AI will provide richer opportunities for people to grow in their careers, shifting their focus away from low-value tasks like data gathering toward more meaningful and strategic work. For those looking to build a long-term career in insurance, this represents a promising outlook.
Despite the potential, most insurers haven’t fully adopted AI. In fact, according to Capgemini’s research, only 27% use predictive modeling like sophisticated algorithms to anticipate risk and behavior, while 37% use advanced third-party data from outside datasets like credit, health, or environmental info. This shows a large gap between ambition and execution.
Why rules-based processes must be left behind
Insurers are facing many challenges as the insurance ecosystem shifts rapidly in favor of technologically savvier players. On one side, customers want faster, more transparent, and fairer insurance experiences like instant quotes, smooth claims, and personalized offers, while new risks and tighter regulations press on the other. Among these risks are cybercrime, climate change, volatile financial markets, and digital fraud, which further complicate what insurers need to assess and account for, while being pressured by authorities for more rigorous oversight, transparency, and fairness in underwriting and claims management.
Together, these expose the limits of rules-based processes. The traditional, fixed decision frameworks that rely on predefined formulas, criteria, and thresholds are designed to handle risks in a consistent way, but they don’t adapt well to complexity and changing conditions. This opens a loophole for a new kind of insurer, the one that relies on intelligent data and instant insight.
Why legacy platforms work no more
Years of market consolidation have left insurers grappling not only with disparate datasets but also with outdated technology platforms, some of which date back decades. These legacy systems, heavily customized around underwriting, claims, and policy administration, once provided stability and accumulated company knowledge. Today, they act as a burden, stalling innovation and draining resources without ROI.
The scale of the problem is striking. According to Earnix’s 2024 Industry Trends Report, around 49% of global insurance executives report that technical debt has become a major barrier to transformation. In another survey, 44% highlight inconsistencies in their data, while 41% cite a lack of interdepartmental transparency, both of which make it difficult to build a unified view of customers and risks. And despite advances in automation, a large part of industry professionals depend on manual data entry for core activities such as underwriting and claims. Perhaps most shocking of all, only 8% of property and casualty insurers leveraged data-driven underwriting enabled by advanced technology capabilities in 2024—a stark reminder of just how far the industry still has to go.
The impact is profound. Product launches could take up to 12 months for legacy systems, (compared with 3–6 months for InsurTech challengers) and adding AI on top of such systems could be particularly challenging, especially when innovation budgets are already constrained because most IT spend is absorbed by legacy maintenance.
Where to start with automation and AI
Automation and AI adoption don’t have to be a sweeping, all-or-nothing transformation. Insurers can take a phased approach—building momentum with targeted initiatives that prove value quickly and create a foundation for broader change.
Move to cloud and modularized platforms
One starting point is adopting cloud-native integration. Modern platforms can be connected to existing policy administration, claims, and CRM systems through API-first architecture, unlocking flexibility without requiring a full system overhaul. From there, insurers can begin shifting toward modularized platforms with microservices, which allow them to scale specific capabilities independently. This not only accelerates delivery and ROI but also makes it easier to embed AI directly into the services where it adds the most value.
Try high-value use cases and learn from other industries
Focusing on high-value use cases is another proven strategy. Claims automation and underwriting AI are strong candidates to demonstrate quick wins, prove ROI, and free up resources to reinvest in broader modernization. Over time, this creates a cycle of efficiency, insight, and innovation. Insurers should also look beyond their own industry, borrowing lessons from sectors such as banking and retail that are further ahead in AI adoption, and adapting those practices to their own use cases.
Prioritize education for both leaders and employees
The most critical investment will be in people and culture. Employees need training, support, and encouragement to embrace AI as an opportunity rather than a threat. Early adopters should be rewarded, best practices shared, and experimentation encouraged.
Understanding AI goes far beyond knowing how to write a prompt.
At the same time, leadership has a pivotal role to play — first by filling knowledge gaps at the board level. Understanding AI goes far beyond knowing how to write a prompt. Executives need to grasp what AI really is (and isn’t), the differences between machine learning and AI, generative AI and agentic AI, and data science, and how emerging tools can be applied to solve their problems. With that depth of understanding, leaders will be better positioned to present opportunities clearly, establish the right controls to mitigate risks, and set a vision that balances innovation with responsibility.
In short, the path forward is not about rushing headlong into AI. It is about building the right foundations, starting with high-impact areas, and creating the cultural and leadership alignment needed to turn automation into a sustainable competitive advantage.
In a recent interview on AI in FSI, our board advisor, Gary Duggan, reflected on HTEC’s approach in helping insurance clients modernize their operations and incorporate AI tools into their processes:
Similar to how organizations were adopting Agile methodologies, we start with rapid prototyping with a select piece of business and a group of forward-thinking employees looking to enrich their careers and do a better job for their customers, who would welcome and test new tools. Then we iterate – build, test, and learn, and then gradually roll it out to other lines of business. HTEC took this approach with an insurance client, and it seems to be working very well in bringing hearts and minds together. Over time, we start to build some real advocates who can champion the change and the improvements it has brought about, and then other people in the organization begin to notice it and want to take part.
Strategic imperative: wider adoption and scale-up of AI
For insurance leaders determined to seize the future rather than be shaped by it, the path forward is clear. The focus is on transforming legacy systems and fragmented data ecosystems into platforms that can fully support the next era of underwriting, claims, and risk management. It requires aligning business vision, technology modernization, and cultural readiness to unlock the full value of AI and automation.
Those who start now, experimenting with targeted use cases and scaling what works, will build the agility and insight needed to compete in a world driven by intelligent data. The insurers that succeed won’t just react faster—they’ll anticipate risk, personalize coverage with precision, and deliver fairness and transparency by design.
