Artificial intelligence has already become an integral part of the day-to-day operations of insurance companies. However, a report by the Capgemini Research Institute shows that simply implementing the technology does not guarantee better business results. The key to gaining a competitive edge lies in the ability to scale it effectively and align it with the organisation’s objectives.
According to the‘World Property and Casualty Insurance Report 2026’, only around 10 per cent of property and casualty insurers are effectively utilising AI on a large scale. This group achieves higher revenue growth and better stock market performance than other market participants. At the same time, as many as 60 per cent of companies remain at the testing or proof-of-concept stage, whilst 42 per cent do not measure the impact of their implementations using any metrics.
This means that, for many organisations, artificial intelligence remains a collection of isolated initiatives rather than a tool supporting the implementation of business strategy. AI is currently used, amongst other things, in customer service, insurance risk assessment, claims settlement and agent support; however, the financial benefits often fall short of expectations.
One of the main problems remains the lack of systematic measurement of outcomes. Without defined metrics, it is difficult to assess which projects deliver value and which require changes or should be discontinued. An increasing number of experts are pointing out that AI initiatives should be assessed in the same way as other business investments, taking into account their impact on revenue, costs and operational efficiency.
The report also highlights an imbalance in how the transformation is funded. On average, 72 per cent of AI budgets are allocated to technology and infrastructure, whilst only 28 per cent goes towards change management. Yet it is precisely the preparation of processes, skills and employees’ working methods that often determines the success of implementations.
“Most insurance companies today are not struggling with AI technology as such, but with scaling it effectively. The real differentiating factor is the ability to industrialise artificial intelligence — integrating it with business processes, translating it into financial results and ensuring adoption across the entire organisation. This is what distinguishes the experimental phase from achieving a real competitive advantage,” comments Fabrizio Casale, Continental Europe – Insurance Practice Head, Financial Services at Capgemini Poland.
The report’s authors point out that the most successful companies treat AI as part of their operating model, rather than as a separate innovation project. Between 2021 and 2024, they achieved a 21 per cent higher revenue growth rate and a 51 per cent greater increase in share value than other insurers. Their advantage stems from the simultaneous development of technology, staff skills, data quality and organisational management.
Team skills also remain a significant challenge. The report shows that 67 per cent of insurers cite a lack of knowledge about AI as one of the main barriers. At the same time, almost half of employees using these tools say that after 18 months, their day-to-day work has remained virtually unchanged. Furthermore, 43 per cent are concerned about the impact of AI on job security, whilst only 14 per cent clearly understand the role of this technology in their duties.
“AI will not deliver benefits if it remains merely a collection of isolated applications. True change lies in the transition from AI that supports human work to AI that actively operates within the organisation — integrated into core processes, supporting and orchestrating decisions, and implemented on an enterprise-wide scale. Such a transformation requires not only the right technology, but also a clear division of responsibilities, new skills and a fundamental change to the operating model,” adds Fabrizio Casale, Continental Europe – Insurance Practice Head, Financial Services at Capgemini Poland.
According to the report’s authors, the next stage in the sector’s development will be the AI-supported insurer model, in which technology takes over repetitive tasks, whilst experts focus on decisions requiring knowledge and experience. However, success will depend on the transition from experiments to full-scale implementations based on data, clearly defined responsibilities and a coherent business strategy.
