Many professionals believe that artificial intelligence (AI) will change the way people work even more rapidly than the Internet did. Insurance is no exception. Traditionally known for its reliance on legacy systems and manual processes, the sector is now undergoing a profound transformation driven by AI.
Although various studies assess the rate of AI adoption in the sector differently, one thing is certain – AI appears to be all-pervasive. Goldman Sachs’ 2025 Global Insurance Survey (covering 405 senior investment professionals in insurance), for example, reveals that 9 out of 10 respondents are currently using or are considering utilizing AI. Out of these, an outstanding 81% highlight reducing operational costs as their primary advantage from AI adoption, whereas 44% find it helpful in assessing insurance risk underwriting.
In this article, we will look into how AI transforms the insurer’s legacy model, what the most common challenges for its adoption are, and how Sirma can help lead the change.
The legacy model: what insurance used to be
Before the advent of AI, insurance relied on static risk models, predominantly manual processes, and broad customer segmentation. Underwriting and claims management were slow, paper-heavy, and reactive, with partial real-time insight. Organizations typically operated with fragmented data, rule-based decision-making, and limited automation. Processes were reactive and resource-intensive, with minimal personalization or predictive capabilities. This resulted in slower claims handling, generic customer experiences, and general inability to adapt swiftly to evolving risks and digital expectations. Claims processing, for example, has long been a pain point. Traditionally, customers had to initiate claims manually, often under stressful circumstances, and then endure a bureaucratic process involving multiple human touchpoints, document submissions, and lengthy wait times. For insurers, this meant high operational costs and exposure to fraud; for consumers, it resulted in frustration and a perception that insurers were more focused on denying claims than delivering support. In an age where digital-first experiences are redefining customer expectations, the traditional model has struggled to keep up, creating both a challenge and a prime opportunity for innovation.
How does AI transform the insurance industry?
AI, including GenAI and recently AI agents, is disrupting the insurance industry by transforming virtually every aspect of insurers’ operations. AI agents can help automate processes, enable predictions, boost customer service, detect fraud in real-time, etc. They can transform both core business functions, such as underwriting, claims processing, customer support, or sales, and back-office operations, such as marketing or product development.
Thus, AI’s uses for insurance are practically limitless. For example, data analytics helps assess risk by analyzing customer behavior. Machine learning detects fraud by spotting unusual patterns. Natural language processing (NLP) enhances customer service through chatbots. Computer vision evaluates images to assess damage in car accidents. Predictive modeling helps set prices and predict customer retention. Generative AI creates policy documents, processes paperwork, and enables personalized emails or messages. For instance, an AI chatbot may respond to policy queries instantly, while an image recognition tool might estimate repair costs from a photo of a damaged vehicle.
One of the most impactful and historically most complex - applications of AI in insurance lies in claims processing. Claims management has always been a cornerstone of insurance operations, yet it remains one of the most labor-intensive and error-prone aspects of the business. This is where AI, especially with the recent advancements in Generative AI and autonomous agents, is making a transformative difference. At Sirma, we have built and deployed a robust AI-driven solution specifically tailored for the health insurance sector to address precisely this challenge. The system allows policyholders to submit medical documents, diagnostic reports, and images via a digital interface. Once submitted, our AI models extract relevant information, classify documents, and structure the data in a standardized format, ready for seamless integration with the insurer’s internal systems.
This end-to-end automation drastically reduces the need for manual data entry and review, cutting down processing time from days to hours or even minutes. It also improves accuracy and consistency in claims handling - two areas that often suffer from human oversight and subjective decision-making. Moreover, this solution sets the foundation for the next level of AI-enhanced operations: intelligent decision support in claims adjudication. In upcoming phases, the system will assist with AI-powered approval or rejection of claims based on pre-defined business rules, past claim patterns, and risk profiles. By embedding AI deeper into the claims lifecycle, insurers not only achieve faster turnaround times but also greater transparency, improved compliance, and ultimately a better experience for both customers and claims processors.
Challenges for AI adoption in insurance
While the application of AI in insurance is vast, it is not without challenges. On the technical side, these may include poor data quality, fragmented systems, and legacy infrastructure that hinders effective integration. Cybersecurity threats increase as data volumes expand, and maintaining AI systems necessitates continuous updates and retraining. It often requires investments in in-house engineering teams and cloud computing resources, among other things. Yet, as research shows, it usually also necessitates organizational changes and the proper involvement of employees, i.e. the vital human dimension.
In analyzing how insurers should approach AI, Boston Consulting Group argues that AI’s early-stage practical applicability is debatable. Even though many early-stage demonstrations and proofs of concepts have shown huge AI’s potential to change insurers’ way of working, many pilots have also fallen short, and scaling has proved to be another problem. Complemented by the growing uncertainty that accompanies the speed of AI development, some insurers have adopted AI, whereas others have remained hesitant. Still, the analysis highlights that insurers should approach AI in all its forms by keeping in mind three key considerations: 1) technology should be focused on priorities, 2) its deployment could start by transforming a selected core function and then moving to the next, and finally, 3) it requires investment in people and processes.
The last perspective has been taken further by Wavestone’s Global Insurance Market Pulse. By arguing that high AI adoption in insurance is often accompanied by low enterprise impact, the research emphasizes that the greater challenge for scaling is aligning technology to the way people work rather than the technical aspects. To achieve a proper transformation, the development of people and technology should go hand in hand. Regardless of the path towards AI adoption, it is crucial that businesses can trust a proven technology partner who not only provides the technology but also leverages its domain expertise.
Leading AI transformation in insurance
Sirma has been a pioneer in the field of artificial intelligence for over 30 years, delivering AI-driven projects and solutions across a wide range of industries. With a particularly strong legacy in finance and insurance, we have consistently helped clients embrace innovation while solving some of the most complex challenges in the sector. Our long-standing expertise in building intelligent systems has uniquely positioned us to support insurers in navigating digital transformation, streamlining operations, and improving customer engagement.
One of our flagship offerings in this space is InSuite, our core insurance platform that enables end-to-end policy and process management. What sets InSuite apart is that it is built on Creatio, the leading AI-native platform for process automation and CRM. This combination allows insurers to rapidly modernize legacy systems, respond to changing customer needs, and launch new products or services without the traditional constraints of software development cycles.
Creatio technology offers a paradigm shift for the insurance industry. It empowers business users, without technical background, to design, test, and deploy workflows, customer journeys, and even AI-powered agents. Creatio’s AI-native architecture further enhances this capability by embedding intelligent agents into all areas of insurance operations, from underwriting and claims to customer service and sales. Through its drag-and-drop interface and extensive integration capabilities, insurers gain the agility to innovate at speed, while maintaining compliance and control.
With the combined strengths of Sirma’s domain expertise and Creatio’s cutting-edge platform, insurers are better equipped than ever to deliver the seamless, data-driven experiences today’s customers expect, and tomorrow’s market will demand.
Julian Masliankov, VP Insurance Solutions, Sirma Group Holding, shared
At Sirma, we’ve spent over three decades pushing the boundaries of AI in real-world applications, particularly in insurance and financial services. Today, by combining our deep industry knowledge with the technology power of Creatio, we’re giving insurers the tools to reimagine their operations from the ground up. AI isn’t just a technological evolution, it’s a strategic imperative for building the insurer of the future.
Final words
In a global, competitive landscape, embracing AI is no longer optional – it is a necessity for responding to evolving customer expectations and staying competitive in a dynamic market. AI adoption brings clear benefits to insurers, including higher productivity, streamlined processes, and empowered employees. Yet, it may also be accompanied by challenges. These can be successfully addressed if the insurer selects the right technology partner, one that can support the technology, is prepared to consider the vital human factor, and is capable of proposing and driving in the long term an overall AI adoption strategy.
