The future of AI in the insurance industry

Every now and then, some technological innovation is released that alters the world and the businesses must adapt or even may fall into irrelevance. A transition between largely agrarian and urban lifestyles became possible due to the steam engine of the Industrial Revolution and the mechanization of production. With the introduction of the internet came the increased real-time communication, e-commerce and cloud computing among others. Now, it's AI's turn. It is a highly powerful technology that is fast changing workflow processes, enhancing innovation, and remaking industries. Like in any other transformative technology effects, it will be difficult or even impossible to see companies including the insurers shun AI. Several decades ago, when e-commerce turned into an inevitable and streamlined and more advanced phenomenon, customers became accustomed to being able to order products easily and deliver them in a short time and expected all traders to have the same features. Likewise, AI has transformed consumer expectations so that now, customers are demanding:

• more accuracy and reliability throughout the consumer journey
• the ability to have human-like dialogues with AI bots (text or voice-based)
• hyperpersonal offers and communication
• on-demand products and interactions that are tailored to their needs

This is one of the differences between gen AI and other technological leaps: Gen AI can reason, make judgments, demonstrate creativity, and empathy on the level never before seen by other technological innovations, and these are particular skill sets that are of particular importance to insurers. This is why gen AI can actually revolutionize the insurance sector. The essence of insurance is acquiring a proper grasp of the specific risk, and helping people in distress in an efficient, effective, and tactful manner as possible.

All of this can be changed by AI

• Traditional analytic AI is cognizant of trends in data
• Gen AI extends these features with a better comprehension of unstructured forms of data and makes it possible to add hyperpersonalization and empathy to the responses
• Agentic AI brings new automation to complex work processes, enabling insurers to extract the maximum value Due to this flexibility, insurers are applying AI to all major functions, such as:
• sales productivity and hyperpersonalization
• automation and enhanced accuracy of underwriting
• augmented claims management
• customer service operations with voice agents
• change of back-office functions, such as finance, actuarial, and IT Like other technological innovations that can be groundbreaking, the consumers will gradually realize that AI can simplify their lives and become accustomed to it on the part of their service providers. The insurers that are able to capitalize on the chance of thoroughly infusing AI in all their activities will be in a position to emerge the best. They will then be in a position to do more business, at a faster rate and in a more personalized way with a better understanding of the underlying risk.

To achieve AI transformation successfully, one must not merely fiddle around the edges and do a few pilots, or that it is possible to achieve actual AI enablement by purchasing a patchwork of software-as-a-service products with a shallow strategic intent, or that workflow transformations will be achieved by off-the-shelf AI solutions. To generate sustained business value through AI, the insurers must:

• establish a bold enterprise-wide vision of what AI could become
• fundamentally, radically re-wire the way they conduct business in all areas of their business (underwriting, claims, distribution, customer service etc.) to make the technology part of the organization
• overhaul workflows
• re-architecture the way of operating
• work towards a modern data and technology stack
• scale AI using reusable components across different use cases and business domains And they will have to do this in a way that brings meaningful changes in unit economics. End to end processes will also have to be redesigned to derive value out of AI, and not just overlay AI over existing processes, or even worse, adding another step in a workflow with an AI tool that is unnecessarily added.

Insurance AI consulting expertise

In insurance AI consulting, the work on AI has been identified as a pioneer of digital changes by Forrester, and has been involved in AI with over 200 insurers worldwide. The division that deals in specialized insurance also has a library of:

• over 50 reusable AI components
• over 20 end-to-end insurance capabilities that a client can use and tailor With a holistic and strategic approach to AI, insurers can be in a position to be AI-native and develop a sustainable competitive advantage.

The screening is one of the reasons why gen AI has so much potential, due to reusable parts, then the technology can be used to revolutionize various aspects of a business. As an example, an AI gen capability that has been trained to respond with customer service is reused to:

• respond to internal IT support questions
• create marketing content
• review response to request-for-proposal
• write legal documents Although the final products differ greatly, the components of AI that lie behind it can be reapplied in different business fields and applications.

The future: AI multiagent systems

AI is still advancing at a fast rate. As an example, almost all the actions of customer onboarding in insurance would be performed by AI multiagent systems in the near future, which would serve as virtual co-workers. Information would be absorbed through an intake agent and then they would contact customers or intermediaries to clarify a point of information and also extract the data smoothly out of the complex documents like medical records or engineering documents. A risk profiling agent may develop a holistic risk profile of every case, based on current underwriting rules. A pricing and product agent could automatically price the case and recommend policy structures on how to satisfy the customer, such as attaching critical illness and disability riders to a life insurance policy. A compliance and fairness agent would make sure that the regulatory standards and high levels of ethical standards are met. The decision orchestrator agent might collate the input received by other agents in deciding whether or not the policy can be automatically approved or whether it has to be forwarded to a human senior underwriter to review it based on the size of the policy or other considerations. An agent based on learning and feedback may also maintain a continuously evolving model, take into account human feedback to enhance, and monitor drift, or performance degradation of a machine learning model over time.

Although there is much potential in AI within the insurance sector, it is difficult to scale it throughout an organization. The problem of:

• security risks
• high costs
• risks of getting locked up with suppliers
• shortage of talents in the organization
• cultural resistance
• lack of proper governance
• the existence of legacy infrastructure tend to slow down progress. It will take a real change to go out to the point and respond to these barriers in a considerate manner that does not result in a legacy of proliferation of approaches and solutions as we are witnessing today.

Change management is critical

This is why change management is a component of AI changes. Change management in our case constitutes half of what it takes to achieve financial and non financial impact and the other half is constituted by efforts to bring clean data to the models, the modeling process itself and integration of AI.

Virtual the insurers have started using AI, and they have many applications in manufacturing. Nevertheless, leaders of numerous insurers believe that their organizations are not AI native, as they have not adopted it in their business processes fully. This feeling promotes the urge to invest more in AI technologies because corporations are seeking to remain competitive. The insurers are keen on discovering the right formula to success yet very few have succeeded in doing so.

AI leaders are outperforming

The AI leaders among insurers already surpass their fellows. As an illustration, according to the research, the AI leaders of the insurance industry have generated 6.1 times more TSR in the last five years than AI laggards (2-3 times in most other industries). Despite the fact that not many insurance companies are deriving meaningful value through AI in the entire value chain on a significant scale, best-in-class insurers are domain-based approaches to change. They select some business functions – distribution, pricing and underwriting, claims, investments and overhaul the operation of that function.

Measurable improvements from AI transformation

To date, AI-based rewiring at the domain level has already affected several critical aspects of insurance enterprises, such as:

• 10 to 20 percent higher success rates of new agents
• 10 to 15 percent higher premium growth
• 20 to 40 percent lower costs of acquiring new customers
• 3 to 5 percent improvement in claim accuracy

The roadmap of transformation has already been released as the guideline companies should use to embrace the power of digital and AI. With the accelerating development and increasing influence of gen AI, this is becoming an urgent requirement. According to this framework, there are six signature moves that insurers can make in order to create organizations that will thrive in the digital and AI era:

1. Align C-suite to business-led AI transformation road map

The major insurers do not consider AI as another instrument of efficiency, but they understand it as a source of change and a possibility to enhance expansion, customer relations, and productivity. The change should be based on business value and the outcomes should be quantifiable. A state of AI transformation among insurers requires that they:

• motivate and concentrate the top team
• center their attention on a limited number of significant business domains and transform them through and through
• connect the outcomes of the transformation with a desired result in operational KPIs, including reduced customer churn It is essential to be wide-ranging, combining AI solutions throughout the business with a clear road map on how to consolidate various applications within a domain instead of implementing isolated applications to various domains.

2. Build the right talent bench

Achieve the status of digital leaders, insurers need to develop their talent pool by possessing a robust in-house digital talent pool, and preferably, 70 to 80 percent of the digital talent should be in-house. Digital leaders employ three major steps:

• they transition to a talent pool of more experienced and highly productive technologists and fewer beginners
• they create granular skill development grids supported by credentials to build excellence and identify unique technologists
• they create a specialized team to remodel their HR practices to recruit and retain digital talent Also, they are organizing a new era where the labour force consists of human beings and AI agents, which will compel the evolution of the organizational practice.

3. Implement a scaled operating model

When an insurer is transforming with the assistance of AI, it is necessary to choose an operating model that can help the insurer implement a strategy. Moreover, the introduction of the effective product management skills is essential, and it may assist in making the transformation a success.

4. Go fast and widely spread innovation by using technology

Best-in-class insurers are based on a malleable AI capabilities stack that is driven by reusable multiagent systems. The current AI technology stack of an insurer is very adaptable to accommodate the rapidly changing technology and is highly modular. The reuse of the underlying AI components, and capabilities is essential, as well as an agentic AI mesh architecture. This distributed, vendor-neutral architectural paradigm allows multiple agents to think, cooperate, and act independently on a variety of systems, tools and language models in a safe and scaleable way. The architecture is also developed in such a way that it can develop with the technology.

5. Embed data everywhere

State-of-the-art data capabilities will always be essential since any AI is data-driven. Although AI itself may contribute to data challenges resolution, the majority of insurers will have to advance their data potentials more radically to succeed in their AI vision. The development of these capabilities entails being able to overcome technical and organization hurdles. Intellectual property at insurance companies may become the capability to entrench and utilize the knowledge and "special sauce" of an insurance organization into agentic AI systems.

6. Invest in change and adoption management

Adoption ranks equally with development. By a norm, one dollar invested on the creation of digital and AI based solutions should be allocated at least one more dollar to make sure that the user adoption and scaled up to the enterprise level.

Domains

Domains represent the most basic elements of the business processes of the insurers. Some of the domains comprise primary functions:

• sales and distributions
• pricing and underwriting
• claims
• policy servicing The insurers normally possess between 10 to 15 areas that can be rewired using AI.

Subdomains

Subdomains are composed of domains. The underlying unit of an AI transformation is a subdomain, which is defined by three main features:

• people, assets, and capabilities collaborating to achieve shared goals
• business core outcomes of each domain
• specific AI applications across several business segments, which requires reusable components and specific success metrics

Insurers going through an AI transformation will be required to take into account which areas they need to change the most. Every typical insurer possesses several areas, including sales and distribution, pricing and underwriting, claims, and policy servicing, each of which has considerable opportunities of being optimized by AI.

The importance of domain-based implementation

The reorganization of a field needs size. In a move to realise the full potential of AI, insurers should not remain in fragmented solutions or use case initiatives and instead proceed to domain-based implementation. Individual applications are commonly created in order to prove the legitimacy of the gen AI technologies in a regulated setting. And as well as they can deliver encouraging outcomes, proof of concept and minimal viable product projects tend to concentrate on quick wins and are not long-term strategic fit, workflow integration, and long-term sustainable benefit capture. An end-to-end transformation of one to three domains to create meaningful impact can be domain-based, without necessarily overwhelming the organization. The required number of use cases to rewire a domain depends, although it is important to make sure that use cases used can result in the meaningful change and that all of them can enhance the performance. Insurers develop data preparation, systems integration and change management synergies through a domain-wide approach and re-organizing of complete workflows. That positions them to make actual improvements in efficiency, resources use, and sustainable competitive advantage. Successful AI scaling by insurers focuses on the most likely use cases with the greatest quantifiable impact on business. These impactful application uses are easy to adapt and implement in various fields by designing and generalizing a few very transferable AI abilities.

Insurers of all types is set to gain advantages in terms of operational efficiency and customer interactions due to the implementation of gen AI that will help optimize different aspects of the work.

Life

Life insurers can compliantly utilize gen AI to enhance risk evaluation and underwriting of policies by creating synthetic data, which supplements current data sets.

Health

In health Insurance, gen AI can be used to predict patient outcome and customize health plans, with the help of large bodies of data.

Commercial property and casualty

Commercial property and casualty insurers can take advantage of gen AI to create a detailed risk model and perform different scenarios to evaluate potential losses in a more efficient way.

Personal property and casualty

The Personal property and casualty insurers would find the use of gen AI relevant to process claims automatically and enhance fraud detection with the help of advanced data analysis.

Claims processing transformation

Among the domain-level, multi-use-case AI transformations in insurance, one can distinguish AI in claims processing. To enhance performance across its claims department, UK insurer Aviva deployed over 80 AI models to:

• reduce liability evaluation duration on challenging cases by 23 days
• enhance the effectiveness of claims being directed to the relevant teams by 30 percent
• decrease customer complaints by 65 percent Aviva informed the investor that refurbishing its motor claims sector saved the company over £60 million in 2024.

Insurance sales automation

An insurer has attempted to become personalized and efficient by creating intelligent automation to give quotes to potential clients and sell the policies. The outcomes were impressive:

• 80 percent of transactions shifted to the online platform
• customer satisfaction scores increased by 36 percentage points

Chatbot implementation

Implementation of a chatbot that works 24/7 contributed greatly to an insurance carrier providing its after-hours customer service, as the number of potential customers that have purchased its policies increased by 11 percent.

AI with empathy

A different carrier is applying AI to produce the approximately 50,000 communications it makes every day relating to claims and finds it easier and more empathetic to read than human-written communications.

After a specific area is identified to undergo a transformation, it should be broken down into a sequence of AI-based modules that may be executed, optimized, and extended. To illustrate, to transform the sales and distribution process, it is possible to integrate capabilities of gen AI and predictive analytics (like propensity models) into interlocked end-to-end AI systems, including multifaceted copilots and next-generation AI chatbots. The insurers can unlock sustainable value by transforming AI on a domain basis. Gen AI has a massive potential in the sales field to enhance the productivity and efficiency of employees by saving time on simple activities.

To embrace a successful AI transformation, a paradigm shift in the way businesses are run is essential to adopt an operating model that will be scalable. A successful implementation of AI throughout the enterprise requires an insurer to possess the appropriate system. Discussing the insurers undertaking the AI transformation journey, they must select an operating model that fits their overall strategy. This may be:

• moving to a digital factory model of between 20 and 50 pods
• a product and platform model that accommodates a much larger number of pods
• a more wholesome enterprise-wide agile business model that extends the agile advantage to the whole business, not just the technology intensive centers

The role of product management

One of the most important elements in the successful implementation of the selected operating model is that there should be sound product management capabilities because these can greatly determine the success of the transformation exercise. The insurers can unite business functions, data, and technology into a team-based and agile operating model, thus:

• silo-breaking
• establishment of an ownership culture
• prioritizing the enterprise with a customer-focused mindset

AI control tower

The AI control tower is also vital as it governs and monitors the AI-driven value creation and adoption organizationally. The central AI teams are getting more and more prominent as the insurers grow increasingly concerned with reuse of components and standard AI. The IT is collaborating more closely with data and AI teams as they use greater engineering and cloud capabilities. Simultaneously, such trends should be balanced in terms of front-line business ownership to orient tech capabilities on the correct issues and to create actual value. One such potential model is the product-based model where teams are based on the core "super products" along the insurance value chain.

The aim is to create sustainable value with technology: insurers must use AI-first solution and revolutionize the entire capabilities stack of their organization. This will help organizations to:

• be flexible
• embrace the most recent AI innovations
• avoid the production of archaic technologies that will repress future growth and innovations The actual modernization will be to make the AI elements and functions reusable, to harmonize the standards throughout the enterprise and with high-quality data to train the models.

Four layers of the AI stack

The capabilities stack is elaborate to ease the adoption of AI in the enterprise. This AI stack has four layers that are crucial:

1. Reimagined engagement
2. AI-powered decision-making
3. Infrastructure
4. Data platform The layers must be strategically invested in to ensure the enterprise-wide use of AI. Banks can inspire insurers to develop their AI stack, but specifically, they can consider significant differences between the two fields.

Insurers should reimagine their engagement with customers, using AI to provide highly personalized experiences with a seamless customer interaction. Major insurance companies are also applying AI to enhance their interaction with customers, and they combine several communication channels that have the following features:

• text chatbots
• images that enable customers to comprehend complicated data
• voice assistants that allow customers to talk to their insurance company rather than typing The AI is used to make sure that the customer experience is not only human like, but also seamless and even channel-to-channel. To illustrate the point, when a customer initiates a conversation using the mobile app and proceeds with it via a phone call, the AI will consider the former input, thus the customer will not need to input the information again.

Gen AI consumer adoption

Gen AI is also getting to be recognized by customers. Out of individuals utilizing the use of gen AI applications like ChatGPT, 29 percent seek financial or investors information, advice, or recommendations. The increase in the number of consumers using gen AI tools to accomplish tasks such as insurance offers will leave carriers with no option but to take their advisory, product value communication, and price transparency to the next level.

The AI-driven decision-making layer is a layer that explores mountains of data that have been created through the different channels to offer a very personal customer and employee experience. This layer:

• augments the current pricing and underwriting decisions
• supports claims decisions
• enhances the accuracy of claims through the dynamical assessment of the data points including adjuster notes, damage pictures, text submission or documents and claim histories Indicatively, a carrier has created a consolidated product storage where policy documents cut across the global enterprise; call center agents are able to respond to coverage, exclusions-related inquiries, among others easily. Within the property and casualty space, AI and claims data are being used together by many carriers to detect new risk factors, including damage estimates caused by climate.

Advanced AI functions

With the development of AI technology, the major insurance companies are leaving the traditional predictive models and turning to advanced functions like:

• multiagent systems
• multi-step reasoning

Gen AI agents

The use of AI is being revolutionized through agentic AI. Gen AI agents are sophisticated AI systems being capable of judgment application and are usually geared towards conversation with a user based on vast scientific understanding and past data. Several agents are involved in collaborative efforts in accomplishment, including satellite and drone imaging to assess and prevent damages. They are also incredibly insightful, offer human agents real-time assistance, and propose appropriate plans of actions. The potential of Gen AI agents is to contribute to:

• greater involvement of customers
• automation of complicated processes
• increased productivity

Multistep reasoning

Multistep reasoning, in its turn, enables an AI system to divide a complex problem into multiple smaller and manageable steps and subsequently address each of those steps consecutively. As an example, a multistep reasoning AI system might be adopted by an insurer and concludes the damage and payment on a claim.

Real-world implementation

As a case in point, one of the North American based leading insurers is adopting agentic operations across its underwriting processes. This implementation has revealed different implicit judgments through which underwriters have long been used and which they have incorporated into new regulations and guidelines to improve the effectiveness and uniformity of its process of underwriting.

Scaling reusable components

The scaling of reusable and standardized components should also be prioritized by the insurers to get maximum value out of AI. AI models and pipelines are to be developed as modular, interoperable, code resources, which can be applied in various areas. Given the example of an AI-assisted document classification engine created to support underwriting, a similar engine can be used to improve claims processing and policy service as well. Standardized AI frameworks, APIs, and code assets can be:

• more cost-effective to invest in
• lessen the time of development
• reduce redundancy
• hasten the implementation of AI throughout the enterprise Insurers can derive far greater value on their AI investments by considering AI as an incremental ability and not a collection of bespoke projects.

A robust infrastructure layer offers the features, which allow AI to operate and generate value, such as machine learning pipelines that can execute large AI models with low costs.

Build, buy or partner decision

Making whether to develop AI solutions internally and establish intellectual property, or outsource the development of AI potentials that may become the sources of the high value IP in the future is a high stakes decision among the insurers that influences their scaling, differentiation, and responsiveness to market.

In-house development

In-house development of AI capabilities can enable custom solutions that better match the needs of a particular business and can hold an insurer's "secret sauce," to the AI capabilities stack with a moat of protection, which gives a better control and differentiation opportunity. Nonetheless, this strategy requires considerable investments to specific talent, infrastructure, and long-term development cycles, which, however, do not always prove to be cost-effective.

Acquiring AI solutions

By contrast, acquiring AI solutions through established vendors will allow faster implementation and will be based on proven technology, but will be limited in terms of:

• customization
• integration
• long-term cost
• third-party product road map dependency
• market-mean execution through the utilization of tools and capabilities already in use by others

Hybrid approach

A mixed solution can compromise on scalability and strategic control. By outsourcing the services of the insurers who provide standardized solutions incorporating gen AI, in particular, in corporate operations, including finance, human resources, and procurement, the insurers can dedicate internal resources to the main operations of the companies that include underwriting and claims management. The tailor-made AI in such locations can be used as a source of differentiation, utilizing the proprietary data and expertise in specific fields to increase competitive advantage. This is a cost-effective, high-speed, and differentiation strategy that allows:

• careful build-versus-buy choices
• consideration of the long-term business goals
• individual business requirements A hybrid system would require the creation of an in house orchestration capacity that would be capable of combining the solutions both internal and external, both technically and in terms of value and ending up giving the insurers another IP that would act as one of the competitive distinguishing factors. Also, the creation of a dynamic network of partners enables the insurers to acquire innovative external knowledge and solutions in the areas where the internal ones might be weak.

Making strategic decisions

Considering the changing AI environment, insurers need to be cost-benefit-driven and have a long-term view of the changes in order to navigate such decisions. To date very little of the insurers have laid down a strict structure of the decision between building and buying, and even fewer are exploring the re-examination of this structure in an age of accelerated technological evolution. The analysis of the need to construct or to purchase AI technologies must have clear:

• value creation
• cost-efficiency
• speed to market analysis
• long-term scalability Technical capability, complexity of integration, regulatory compliance, and data security are also factors that need to be put into consideration by the insurers. In case of outsourced capabilities, the management of technology vendors needs the most attention where the insurers should select, assess, and manage AI and cloud service providers with care to achieve compliance, interoperability, and scalability in the long-term. The right decisions make AI investments business-focused, agile, innovative and competitive at all times.

Addressing legacy systems

Another significant problem in IT transformation is the lack of documentation of legacy systems - systems that are using outdated technology that is not supported by the vendor anymore. This legacy system infrastructure will require insurers to upgrade it to the fullest to realize the benefits of AI, defeating the strict IT systems that are simply not scalable and cannot process in real-time. Gen AI assists in de-mystifying legacy systems, code is analyzed to create structured documentation, organizations can maintain institutional knowledge with the assistance of gen AI. Gen AI also improves the productivity of developers, using:

• automated code creation and testing
• reducing manual effort
• speeding up the release process CIOs and CTOs can use lessons gained in past technological revolutions to advise the C-suite in expanding gen AI beyond pilot projects to sustainable business value.

Cost reduction examples

As one example, several years ago, a major financial institution needed a bill of over $100 million to have an upgrade in a transaction processing system. Gen AI has reduced the costs by more than half. In the meantime, one of the leading 15 insurers worldwide applied gen AI to achieve:

• over 50 percent efficiency in the efficiency of modernization of codes and testing
• over 50 percent faster achievement of coding work

Insurers need to invest in the data infrastructures needed to train and scale multiagent AI systems, and have a smooth integration of business functions. A hybrid cloud infrastructure, i.e. a mixture of on-premises data center and a public cloud environment, must be developed in such a way as to be scalable; furthermore highly configurable core product processors are also designed to ensure flexibility and efficiency. The insurers need to consider the data quality and availability on the data governance front, but also the challenge of handling sensitive information. When legacy systems become a challenge, then the insurers might be forced to upgrade their IT environment so that it could be able to support the large-scale adoption of AI.

In order to achieve successful AI implementation, a culture of innovation, mindset shift and capability development are necessary, yet the organisations tend to underestimate the resistance levels and willingness to adopt a new approach to work. It is necessary to provide employees with the appropriate skills and form a clear vision of the facilitating role of AI in assisting them to work. The top insurers have organized change management programs that focus on:

• leadership role modeling
• articulating the value of AI
• comprehensive capability building programs
• the advancement of the right performance frameworks

Addressing employee concerns

The adoption of AI technologies in organizations can make employees anxious about their functions. Nevertheless, history has demonstrated that technology usually introduces new needs and opportunities and thus new roles and responsibilities are created. Finally, the implementation of AI in workflows will have to be based on developing the sense of shared ownership and responsibility towards AI implementation in the organization.

Technology is only half the battle

Excellent technology is not enough as it is half the battle. The other half is to make employees actually apply AI in their daily routine, and to move the needle in the manner in which work is performed, be it automation or augmentation.

Experience suggests that the use of AI is an inevitable requirement to remain competitive. There are not many dominating insurers that have completely AI operationalized, yet it is an enticing example that other companies looking to seize the chance to change should follow. These insurers already have the lead and new technological developments have provided them with the means to accelerate even more.

The pilot purgatory trap

The rest are trapped in pilot purgatory and fall into the numerous traps:

• They do not have the aggressive, companywide AI approach with quantifiable financial results that can move the company out of its AI inertia
• They fail to recognize the entire scale of investment requirements and thus make small-scale and piece-meal efforts with less ROI
• They are narrow use case oriented rather than transforming the domain
• They do not develop business line reusable components to reduce the value of AI in the long run
• They also overly depend on off-the-shelf solutions, which make them less aligned to their business peculiarities and cannibalize their own capability to produce new-age intellectual property

In order to stay up to date with the fast changing world, insurance companies need to take a radical enterprise-wide perspective on AI, completely rewiring the business and integrating AI into all their processes. This includes:

• building enterprise-specific systems
• training AI models on internal data
• re-tooling business processes to outcompete in selected lines and markets
• re-evaluating the operating model
• re-using AI with reusable components to achieve its transformative potential Rewiring the operations with AI first will provide insurers with a long-term business value and outperform their competitors.