All posts by : Author3

For more than a decade, insurers have been examining how technology is changing underwriting.

Yet one challenge has remained remarkably persistent: underwriters spend too much time doing work that is not actually underwriting.

Across the industry, professionals have traditionally devoted a significant portion of their working day to activities such as collecting information, checking documents, entering data, coordinating administrative tasks, and navigating multiple systems.

Recent research suggests that this is beginning to change.

The improvement may have been gradual so far, but the expectations surrounding artificial intelligence and automation are anything but incremental.

For the first time, many insurance executives appear to believe that technology could fundamentally reshape how underwriting is performed—and do so at a much faster pace than previous waves of innovation.

The Difference Between Another Technology Wave and a Real Shift

Insurance has experienced its share of technological revolutions.

Knowledge-management systems promised easier access to information. The Internet of Things introduced new sources of real-time data. Advanced analytics gave insurers increasingly sophisticated ways to identify patterns and assess risk.

Each became part of the broader insurance technology landscape.

But none completely redefined the underwriter’s role.

AI may be different.

The combination of generative AI, automation, advanced data ingestion, natural-language processing, and increasingly intelligent decision-support tools has the potential to address one of underwriting’s most persistent problems: the amount of time spent assembling and processing information instead of applying expertise to risk.

That distinction is important.

The goal is not simply to make existing underwriting faster.

It is to rethink what the underwriter should actually be doing.

Automation Could Change the Equation

Recent executive research points toward a significant reduction in the amount of time underwriters may spend on non-core activities as AI and automation mature.

Across different insurance segments, executives increasingly expect these technologies to have a meaningful impact on underwriting.

The change is already underway.

Over the past several years, insurers have experimented with AI in areas such as data collection, information synthesis, risk analysis, and underwriting support.

Not every experiment has delivered the expected results. But the broader direction is becoming clearer: AI is increasingly being viewed as a practical tool for removing friction from underwriting rather than simply an experimental technology.

Several workforce expectations illustrate the scale of the change:

• 81% of surveyed underwriting executives expect AI and generative AI to create new roles to a large or very large extent.
• 65% believe their workforce will require additional skills as AI becomes more deeply integrated into underwriting.
• 42% expect they may need access to external talent pools to fully capture the technology’s potential.

These figures point to an important conclusion.

The AI transformation of underwriting is not only a technology story.

It is a workforce story.

The Rise of the AI-Augmented Underwriter

The underwriter of the future is unlikely to be replaced by a machine.

Instead, the role may increasingly become a collaboration between human expertise and machine capabilities.

AI can already support many activities that traditionally consume substantial amounts of underwriting time.

Natural-language systems can interpret requests from customers and brokers, identify relevant information, and route inquiries toward appropriate workflows.

Automated data ingestion can collect and organize information from multiple sources.

Pattern-recognition models can identify relationships and anomalies that might otherwise require significant manual investigation.

Decision-support tools can help assess straightforward cases, while automated workflows can coordinate multiple steps within a single process.

The result is a potential shift in the division of labor.

Machines handle more of the information-heavy work. Humans spend more time on judgment, relationships, exceptions, and complex risk decisions.

That does not make underwriting less important.

It makes the human contribution different.

From Data Collectors to Risk Decision-Makers

Consider how much of an underwriter’s expertise can be buried beneath administrative work.

A professional may have years of experience assessing complex risks, yet much of the working day can still be consumed by finding documents, reconciling information, entering data, requesting missing details, and moving between systems.

AI has the potential to absorb more of these activities.

Instead of beginning every assessment with a blank screen and a collection of fragmented information, an underwriter could increasingly begin with an AI-generated view of the risk, supported by relevant data, identified patterns, and suggested next steps.

The human then becomes the critical layer of judgment.

They can challenge assumptions, investigate unusual circumstances, apply contextual knowledge, communicate with brokers or customers, and make decisions where automated systems are less reliable.

This is not the disappearance of underwriting.

It is the reinvention of underwriting work.

Three Priorities for an AI-Enabled Underwriting Future

Technology alone will not deliver this transformation.

Insurers will need to rethink strategy, talent, workflows, and organizational culture at the same time.

1. Build an AI-Led Strategy

AI initiatives should not exist as disconnected experiments.

Insurers need a clear strategy for how AI will operate within their broader technology environment, supported by a strong digital foundation.

As AI systems become increasingly agentic, the opportunity becomes even broader.

Instead of simply using AI to answer questions or summarize information, underwriters may eventually be able to delegate individual workflow tasks to specialized AI agents.

An agent could gather information, another could organize documents, another could compare relevant risk factors, and another could prepare a preliminary assessment.

The underwriter remains responsible for the overall decision while AI coordinates more of the surrounding work.

2. Reimagine Talent and Workflow

Introducing AI without redesigning the underlying workflow can limit its value.

Insurers should consider how work should be divided between people and machines, which skills will become more important, and where human expertise will deliver the greatest value.

A skills-based approach can help organizations identify emerging capabilities, retrain existing employees, and prepare teams for new responsibilities.

At the same time, AI adoption needs to be connected to broader process redesign.

Simply adding an AI tool to an inefficient workflow does not create an efficient workflow.

The process itself may need to change.

Responsible AI principles should also be embedded throughout this transition, particularly when automated systems influence important underwriting decisions.

3. Create a Culture of Experimentation

AI is developing too quickly for organizations to rely entirely on traditional top-down innovation models.

Employees working closest to underwriting processes often have the clearest understanding of where technology could remove unnecessary effort.

Giving teams room to experiment can reveal valuable use cases that may not emerge from a centralized technology strategy.

The objective is not uncontrolled experimentation.

It is structured curiosity: allowing employees to test new capabilities while maintaining appropriate safeguards around core decisions, data, security, and risk.

The organizations that learn fastest may be those that create enough freedom to experiment without losing control of the decisions that matter most.

The Underwriter Is Not Disappearing

Technology has repeatedly changed the tools underwriters use.

AI may change the work itself.

But that does not mean human expertise becomes less valuable.

In a more automated environment, underwriters may spend less time collecting information and more time interpreting it. Less time navigating administrative processes and more time evaluating complex risks. Less time performing repetitive tasks and more time exercising judgment.

The central question is therefore not:

“Will AI replace the underwriter?”

A more useful question is:

“What could an underwriter accomplish if AI handled more of the work surrounding the decision?”

That question opens a much broader vision for the future.

From Automation to Augmentation

The next chapter of underwriting is unlikely to be defined by technology alone.

It will be defined by how effectively insurers combine human judgment, intelligent automation, data, and increasingly capable AI systems.

Previous technology waves changed individual parts of underwriting.

The current generation has the potential to connect those parts into something much more integrated.

If insurers build the right digital foundations, rethink workflows, invest in new skills, and encourage responsible experimentation, AI could help move underwriting away from administrative complexity and toward what it does best: understanding risk and making informed decisions.

The future underwriter may not be less human.

They may simply have a much more capable machine working beside them.

Artificial intelligence is moving beyond the stage of being a tool that people use.

It is increasingly becoming a system that can reason, coordinate, create, test, learn, and act.

This shift is particularly significant for insurance, an industry built around information, rules, decisions, documentation, and complex workflows. As generative AI evolves into more autonomous, agentic systems, insurers are beginning to reconsider not only what technology can do, but how technology itself should be built and integrated into the enterprise.

One emerging concept captures this transition: the Binary Big Bang.

It describes a defining moment in the evolution of AI and software development, where autonomous systems begin challenging long-standing assumptions about how digital products are created, how much they cost to build, and who—or what—participates in their development.

The implications for insurance could be substantial.

Breaking Through the Natural-Language Barrier

Foundation models changed the relationship between people and software by making natural language a powerful interface for interacting with technology.

Instead of translating an idea into highly structured instructions, people can increasingly describe what they want in ordinary language and allow AI to interpret, develop, and refine the underlying solution.

This dramatically expands the possibilities for software development.

For insurers, generative AI is therefore more than another layer of automation.

AI models and agents are becoming potential components of the enterprise itself, with applications spanning customer service, underwriting, claims, risk assessment, product development, and operational management.

The opportunity is not simply to automate today’s processes.

It is to rethink the processes themselves.

Insurance executives can begin building what might be described as a cognitive digital brain—an interconnected environment in which data, AI models, workflows, organizational knowledge, and autonomous agents work together.

The value comes from the connections between these components.

From AI Assistants to AI Agents

The next stage of this evolution is agentic AI.

AI agents are designed to pursue goals, reason through problems, use external tools and information, make decisions, and take actions with varying degrees of autonomy.

For insurers, this opens the possibility of distributing parts of the technology development lifecycle across specialized AI agents.

A requirement-management agent, for example, could interpret business needs, organize priorities, track progress, and ensure that development remains aligned with defined objectives.

A code-development agent could translate requirements into structured software components while maintaining traceability between business needs and technical implementation.

A testing agent could simulate different user scenarios, identify potential issues, and repeatedly test applications throughout development.

A deployment and support agent could assist with releasing applications into production and identifying or resolving environment-specific issues after launch.

Instead of software development being a linear sequence of human-led activities, it could become a coordinated ecosystem of specialized digital workers.

That has the potential to change both the speed and economics of building technology.

Three Forces Reshaping Insurance Technology

As AI becomes increasingly embedded into technology environments, three interconnected forces are emerging: abundance, abstraction, and autonomy.

1. Abundance: More Technology, Faster

Legacy technology remains a major challenge for insurers.

Maintaining aging systems can be expensive, while modernization efforts often require significant time, specialized skills, and investment.

AI could change the economics of this equation.

Generative AI can accelerate software development, help interpret legacy code, identify technical debt, generate documentation, and support the migration of older applications into modern environments.

The result could be a greater capacity to build and improve digital systems without relying entirely on traditional development models.

Research indicates that 78% of insurance executives believe AI agents will reinvent how their organizations build digital systems.

The demand for this additional capacity is also clear. If software engineering resources were unlimited, 62% of executives would prioritize launching new products and services, while the same proportion would prioritize adding new features to existing offerings.

AI-driven development could help narrow that gap.

2. Abstraction: Making Complexity Easier to Navigate

Insurance contains enormous amounts of complexity.

Underwriting decisions, claims processes, policy rules, customer interactions, regulatory requirements, and internal workflows all involve multiple layers of information.

Generative AI can help make that complexity more manageable.

Instead of forcing employees to navigate numerous systems and information sources independently, AI can summarize information, surface relevant insights, provide recommendations, and create more intuitive interfaces.

In underwriting and claims, AI can support decision-making by bringing together relevant information at the right moment.

In customer service, agentic systems can use customer context to create more personalized interactions.

The technology essentially becomes a layer of abstraction between people and underlying complexity.

Employees do not necessarily need to understand every technical detail behind a system to use its capabilities effectively.

3. Autonomy: Moving From Assistance to Action

The most significant change may be the transition from AI that assists people to AI that can perform defined activities independently.

Autonomous systems can increasingly analyze information, make decisions within established parameters, execute workflows, and respond to changing conditions.

This does not mean removing humans from the equation.

Instead, it creates the possibility of designing workflows in which technology handles predictable, information-intensive activities while people remain responsible for oversight, judgment, exceptions, and strategic decisions.

As data becomes more integrated, insurers could potentially encode business processes, institutional knowledge, rules, and workflows into interconnected AI environments.

The result is an operating model that can respond dynamically rather than simply following rigid sequences of instructions.

AI Turns Data Into a Working Asset

Insurance has never suffered from a lack of data.

The challenge has often been making that data accessible, understandable, and useful at the moment a decision needs to be made.

AI can help change that.

Modern AI systems can identify patterns, connect information from different sources, surface previously overlooked relationships, and deliver relevant information to employees when it matters.

This can influence virtually every stage of the insurance technology lifecycle.

AI can support:

• Generating documentation, use cases, data dictionaries, and user stories
• Configuring information for modern technology platforms
• Rewriting legacy applications for newer technology environments
• Reconsidering requirements earlier in the development process
• Creating comprehensive test cases before a new application is built
• Connecting business requirements more directly with technical implementation

This creates a different development philosophy.

Instead of waiting until the end of a technology project to test whether the solution meets business needs, AI can help validate assumptions much earlier.

That can reduce rework, accelerate development, and improve the connection between technology and business outcomes.

The New Generation of AI-Powered Underwriting

Underwriting provides a particularly clear example of how these capabilities can come together.

AI-powered underwriting systems can analyze submissions, identify missing information, assess whether a risk fits established criteria, and surface insights that help underwriters make decisions.

The potential value is not simply speed.

It is the ability to process larger volumes of information consistently while giving skilled professionals better context for complex decisions.

Similar approaches are emerging in reinsurance, where AI assistants can monitor information from a broad range of sources, synthesize relevant developments, and provide underwriters with a more current view of potential risks.

As these systems mature, the underwriting process could become less dependent on manually searching for information and more focused on interpreting insights and exercising professional judgment.

The human role does not disappear.

It becomes more concentrated around the decisions where expertise matters most.

A New Architecture for Insurance

The Binary Big Bang represents more than another stage in the technology cycle.

It points toward a different way of building and operating insurance businesses.

Software may become easier to create. Digital capabilities may become more abundant. Complex processes may become easier to navigate. And autonomous systems may increasingly perform work that previously required significant human intervention.

But the real transformation comes from combining these capabilities.

An insurer’s competitive advantage may increasingly depend on how effectively it connects AI, data, people, workflows, and institutional knowledge into a coherent digital environment.

That requires more than adding AI tools to existing systems.

It requires rethinking the architecture of the business itself.

From Automation to Reinvention

The most important question is no longer simply:

“What can AI automate?”

A more consequential question is:

“What could insurance become if technology could build, understand, and operate parts of the business alongside people?”

That is the deeper significance of the Binary Big Bang.

AI is moving from the edges of insurance technology toward its core. As autonomous agents become more capable, insurers have an opportunity to redesign how products are built, risks are evaluated, claims are processed, customers are served, and decisions are made.

The organizations that embrace this shift will not simply have faster technology.

They could have a fundamentally different way of working.

The next chapter of insurance technology may not be about adding more software. It may be about creating software that can increasingly build, understand, and improve itself.

For decades, robots have largely existed within carefully controlled environments. They assembled products, transported materials, performed repetitive tasks, and followed precisely defined instructions.

That model is beginning to change.

The combination of large language models, advanced AI reasoning, and increasingly capable physical hardware is creating a new class of machines: generalist robots that can interpret instructions, understand their surroundings, adapt to changing situations, and perform a much broader range of activities.

In other words, AI is beginning to move beyond the screen.

It is gaining a physical presence.

For the insurance industry, this development represents more than another technological milestone. As robots become capable of interacting with people, property, workplaces, healthcare environments, and infrastructure, they will also interact with risk in entirely new ways.

The opportunity is significant. So are the questions.

From Programmed Machines to Generalist Robots

Traditional robots have typically been designed for a specific purpose. A robotic arm might repeatedly perform the same manufacturing task, while an automated vehicle might follow a predetermined route.

These systems can be highly effective, but their flexibility is limited.

Generalist robots represent a different approach.

Powered by increasingly sophisticated AI models, they can potentially interpret natural-language instructions, recognize objects, understand spatial relationships, respond to environmental changes, and determine how to complete unfamiliar tasks.

Imagine telling a robot to retrieve a particular item from another room. Instead of requiring a sequence of pre-programmed commands, the robot could interpret the request, locate the object, navigate its surroundings, avoid obstacles, and return with the item.

This ability to combine perception, reasoning, and physical action opens the door to a much wider range of applications.

Consider an autonomous mobility assistant operating in a busy public environment. It could potentially navigate around people, identify obstacles, respond to verbal instructions, and help someone reach a particular destination.

The machine is no longer simply executing a predefined task.

It is interpreting the world around it.

That shift has profound implications for insurance.

When Physical AI Becomes Part of the Risk Landscape

Every new capability creates a corresponding set of questions for risk professionals.

What happens when an autonomous machine makes an incorrect decision?

Who is responsible when someone is injured?

Does liability rest with the owner, manufacturer, software developer, operator, technology provider, or another party?

And how should responsibility be determined when several systems contribute to a single decision?

These questions are familiar from other areas of automation and autonomous technology, but physical AI introduces additional layers of complexity.

A robot operating in the real world can affect people and property directly. It may encounter situations its developers did not anticipate, interact with systems it was never specifically designed to work with, or respond to circumstances that fall outside its original training data.

As robots become more capable, insurers will need to understand not only what these systems are designed to do, but also how they behave when conditions change.

1. Robots Could Transform Risk Assessment and Claims

One of the clearest opportunities lies in property inspection, risk assessment, and claims management.

Generalist robots could potentially enter environments that are dangerous, inaccessible, or impractical for people.

After a natural disaster, for example, autonomous machines could enter damaged buildings, inspect infrastructure, capture images and video, and collect information without immediately exposing human assessors to hazardous conditions.

On construction sites, robots could monitor working environments and identify potential safety issues. In industrial settings, they could inspect equipment or hard-to-reach areas.

Even wearable robotic systems such as exoskeletons could support professionals performing physically demanding inspections or claims assessments.

The result could be faster assessments, richer evidence, and reduced exposure to dangerous environments.

But there is another dimension to consider.

When Machines Start Finding Patterns

Advanced AI systems can identify patterns that humans may overlook. That capability can be extremely valuable when assessing the cause or severity of a loss.

A robot could potentially combine visual information, environmental conditions, historical data, and contextual signals to form a view of what happened.

But insurers should not assume that an AI-generated conclusion is automatically correct.

Machine-learning systems can identify relationships that are difficult for humans to explain. They can also develop unexpected behaviors when exposed to new data or when multiple AI systems interact.

Research into phenomena such as unexpected or indirect learning in AI systems illustrates just how difficult it can be to understand every behavior emerging from complex models.

For insurers, this creates an important principle:

More data does not automatically mean better decisions.

The data generated by physical AI could eventually influence claims, underwriting, risk models, pricing, and product design. Strong governance, validation, human oversight, and clear accountability will therefore become increasingly important.

2. A New Workforce, and a New Workers’ Compensation Question

The impact of robotics will not stop with insurance companies themselves.

The businesses insurers cover are also likely to become increasingly automated.

Factories, warehouses, construction sites, logistics operations, healthcare facilities, and other workplaces may gradually integrate more autonomous machines into everyday operations.

This could reduce certain types of workplace risk while creating entirely new ones.

Robots might monitor working environments, detect unsafe conditions, or perform dangerous tasks that would otherwise expose employees to injury.

But what happens when an autonomous machine makes a mistake?

Traditional workers’ compensation and liability frameworks are built around human activity and relatively understandable chains of responsibility. Physical AI can introduce much more complicated relationships between employee, employer, machine, manufacturer, software provider, and operator.

As robots become more autonomous, insurers may need to reconsider how workplace risks are classified, monitored, and transferred.

The question may no longer simply be:

“Who was operating the machine?”

It may become:

“Who designed, trained, deployed, maintained, supervised, and ultimately controlled the machine’s behavior?”

That distinction could have significant implications for future insurance products and coverage structures.

3. An Aging Population Could Accelerate Robotic Care

Another major opportunity—and challenge—lies in healthcare and long-term care.

Longer life expectancy and changing demographic patterns are placing pressure on care systems in many parts of the world. At the same time, many healthcare organizations face shortages of skilled workers.

Robotic assistants could potentially help address some of these challenges.

Machines may eventually support patients with mobility, transportation, medication reminders, household tasks, monitoring, or other activities of daily living.

For families and care providers, this could offer valuable additional support.

But care environments involve some of the most vulnerable people in society, making risk management especially important.

What happens if a robotic assistant incorrectly interprets an instruction? What if a patient falls while being supported by a machine? What happens when a system encounters a situation it was never trained to handle?

These are not simply technical questions.

They are questions of responsibility, safety, accountability, and trust.

Insurance will have an important role to play in understanding these emerging risks as robotic systems become more integrated into care environments.

Cybersecurity Becomes Physical Risk Management

The more connected robots become, the more important cybersecurity becomes.

A compromised digital system can already cause significant financial and operational damage. A compromised physical system could potentially create consequences in the real world.

Imagine a connected machine responsible for moving people, inspecting infrastructure, assisting patients, or operating within an industrial environment.

A cybersecurity vulnerability could potentially become a physical safety issue.

This creates a convergence between cyber risk and physical risk.

Insurers will therefore need to consider questions such as:

• How securely are robots connected to external systems?
• Who can access their software and data?
• How are updates and patches managed?
• What happens if connectivity is interrupted?
• How quickly can a compromised system be isolated?
• Who is responsible for monitoring autonomous behavior?
• How is evidence preserved after an incident?

Cybersecurity can no longer be treated solely as an IT concern when software has the ability to control physical machines.

Responsible AI Moves Into the Physical World

AI governance becomes even more important when algorithms can directly affect people and environments.

Transparency, fairness, explainability, accountability, privacy, and human oversight are already central considerations for responsible AI.

Physical robots add another layer: real-world consequences.

An incorrect recommendation in a digital environment may require correction. An incorrect physical action could potentially result in injury or property damage.

That means insurers and the businesses deploying these technologies will need to think carefully about how AI systems are tested before deployment and monitored afterward.

Responsible AI should not be treated as a compliance exercise performed at the end of a technology project.

It needs to become part of the entire lifecycle—from design and training through deployment, monitoring, incident response, and continuous improvement.

Insurance Will Need to Insure the Transition, Not Just the Technology

The emergence of generalist robots creates an unusual situation for insurers.

They will not simply be insuring robots.

They will be insuring the new ecosystems created around them.

Manufacturers, software developers, operators, businesses, healthcare providers, infrastructure owners, technology platforms, and consumers may all become connected through increasingly autonomous systems.

This could create new forms of liability, new cyber exposures, new workers’ compensation considerations, new property risks, and potentially entirely new insurance products.

At the same time, the data generated by robots could improve the industry’s ability to understand existing risks.

The challenge will be finding the balance between using that information to improve decision-making and recognizing the uncertainty that comes with systems whose behavior may not always be completely predictable.

Preparing for a World of Physical Copilots

The emergence of generalist robots marks a significant shift in the relationship between humans and technology.

AI is no longer confined to applications that read, write, analyze, or recommend.

Increasingly, it can see, move, interact, and act.

For insurance, that means the future of robotics cannot be viewed solely as a technology story. It is simultaneously a story about liability, cybersecurity, workplace safety, healthcare, claims, underwriting, risk modeling, and customer protection.

The most important question may not be whether robots will become more capable.

It is how society, businesses, regulators, and insurers will adapt when machines become active participants in the physical world.

The opportunities are substantial: safer inspections, faster claims, better risk intelligence, additional support for workers and caregivers, and entirely new ways of managing complex environments.

But every new capability also introduces new uncertainty.

The insurers best prepared for this next chapter will need to do more than understand what robots can do today. They will need to continuously evaluate what these systems are learning, how they behave in unfamiliar situations, and where responsibility sits when something goes wrong.

When AI gets a body, risk gets a new dimension.

And insurance will be one of the industries responsible for understanding it.

The life and annuity industry experienced a period of exceptional momentum between 2022 and 2024. Strong sales, improving margins, and substantial capital flows created favorable conditions for insurers and encouraged continued investment across the sector.

But markets rarely stand still.

As conditions began changing, questions emerged about whether the strategies that worked during the recent growth cycle would remain effective in a more constrained environment. Lower interest rates, evolving customer expectations, regulatory pressure, technological change, and shifting distribution models are creating a different set of challenges.

For life and annuity executives, the next phase may require less focus on repeating the successes of the past and more attention to building businesses that can adapt to what comes next.

Here are eight strategic areas worth watching.

1. Rethink the Architecture of Insurance Products

The interest-rate environment can have a significant influence on the economics of life and annuity products.

When yields are attractive, relatively straightforward products may be easier to design and price competitively. When rates decline, however, insurers may have less room to offer compelling returns while maintaining sustainable economics.

That makes product architecture increasingly important.

Rather than focusing exclusively on individual products, insurers can explore solutions designed around broader retirement needs—including income stability, flexibility, liquidity, longevity protection, and growth potential.

The opportunity lies in creating products that work together as part of a larger financial strategy rather than treating each offering as an isolated transaction.

2. Build Connected Product Ecosystems

Customers rarely think about their financial lives in product categories.

They think about retirement income, savings, financial flexibility, and long-term security.

Insurers can respond by developing interconnected product ecosystems that address different stages and needs throughout a customer’s financial journey.

For example, growth-oriented products could potentially be combined with solutions designed to provide guaranteed income or liquidity. The value comes not simply from having several products available, but from making them easier to understand, combine, and manage.

Achieving this requires more than product development. It may also require integrated technology, consistent customer experiences, better advisor tools, and systems capable of connecting different parts of the insurance portfolio.

3. Move AI From Experiment to Infrastructure

Artificial intelligence is rapidly moving beyond pilot programs and isolated experiments.

Across the insurance value chain, AI can support underwriting, claims, customer service, distribution, operations, compliance, and product development. Generative AI is expanding what employees and advisors can accomplish, while more autonomous forms of AI could eventually perform multi-step tasks with limited human intervention.

But technology alone does not create transformation.

Insurers seeking meaningful value from AI may need to redesign processes, improve data foundations, establish appropriate governance, and prepare employees for new ways of working.

The question is increasingly shifting from “Where can we use AI?” to “How should the business be redesigned around what AI makes possible?”

4. Look Beyond Investment Performance

Investment expertise remains important, but long-term differentiation may depend on much more than investment performance.

Product innovation, actuarial capabilities, distribution, customer experience, technology, and operational efficiency can all influence an insurer’s ability to compete.

AI and automation may also create opportunities to rethink the underlying cost structure of the business.

The insurers that combine financial expertise with operational and technological capabilities may be better positioned to adapt as market conditions change.

5. Treat Regulation as Part of the Strategy

Regulatory expectations continue to evolve alongside changes in ownership structures, risk profiles, technology, and market practices.

Instead of treating compliance as a separate function that reacts to new requirements, insurers can integrate risk management into broader transformation efforts.

Modern stress-testing capabilities, stronger data infrastructure, automated monitoring, and AI-supported compliance tools can help organizations identify potential issues earlier and respond more efficiently.

A proactive approach can turn regulatory readiness into part of a company’s operating model rather than simply another layer of oversight.

6. Make Distribution More Focused

The insurance distribution landscape is becoming increasingly diverse.

Independent advisors, traditional agents, financial institutions, digital channels, and other distribution models can have very different needs and customer relationships.

Trying to serve every segment in exactly the same way may make it difficult to create meaningful differentiation.

A more focused strategy could involve developing specialized tools, experiences, and support for specific distribution channels.

For example, advisors may benefit from technology that helps analyze customer portfolios and develop personalized proposals, while other distribution networks may require different forms of training, technology, or sales support.

7. Orchestrate Capabilities Instead of Building Everything

Insurance transformation does not necessarily require every capability to be developed internally.

As technology evolves quickly, strategic partnerships can provide access to specialized expertise, platforms, data, and innovation without requiring insurers to build every solution from scratch.

The challenge is finding the right balance between internal capabilities and external partnerships.

Successful orchestration means knowing which capabilities are strategically important to own, which can be sourced externally, and how different technologies and partners can work together within a coherent operating model.

8. Reconsider the Mass-Market Opportunity

One of the industry’s biggest opportunities may also be one of its most difficult challenges: making sophisticated financial solutions more accessible to people with modest assets.

Large portions of the population approach retirement without sufficient financial preparation. Traditional advisory models may not always be economically practical for every customer segment.

Technology could change that equation.

AI-powered tools may help automate research, personalize education, simplify complex financial concepts, and support advisors serving a broader customer base.

The objective is not necessarily to replace human advice, but to make expertise more scalable and potentially more accessible.

Preparing for a Different Insurance Cycle

The next phase of the life and annuity industry may look very different from the conditions that supported the rapid growth of recent years.

If interest rates remain constrained, insurers will need to think differently about product design. If customers expect more personalized experiences, distribution models may need to evolve. If AI continues advancing rapidly, operating models and workforce skills will have to change alongside it.

The central question is therefore not simply how to maintain growth in a favorable market.

It is how to build an organization capable of competing when the market is no longer favorable.

That means connecting product innovation with distribution, technology with operations, and investment expertise with customer needs. It also means treating AI, regulation, demographic change, and retirement readiness not as separate trends, but as interconnected forces shaping the industry’s future.

The next chapter of life and annuity may not be defined by another boom. It may be defined by how effectively insurers adapt when the rules of the market change.

For many patients, the healthcare journey can become complicated long before a claim is ever submitted.

Imagine a policyholder who begins experiencing severe abdominal pain but struggles to secure a timely appointment. What could have been a straightforward diagnosis develops into a much longer journey involving repeated examinations, additional tests, extended hospital stays, and increasingly complex treatment decisions.

Along the way, inadequate pain management or unclear discharge instructions can add another layer of frustration. And when the final insurance reimbursement does not align with expectations or medical expenses, dissatisfaction can extend beyond the claims process to the entire healthcare experience.

This scenario illustrates a broader challenge facing health insurers: a claim is rarely an isolated financial transaction. It is part of a much larger healthcare journey.

When the Claims Experience Becomes the Customer Experience

Health insurance claims sit at the intersection of healthcare, technology, finance, and customer service. When these elements are disconnected, policyholders can experience delays, uncertainty, inconsistent decisions, and unnecessary administrative effort.

Research has shown that a meaningful share of consumers remain dissatisfied with their health insurance claims experiences, with dissatisfaction particularly pronounced in parts of the Asia-Pacific region.

One contributing factor is the continued reliance on legacy claims environments and traditional cost-management approaches. Important information may exist across different systems, documents, providers, and historical records, but claims professionals may still need to manually piece those fragments together.

The result can be a decision-making process that is slower and less consistent than it needs to be.

The industry is beginning to move toward AI-enabled claims operations, but adoption remains uneven. Many insurers are experimenting with generative AI for claims intake and related activities, while a much smaller proportion have successfully scaled these capabilities across their organizations.

This gap matters.

The difference between experimenting with AI and redesigning claims around AI can be substantial.

Modernizing the Claims Platform Is Only the Beginning

For insurers seeking to improve both customer experience and operational performance, claims modernization needs to go beyond faster processing.

Accuracy matters. Speed matters. Explainability matters.

But so does empathy.

A policyholder dealing with illness does not experience a claim as a data point. They experience it as one part of an often stressful personal situation.

Modern claims platforms can help insurers connect information across the healthcare ecosystem, integrate data from multiple sources, and support more consistent decision-making. When combined with stronger collaboration between insurers, healthcare providers, technology partners, and distribution channels, modernization can create a more connected journey from diagnosis through treatment and reimbursement.

The objective is not simply to process claims more efficiently.

It is to create an environment where the right information reaches the right person at the right moment.

The Rise of AI Agents in Claims

Generative AI introduces another opportunity: moving from systems that simply analyze information toward systems that can actively support and coordinate parts of the claims workflow.

This is where agentic AI enters the picture.

AI agents can be designed to perform specific tasks, interpret information, interact with systems, and make recommendations with varying levels of human oversight.

A useful way to think about an AI-enabled claims environment is through two complementary roles: Super Agents and Utility Agents.

Super Agents: Orchestrating the Claims Journey

Super Agents can support broader stages of the claims process, bringing together multiple capabilities within a single workflow.

They may assist with:

• Digital claims intake
• Case summarization
• Information verification
• Claims assessment
• Adjudication support
• Fraud, waste, and abuse detection
• Communication and workflow coordination

Rather than forcing claims professionals to move between disconnected tools, these capabilities can be brought together into a more coherent experience.

Utility Agents: Supporting the Details

Utility Agents can focus on narrower, specialized tasks that feed information into the wider claims process.

For example, they can help extract information from documents, validate data, identify inconsistencies, surface relevant historical information, and provide actionable insights to claims assessors.

Together, these two layers can help create a claims environment where AI handles repetitive information-intensive work while people remain responsible for judgment, oversight, and complex decisions.

AI Does Not Always Require a Complete Technology Overhaul

One of the most important opportunities presented by modern AI is its ability to work with information trapped inside existing technology environments.

Legacy systems remain deeply embedded across the insurance industry, and replacing them entirely can be expensive, disruptive, and time-consuming.

AI models can potentially help insurers extract, summarize, organize, and synthesize information from existing systems, allowing organizations to unlock more value from historical data without immediately rebuilding every component of their technology architecture.

This does not eliminate the need for modernization.

Instead, it can create a bridge between today’s technology environment and a more intelligent future claims operating model.

Connecting the Healthcare Journey From Online to Offline

Claims modernization becomes even more powerful when it extends beyond the insurer’s internal processes.

Consider the earlier patient scenario.

If relevant information had been available earlier, if healthcare access had been better coordinated, and if treatment decisions had been supported by connected data, the patient’s journey might have looked very different.

This points toward a broader concept: connected customer healthcare.

The healthcare experience should not begin when a claim is filed. It begins when a person first notices a health concern.

Insurers can contribute to this journey by strengthening relationships across healthcare networks and creating easier connections between digital services and physical care.

Mobile platforms, healthcare-provider networks, digital appointment services, diagnostics, and claims information can work together to give policyholders a clearer path through the healthcare system.

Distribution partners can also play an important role by providing human support when customers need reassurance, explanation, or guidance.

Technology can improve efficiency, but human interaction remains an important part of an empathetic healthcare experience.

From Treatment to Prevention

The opportunity extends beyond managing illness.

Health insurers can increasingly support preventive care by connecting policyholders with wellness resources, screenings, diagnostics, health-management programs, and other services.

Digital platforms can make these services easier to access across different stages of life, while partnerships with healthcare providers can expand the range of available options.

Integrated data can add another layer of value.

When claims information, electronic health records, health assessments, wearable-device data, and broader health trends can be responsibly connected, insurers may gain a more comprehensive understanding of emerging needs.

For customers, this could mean more relevant health insights and earlier opportunities to address potential concerns.

For insurers, it can support more personalized services, better-informed products, and greater visibility into healthcare costs.

The long-term opportunity is therefore not simply to pay for healthcare after something happens, but to become part of a broader ecosystem that supports healthier decisions before problems become more complex.

Building a More Empathetic Claims Future

The evolution of AI in health insurance is about more than automation.

At its most meaningful, it represents an opportunity to rethink the relationship between insurers, healthcare providers, technology, and policyholders.

Modern platforms can connect fragmented information. AI agents can coordinate repetitive and data-intensive tasks. Human professionals can focus on judgment and empathy. Healthcare partnerships can connect digital services with real-world care. Preventive programs can shift attention from reacting to illness toward supporting healthier outcomes.

None of these changes should be treated as a universal blueprint.

Every insurer has a different technology landscape, operating model, workforce, customer base, regulatory environment, and strategic priority. The right approach will therefore depend on the context.

But the direction is becoming clearer.

The future claims experience may be less about submitting information, waiting for a decision, and navigating disconnected processes—and more about creating a continuous, connected journey in which information, technology, healthcare, and human support work together.

The real opportunity is not simply to make claims faster.

It is to make them smarter, clearer, more connected, and more human.

Natural catastrophes are becoming harder to treat as occasional disruptions.

Floods, wildfires, storms, earthquakes, and other climate-related events are placing increasing pressure on communities, businesses, governments, and insurers. In the first half of 2025 alone, global insured catastrophe losses reached an estimated $84 billion, putting the year on track to become another in a growing run of years with losses exceeding $100 billion.

For insurers, this is more than a claims-volume problem.

It represents a fundamental shift in the underlying risk environment.

As the frequency, severity, and unpredictability of catastrophic events evolve, insurers are being forced to reconsider not only how they price and manage risk, but also how they support customers before, during, and after a loss.

The traditional insurance model has largely been built around a simple sequence:

Risk occurs → damage happens → claim is submitted → insurer pays.

That model is increasingly being challenged.

The emerging opportunity is to create something more proactive:

Understand the risk → help prevent the loss → respond quickly → support recovery.

When a Claim Becomes a Moment of Truth

Few interactions between an insurer and its customer are as emotionally significant as a catastrophe claim.

When a home is damaged by flooding or fire, a business is forced to close, or a family suddenly loses access to essential belongings, customers are not simply evaluating a financial transaction.

They are asking whether the organization they trusted will actually be there when they need it.

This makes claims a defining moment for the insurance brand.

A slow response, unclear communication, or complicated settlement process can turn an already difficult situation into a deeply frustrating experience. In an era of social media and immediate communication, those experiences can also quickly become public.

As a result, claims quality is increasingly a brand issue.

Despite significant investment in digital transformation and AI, improvements in several customer-experience measures have remained relatively modest in recent years. Some insurers have seen progress in customer satisfaction, but broader measures such as loyalty, effort, and long-term relationship value continue to present challenges.

The message is clear: improving the claims experience is not simply about operational efficiency.

It is about building lasting trust.

From Paying for Losses to Helping Prevent Them

One of the most important changes taking place across insurance is a shift from a payout mindset to a protection mindset.

Historically, insurers have primarily responded after an event occurred. Increasingly, technology allows them to intervene earlier.

Connected devices, predictive analytics, generative AI, agentic AI, satellite information, environmental data, and other technologies can help identify potential risks before they become costly claims.

Imagine a connected property where a system detects an electrical hazard before it triggers a fire.

Or a building where sensors identify a water leak before significant structural damage occurs.

Or a community where predictive models identify increasing wildfire risk and trigger preventative measures before flames reach vulnerable properties.

In each case, the insurer is doing more than preparing to pay a claim.

It is helping reduce the probability or severity of the loss itself.

That represents a fundamental change in the role insurance can play.

Three Ways the Claims Model Is Changing

1. From Reactive Claims to Proactive Protection

The first opportunity is to identify and address risks before they become losses.

IoT devices can continuously monitor properties and equipment. AI can analyze large volumes of information to detect unusual patterns. Predictive models can help identify emerging risks.

Together, these technologies can support earlier intervention.

For insurers, that may mean fewer severe claims and more efficient operations.

For customers, it can mean something even more valuable: avoiding the loss altogether.

The future of claims may therefore begin before a claim exists.

2. From Transactional Service to Customer Experience

Technology should not make the claims journey more complicated simply because the underlying systems are becoming more sophisticated.

Customers generally want the opposite: fewer steps, clearer communication, faster answers, and greater visibility into what happens next.

AI can help by summarizing complex claim information, identifying missing actions, routing cases, supporting employees, and giving customers more timely updates.

But technology is only part of the equation.

The best digital claims experiences should combine speed with empathy, automation with human judgment, and efficiency with transparency.

The goal is not to remove people from the claims journey.

It is to remove unnecessary friction so people can focus on the moments where human interaction matters most.

3. From Catastrophe Response to Catastrophe Resilience

Traditional catastrophe models have primarily focused on estimating potential losses.

That remains important, but the scale and complexity of emerging risks are encouraging insurers to think more broadly about resilience.

Instead of asking only:

“How much could this event cost?”

insurers can increasingly ask:

“What can we do to reduce the damage before the event occurs?”

This could include preventative property measures, automated alerts, environmental monitoring, physical risk mitigation, rapid-response services, and partnerships with organizations capable of acting on the ground.

Resilience can therefore become more than a pricing consideration.

It can become a source of product innovation.

Data Could Change the Way Insurers See Risk

The increasing availability of real-time and historical data is another major driver of this shift.

Property sensors, connected devices, satellite imagery, weather information, claims histories, customer interactions, and other data sources can provide insurers with a much richer picture of risk.

But data alone is not the answer.

The real value comes from turning information into timely action.

An insurer that knows a property is at elevated risk but cannot communicate with the customer or initiate preventative support has only partially solved the problem.

The future model will require stronger connections between data, prediction, decision-making, and action.

This is where AI agents and automated workflows could become particularly important.

Instead of simply identifying a risk, intelligent systems could potentially help initiate the next appropriate step—whether that means notifying a customer, escalating a case, coordinating an inspection, or supporting a claims professional.

Catastrophe Risk Is Also a Test of Operational Resilience

As catastrophe events become more frequent or severe, insurers may face sudden surges in claims volumes.

Thousands of customers may need assistance at the same time.

Traditional manual processes can struggle under these conditions.

Automation and AI can help insurers scale certain activities more effectively, from initial claims intake and document processing to case summaries, customer communications, and workflow management.

This can allow human teams to focus on complex cases while technology handles more repetitive tasks.

However, resilience also requires preparation.

Systems need to be tested under pressure. Data needs to remain accessible. Communication channels need to function during disruption. Employees need clear processes. Customers need reliable information.

A resilient claims operation is therefore not simply one that processes claims quickly.

It is one that can continue functioning when demand suddenly exceeds normal capacity.

Innovation Needs to Be Measured by More Than Technology

The insurance industry has been investing heavily in innovation, and evidence suggests that many initiatives are producing meaningful results.

Research has found that a large majority of innovation programs achieve or exceed their expected financial outcomes. Even more report progress against non-financial objectives such as customer engagement, satisfaction, brand strength, and employee experience.

This matters because the value of innovation cannot always be captured in a single financial metric.

A successful claims transformation may reduce expenses.

But it may also shorten customer wait times, improve employee productivity, strengthen communication, reduce preventable losses, and help customers recover more quickly.

Those outcomes are interconnected.

The New Claims Equation

The changing catastrophe landscape is forcing insurers to reconsider what a successful claims experience looks like.

It is no longer enough to simply calculate the loss accurately and issue the appropriate payment.

Customers increasingly expect insurers to help them understand risk, prevent avoidable damage, respond quickly when something happens, and guide them through recovery.

That requires a different model of insurance.

One built around prevention as well as compensation, prediction as well as reaction, and relationships as well as transactions.

The insurers that adapt successfully will not necessarily be those with the most technology.

They will be those that connect technology to a clear purpose: helping customers experience less disruption, recover faster, and feel supported when uncertainty becomes reality.

As catastrophe risk continues to evolve, insurance has an opportunity to become more than a financial safety net.

It can become part of the resilience system itself.

The future of claims is not simply about paying faster. It is about preventing more, responding smarter, and helping people recover with greater confidence.