I'll be quick this week as long as the U.S. seems to already be slipping into a looong Fourth of July weekend. I'm about to start packing my car and don't want to delay anyone else from doing so, too.

I just want to note that the "insurtech spring" possibility that I raised in a commentary in March seems to be well under way, based on recent funding rounds, a high-profile IPO and the continued stock market success of the high-profile, full-stack insurtechs. 

While I think my earlier analysis holds up pretty well, the more recent evidence points to some important trends about where insurtechs and, more broadly, innovation in insurance is heading.

First, the evidence:

• Pitchbook reports that VC investing in insurtechs surged 65% in the first quarter. Admittedly, that's up from a rather small base, but it still reflects enthusiasm, especially for innovation in claims and underwriting, after a long series of declines. Pitchbook singles out Assured, a claims automation platform that secured $23.3 million in a Series B round.
• Slide Insurance went public in June and quickly achieved a market cap north of $2 billion. It's a bit of a one-off, given that it's mostly picking through policies that Florida's Citizens Property Insurance wants to shed after years of having to support the state's dysfunctional homeowners market. The company is also controversial because its CEO and his wife, also a company officer, took home more than $50 million in 2024 at the startup, while Florida homeowners are struggling. Whether Slide has separated the wheat from the chaff remains to be seen, but the enthusiasm for an insurtech IPO is undeniable.
• The marquee names among insurtech carriers continue to do well after years of struggle. Lemonade stock price is almost 2 1/2 times what it was in October. Hippo is up 86% since then. Root's price is 3.2 times where it was in October. 

As I said in March, that AI is leading the current round of innovation means VC investments don't have to be huge. While those developing the large language models for generative AI will spend some $320 billion this year on capital expenditures -- Nvidia chips aren't cheap -- the rest of us get to basically plug into AI as we would a wall socket.  

Still, even if the insurtec investment numbers aren't overwhelming, my mantra for decades has been, "Nobody is as smart as everybody," so I believe that even the biggest, smartest insurance companies shouldn't assume they're going to get everything right in this fertile stretch. Venture capitalists certainly won't assume that, so they're going to finance startups that seem to have an unusual insight into a market or a technology. The smart insurers will keep their eyes wide open and buy from, partner with, or simply acquire those with promising propositions that internal teams didn't pursue. 

Pitchbook is right to highlight claims and underwriting. Generative AI can provide huge gains there, starting with operational efficiency and moving well beyond. 

I'd add agencies and brokerages to that list. They're already using generative AI to radically improve productivity and are branching out into agent copilot sorts of scenarios, but I'm sure some smart entrepreneurs have ideas to add to the mix. 

I imagine parametric and embedded insurance will see considerable insurtech innovation, too. Some of the innovation may have to happen at the incumbents. The appeal of parametric, for instance, is often as part of a stack -- a policyholder wants to get some money immediately, knowing that working through a claim on, say, crop failure will take a long time -- and major carriers are well-positioned to offer that sort of coverage. But insurtechs will surely find ways to augment what carriers are doing both with parametric insurance and with insurance embedded into purchases that the incumbents don't currently recognize as opportunities. 

So here's to an insurtech summer -- just with less heat and humidity, please.

Now I'm off to pack my car on my way to what should be a great time with my many siblings and our families at the New Jersey shore. I hope you have a great loooong weekend, too.

Cheers,

Paul 

P.S. If you're interested in a deeper dive on the rationale for an insurtech boom, here is a very smart piece by Teddy Himler of Optimist Ventures, who is the one who put me on to the "insurtech spring" idea in the first place. 

Paul Carroll

I’ve been intrigued by the Internet of Things since I first heard the term, maybe 15 years ago. How have its capabilities and uses evolved in the insurance industry?

Nga Phan

IoT has become one of the major trends driving technological innovation, and the intersection with insurance has been particularly fascinating, as it addresses the longstanding challenge of identifying concrete, tangible value from IoT devices.

At Bolt, we've found that IoT integrated with insurance provides that valuable opportunity. We've implemented water sensors with HO3 insurance, coupled with comprehensive prevention and protection services. This approach shifts the paradigm from simply repairing or replacing after damage occurs to preventing and protecting against potential issues. This represents a fundamental change in how insurance can work in the connected age.

Paul Carroll

I think there’s a ton of potential for sensors that can alert property owners to leaks and perhaps even automatically shut off water. How does your program work?

Nga Phan

We focus on water damage specifically for homeowners insurance. This involves attritional water losses – damage that occurs due to water issues.

Water damage accounts for approximately 40% of the premium, which many people may not realize. Once water damage occurs, it can actually represent the majority of losses for a home or property.

Paul Carroll

How have water sensors developed over the last five to 10 years, and what advancements do you anticipate in the next three to five years?

Nga Phan

Sensors for water leaks and other applications continue to improve each year. They are more powerful, cheaper, and easier to handle, deploy, install, and extract data from. They're also easier to replace and integrate with other workflows within homes or with insurance carriers.

These advancements collectively ensure that prevention technology becomes more scalable. For our program specifically, the decreased cost and simplified deployment means more homes can access this technology. Homeowners are more compliant in deploying sensors and maintaining them throughout the policy life.

This creates opportunities to integrate with more devices and incorporate their data directly into policy underwriting. We're witnessing the scalability of this technology and program right now, benefiting from many years of advancements in IoT technology. 

Looking forward three to five years, devices will likely become even smarter while continuing to decrease in cost. They will probably collect more diverse data about the home, expanding the scope of monitoring beyond leak detection to include water flow and humidity tracking—essentially capturing leading indicators of potential loss events.

Future devices will transmit data more in real time and in larger volumes, leading to better insights into water flow patterns or leak event patterns. This will enable more comprehensive preventative programs for carriers. By analyzing their portfolio data, they might see that properties of certain types or ages or with specific pipe types are more prone to particular kinds of damage.

Carriers could also combine this rich sensor data with environmental metadata. Homes in specific areas of Delaware, Florida, or elsewhere might show certain damage patterns that can be prevented. The combination of more powerful sensors collecting and sharing more data in real time with other data types will provide opportunities to apply intelligence toward more preventative programs.

Paul Carroll

Help me visualize this. What does participation in your program look like from the homeowner's and carrier's perspectives?

Nga Phan

Let's say you own a home in Florida. We know that in Florida it is really hard to get homeowners insurance. Having a program like this, which reduces attritional water damage, gives carriers more options to provide coverage within that area. As a homeowner, I now have access to homeowner insurance even in places where it's hard to find.

Secondly, because of this program, I am able to receive reduced rates, making protection more affordable for me as a homeowner. Because deployed sensors must be kept active, I know that I am being actively protected and that big losses will be prevented due to the alerts I receive and the associated emergency response provided as part of the program.

If a loss does occur, because of the alerts and prevention measures, that loss will probably be smaller than it would otherwise be. That's another benefit to me as a homeowner.

To the carriers, there are several advantages. They continue to be able to participate in markets they might otherwise avoid because they can now make it economical to cover these areas. This means they've got larger access to potential new policyholders, which is a way to drive growth.

We’ve seen a significant reduction in losses, with up to 55% total premium impact. More than 40% of that comes from avoided loss events, and the remainder is driven by reduced severity, which can be as much as 28%. These are tangible, verified results that carriers can expect when incorporating this program into their underwriting strategy.

Carriers are also going to be viewed as more innovative, improving their brand image and relationship with customers because they're now participating in prevention. This translates into a better customer experience overall and will lead to better renewal rates and retention for the carriers.

Paul Carroll

What does the home water monitoring system look like physically, and how does the information flow from the sensors to insurance carriers? The ones I’m most familiar with look sort of like hockey pucks and can be set down anywhere that there might be a leak.

Nga Phan

That is exactly it. The sensors are easy to set up with just a few clicks, thanks to the WiFi enablement and an app that provides real-time, interactive guidance.

The system creates a regular stream of data being shared with the manufacturers through the apps and WiFi network. This data is also shared with the carriers.

On the back end, all these events and data are integrated into the policy administration system and underwriting workflows for the carriers in real time.

One of the most compelling things is that carriers can be confident about compliance. Before, if a carrier offered a discount for having a water sensor, they had no way to verify if customers were actually using it, keeping it updated, or ensuring it worked effectively.

With this program, carriers can confirm the sensor is active, in use, and doing its job. This visibility is enabled by our direct data feed that connects all components in the system. The data flows from the sensors to the apps, device manufacturers, our platform, and ultimately into the carrier's operations. 

The program has proven to be economical for carriers to invest in. Because carriers experience lower loss ratios, they are able to realize significant savings. Even when they pass these savings to insureds in the form of discounts and cover the cost of the devices, they still come out ahead.

Year one is typically about breaking even. However, starting from year two and throughout all subsequent years, carriers actually experience a positive return on investment.

Paul Carroll

If carriers can provide these sensors for free and still come out ahead, that’s a big change. A few years ago, the economics only worked if the cost of deployment was matched against the whole range of costs from a loss – the cost to the insurer, the deductible paid by the policyholder and the hassle factor of recovering from major water damage.  

Nga Phan

Yes, we've reached a tipping point. The economics have become positive, allowing carriers to subsidize the device cost, which is crucial for scaling this technology.

Paul Carroll

Some leak prevention systems use automatic water shut-off valves, but they’re expensive, especially if you’re retrofitting a building rather than including them in the original construction. Where do you stand on shut-off valves?

Nga Phan

Today, we have not incorporated shut-off valves, but that’s actively being researched by us for the proper adoption and economics. The technology around that was still expensive a couple of years ago, which would have made the economics more challenging and prevented the broad rollout of our program.

That's why we went with water leak sensors first. It's important to prove out the concept. 

Now that we have proven the economics, I think it will be very powerful to combine the current approach with shut-off valves in places where it makes economic sense.

Paul Carroll

What is the current deployment scale of your program, and what expansion plans do you have for the coming years?

Nga Phan

Currently, we have 25,000 devices deployed across our network. As I mentioned, these installations have resulted in a 55% pure premium reduction for carriers driven primarily by avoided loss events, which account for over 40% of the reduction, with the remainder coming from reduced severity when losses do occur. These are verified results that we’re confident can scale across more carrier partners. Our immediate focus is bringing more carriers into the program. We're also expanding our sensors to monitor additional types of events, including freezing, fire, and theft.

Commercial property represents another significant area of expansion for us. We began with attritional water damage and have proven our model's effectiveness and scalability. This established playbook now allows us to extend into other event categories and implement additional types of sensors.

Paul Carroll

A device made by Whisker Labs, called a Ting, prevents fires by monitoring for electrical problems in a building. As it’s been widely deployed, the device has been shown to be able to spot problems in the grid, too, including ones that have led to major wildfires. Could broad deployment of water sensors provide the same sort of benefits to the whole system?

Nga Phan

I think so. With water, understanding the flow for all homes in a particular area enables detection of surges and allows for implementation of region-wide alerts.

While responding to one house at a time might be manageable, addressing multiple homes simultaneously presents challenges. Anticipating potential issues affecting numerous households allows for more efficient mitigation actions. You could analyze correlations between outside temperature and the temperature of main pipes across homes. This application of big data helps identify larger risks affecting a greater number of properties.

By analyzing patterns from everywhere—combining factors like pipe age, temperature, and humidity—we can predict the likelihood of incidents before receiving an actual leak alert. This represents a shift from macro to micro focus, enabling more personalized risk prediction for specific properties.

Paul Carroll

I assume the learnings could then be fed back to municipal regulators and builders to improve construction practices.

Nga Phan

Absolutely. The potential applications are significant for the industry.

We are extremely excited about this initiative. We embarked on this journey as a proof of concept because we had the intuition that it would provide meaningful impact. Now we have proven it works.

The economics and the scale are there. The program is real and tangible, and both the insured and the insurers are reaping the benefit. We cannot wait to scale this up with more carriers and more end consumers.

Paul Carroll

That sounds great. Best of luck. 

About Nga Phan

As head of product, Nga is responsible for leading bolt’s product development in support of the company’s long-term growth plans. Her role is instrumental in steering the strategic direction of the company’s product offerings, focusing on innovation and enhancing the insurance market. Nga has nearly two decades of experience in executive leadership roles and advising C-suite leaders at enterprise SaaS companies, including Salesforce, ServiceNow and Risk Management Solutions. She also spent time at Bain & Co., where she contributed significantly to the telecommunications, media and technology practice. A transformational leader in the technology and insurance sectors, Nga’s deep expertise in product management, strategic planning, and operational leadership has led her to make significant contributions across organizations, having driven product innovations that transformed customer service experiences, including leveraging generative AI. Nga holds an MBA from MIT Sloan School of Management and a BA in economics from Bates College. She lives in the San Francisco Bay Area and enjoys exploring the outdoors and spending time with her family, which includes two canine children.

As macro volatility, climate events and litigation risks continue to challenge pricing stability, insurers are responding with a sharper lens and rewarding the well-prepared.

This is the era of precision risk, where even two companies in the same sector can see drastically different renewal outcomes depending on how well they present, quantify and control their exposure. Rising capacity and heightened carrier competition — particularly in property and cyber — have started to temper rate increases, especially for insureds with robust risk management programs.

The commercial insurance landscape showed mixed signals in the first quarter of 2025. Rate guidance ranged from -20% to +15% across major lines, with results largely influenced by risk profile, loss history and industry classification. Certain lines — specifically cyber, workers' compensation and D&O — saw decreases. Others — notably commercial auto and umbrella liability — encountered capacity constraints and sharp price increases, particularly for higher-risk clients.

Consider cyber: Organizations with strong controls, documented internal processes and alignment between IT and risk teams are seeing more favorable renewals, with modest rate decreases of up to 10% and in some cases more substantial decreases (over 50%), depending on how the organization was presented to the marketplace in prior years. Underwriters are rewarding maturity, not just software. Companies that pair solid cyber hygiene with sophisticated modeling and vulnerability scoring have more options, like higher sublimits and broader terms.

Meanwhile, lines like commercial auto and umbrella and excess are less forgiving. First quarter rate increases in these segments typically ranged from +5% to +15%, with steeper spikes in high-risk sectors like transportation and construction. In these lines, a weak risk story — whether from outdated safety protocols or limited data — almost guarantees above-average increases.

Underwriters Want Answers, Not Just Applications

For the policyholder, the message is clear: Insurers expect more. Not more paperwork, but more insight. They're asking:

• How current are your property valuations?
• What business continuity plans back your business interruption limits?
• Can your safety programs be backed by outcomes, not just intentions?
• How are you addressing location-based CAT exposure or litigation risk?

Insurers are under pressure to improve underwriting performance, control loss ratios, and manage capacity. Years of continuing catastrophic weather losses — like this year's California wildfires, expected to result in as much as $50 billion in insured losses — have made profitability harder to maintain. Add to that social inflation, unpredictable litigation and widening gaps in coverage, and underwriters can't afford to treat all submissions equally.

They are triaging submissions more aggressively and rewarding those who come prepared with well-supported documentation, including safety protocols, updated valuations, business continuity and disaster recovery plans and clearly articulated risk mitigation efforts. This is particularly true in distressed or volatile sectors like transportation, real estate and healthcare, where the gap between well-managed and high-exposure accounts continues to widen.

The Well-Prepared Renewal Playbook

In a segmented market, what you do between renewal cycles matters as much as what happens during them. Risk leaders seeing the best outcomes in 2025 are doing the following:

• Start early. Starting the renewal process early creates the breathing room to shape a stronger outcome. It allows time to revisit property valuations, update exposure data, clarify loss narratives and address underwriter questions. Even 60 to 90 days earlier can make a meaningful difference in how your submission is received.
• Quantify the right data. Underwriters are applying stricter triage, prioritizing submissions backed by meaningful metrics, including CAT modeling, updated total insured values (TIVs), measurable risk control strategies backed by loss data, cyber risk scoring and business interruption calculations. Accounts with strong safety protocols, up-to-date valuations and clear mitigation stand out.
• Align across departments. The strongest risk presentations reflect input from across the business, including IT (for cyber), HR (for EPL and benefits-related exposures), finance (for valuation and liquidity) and operations (for safety programs and continuity planning). Businesses that bring cross-functional alignment to the table are showing insurers that they've done their homework and that their insurance strategy is backed by broader operational readiness.
• Be open to structural changes. Nuanced markets require flexible thinking. Structures like captives and parametric solutions are no longer reserved for massive enterprises. They're viable tools for managing cost, volatility, or coverage gaps, especially in CAT-exposed or litigation-prone lines.
• Set internal expectations. Even well-managed risks may face increases in high-pressure lines like commercial auto or excess liability. Historical data, market insights and program benchmarking can be used to explain why a change is happening to stakeholders, and how your strategy is mitigating its impact.

2025 Is a Test and a Turning Point

This year won't reward the default submission. It will reward organizations that demonstrate control, strategy, and responsiveness. C-suites and risk managers who treat insurance as a year-round process instead of a last-quarter task are finding themselves in stronger negotiating positions, with more stable programs.

The gap between well-managed and distressed risk is widening. It's no longer just about loss history. It's about how effectively the business tells its risk story, how well its internal systems support that narrative and how early it engages with the market.

Across every line of coverage, the first quarter of 2025 showed that organizations with accurate valuations, active mitigation efforts and a clear risk strategy are outperforming peers. As the year progresses, volatility will remain a constant. But for the well-prepared, so will opportunity. In this market, readiness is leverage.

The property insurance market has reached a point of stabilization after years of volatile rate hikes and financial losses. However, external threats—especially those posed by natural disasters—continue to challenge property owners and insurers alike. The increasing frequency and severity of extreme weather events, coupled with inflationary pressures and shifting geographic risk zones, have made risk management more critical than ever.

Managing property risk is no longer about reactive solutions; it requires a proactive, data-driven strategy. From accurate property valuations to leveraging advanced event monitoring technologies, advisors must encourage property owners and businesses to take decisive action to mitigate vulnerabilities and secure insurability. 

Here's what your property stakeholders need to focus on to navigate the evolving risk landscape.

Accurate property valuation: A foundation for proper coverage

One of the most significant issues to emerge in property insurance over the past several years is undervaluation. Many properties have been historically underinsured, leading to inadequate coverage when disasters strike. In high-risk areas, rising property values have made this problem even more pronounced.

For instance, a building insured for $3 million may now require $4 million-plus for reconstruction due to inflation, supply chain disruptions and escalating costs of materials and labor. Without regular property valuations, policyholders may find themselves facing significant out-of-pocket expenses if a claim exceeds their coverage limits.

To mitigate this risk, advisors should instruct property owners to:

• Conduct property valuation reviews and updates to reflect true replacement costs
• Work with insurance brokers and valuation experts to determine accurate market rates and trending factors
• Maintain clear documentation of valuation methodologies to justify insured values

By staying ahead of valuation shifts, property owners can reduce the likelihood of cost increases while ensuring they have adequate coverage in place.

The rising threat of natural disasters and secondary perils

Catastrophic weather events are reshaping risk landscapes across the U.S.. In 2024, there were 27 climate-related events that each exceeded $1 billion in insured property losses. Regions once considered safe are now experiencing severe storms, wildfires and flooding, making preparedness a necessity rather than an option.

Tornado Alley, for example, has shifted eastward, while cold snaps now extend into Southern states. Florida and California continue to face increasing risks from hurricanes and wildfires, respectively, exacerbating insurance costs in those regions.

Given this, advisors need to encourage property owners to integrate advanced risk modeling and predictive analytics into their disaster preparedness plans. These tools can help assess potential vulnerabilities and guide risk mitigation efforts, such as:

• Reinforcing building structures to withstand harsh weather
• Upgrading roofs, windows and drainage systems to minimize storm damage
• Implementing wildfire-resistant landscaping and maintaining defensible perimeters

For property that constitutes a business and generates revenue, continuity insurance and a disaster recovery plan are paramount.

The role of smart technology in risk prevention

The emergence of smart home and commercial property technologies is changing how insurers assess and price risk. Property owners who invest in these solutions are more likely to secure favorable coverage terms and lower premiums.

Some of the most effective smart technologies include:

• Water leak detection systems: Water damage is one of the most common non-catastrophic claims, with leaks from plumbing failures or faulty appliances accounting for billions in losses annually. Smart sensors can detect abnormal water flow and alert homeowners before minor leaks escalate into major damage.
• Security and surveillance systems: Many burglaries occur without forced entry, often due to unlocked doors or weak security measures. Smart locks, surveillance cameras and alarm systems reduce theft risk and provide insurers with verifiable proof of security measures.
• Fire and electrical monitoring: New technology can now detect electrical arcing behind walls, a common precursor to house fires. Advanced remote fire monitoring and suppression systems can reduce both loss frequency and severity.

Insurance providers increasingly require these measures as eligibility criteria rather than the basis for optional discounts. As a result, it's important that property owners are aware that failing to adopt risk-mitigation technologies may make it harder to obtain affordable coverage.

The growing cost of insurance and alternative risk solutions

Homeowners and businesses alike are grappling with rising insurance premiums, particularly in catastrophe-prone areas. In Florida, for example, the average annual homeowners' insurance premium is four times the national average, making it difficult for many property owners to afford coverage.

In response, advisors should encourage clients to consider alternative risk solutions, including:

• Captive insurance: Business owners may explore forming captive insurance entities, allowing them to assume some or all of their own risks in a structured and financially beneficial manner over time.
• Parametric insurance: Unlike traditional insurance, which reimburses losses after a claim process, parametric policies provide predefined payouts when certain conditions—such as wind speeds or earthquake magnitudes—are met.

These alternative solutions require careful planning and financial analysis, but they can provide viable options when traditional insurance becomes cost-prohibitive or unavailable.

Modern risk management

Managing property risk in 2025 requires a multi-faceted approach that goes beyond traditional insurance. The increasing frequency of natural disasters, rising insurance costs and emerging global threats demand strategies to safeguard client homes, businesses and assets in an unpredictable world.

By prioritizing accurate property valuations, investing in smart risk mitigation technologies and exploring alternative insurance options, property owners can position themselves as lower-risk policyholders. This not only enhances their insurability but also ensures long-term financial stability.

Now that Elon Musk has finally begun a robotaxi service, after nearly a decade of hype, fans are declaring that we've entered a whole new world that will, among many other things, mean the phaseout of personal auto insurance. 

The stock market seems to agree that what one analyst calls the "golden age of Tesla" is here: It carries a market cap of $1.09 trillion, and analysts say some 75% of that valuation stems from projections about the company's driverless technology. 

I'm unconvinced, including for a reason that has been ignored in all the coverage I've seen. 

I'll explain. 

Let's start with the piece I think has been overlooked: the operational complexity of running a network of what Musk has said will be millions of Teslas owned by individuals who have made them available to function as robotaxis. 

Even if you assume that Tesla's autonomous driving software works perfectly (which I'm not at all prepared to do), ponder for a minute all the work that has to go into managing a network of millions of cars that you don't own and have to essentially borrow from their owners. 

I spent several months doing that sort of pondering with two colleagues as part of a consulting project in 2017 for a major company that was considering a big move into AVs, and potential problems popped up all over the place. 

If a car needs to be recharged while in service, where does it go? Who plugs in the charger? How do you position cars so they can get to those hailing them as quickly as possible? How does rush hour complicate that positioning? How do you know when a kid covered in sand from a trip to the beach has shed in the car, or when a couple has sex in the back seat? Who cleans the car? What if it's used in a drug deal or other crime? How do you keep a group from carjacking the AV by having someone stand in front of it and behind it, knowing the car won't run them over? And on and on and on.

And those were just the problems facing a company that would have owned all its robotaxis. What Musk is promising is far more complicated.

What if I've declared my car available as a robotaxi for a stretch but want it back? What if I don't maintain my car as well as I should? How does Tesla enforce standards? Who is responsible for the depreciation on my car based on the miles spent to position it for ride hailers? How does Tesla deal with the fact that most rides are hailed during the same times of the day when owners are most apt to use them?

Basically, Tesla will have to build the equivalent of an air-traffic control system but on a far bigger scale. Musk is talking about coordinating tens of millions of rides per day, vs. 45,000 plane flights a day in the U.S. Airplanes travel point to point at assigned times, while Teslas will have to be able to go anywhere at any time. And automatic pilot software on airlines doesn't have to worry about pedestrians and bicyclists. 

That system can be built, but it won't be easy and will take years to develop. 

That timeline, alone, argues for dismissing the hyperbole about imminent disruption to personal auto insurers.

There's more, too. Tesla is years behind the competition. While Musk has long pointed to the inadequacies of competitors' capabilities, the Tesla rollout puts it about where Google's Waymo was in 2017. If you want to be charitable, you could say Tesla is where Waymo was as recently as 2020. 

Tesla has about 10 robotaxis that have been operating in the wild since Sunday, while Waymo has more than 1,500 and has been offering autonomous rides for years — Waymo operates about 250,000 paid rides per week. Waymo's rides are fully autonomous, while Tesla has a monitor sitting in the passenger seat. The informed speculation is that Tesla also has someone remotely monitoring each car, with the ability to intervene if it makes a mistake; Tesla hasn't commented, as far as I can tell. Waymo operates in large areas of many cities, while Tesla is limited to a small section of Austin, Texas, away from downtown and other complexities (even though Musk has long said a car can't be considered autonomous if it's limited to a "geofenced" area that has been carefully mapped).

As I've written previously, I believe Tesla also faces an insurmountable technology problem. Musk took the cheap route on sensors, relying only on cameras, while Waymo and other competitors also use lidar, radar and high-definition mapping. 

Musk fans argue that he has a massive data advantage because he has so many cameras on the road capturing data and has been running what is essentially a huge pilot for years, based on drivers who used early versions of his autonomous software. But early results haven't exactly been promising. Tesla vehicles being operated by that software have been involved in quite a number of accidents, including fatal ones, and Tesla's defense has been that it told drivers they couldn't trust the software. 

Videos from the robotaxi rollout show a Tesla stopping abruptly twice for no apparent reason, once in the middle of an intersection. (Apparently, the robotaxi saw police cars in a parking lot adjacent to the roadway.) Other videos shared by riders showed robotaxis speeding (though moderately) and, in one case, getting confused at an intersection and driving into a lane for oncoming traffic. 

That's just in the past two days, with only 10 cars on the road, operating in a tiny, uncomplicated, well-mapped area between 6 a.m. and midnight. And those sharing the problems are hardly critics; they were among the big fans selected by Tesla for early rides.

I'm not saying the robotaxi launch was a disaster. It wasn't. As Reilly Brennan, a prominent analyst in this space, says, "This wasn't the unsupervised dream realized, but if it wasn't Tesla I think people would be saying this was a solid mini launch."

Some go much further, especially on Wall Street, as this article in Quartz details. (If the surname in the byline looks familiar, that's because my daughter wrote it.) Goldman Sachs went on at length about the impact on insurers, saying auto insurance costs would plunge by 50%.

But some are even more cautious than I am, as the Quartz article also details. And if you want to get the really negative view, read this piece in Forbes, which dismisses the Tesla robotaxi as "not ready."

I come down on the side of Amara's law, as I usually do. It observes that technologies tend to be overhyped in the short term but underhyped in the long term. I firmly believe in the long-term prospects for autonomous vehicles, which will, among many other things, shift auto insurance from individuals to the makers of the software and the operators of the vehicles. But I see no need to hyperventilate about Tesla's robotaxi mini launch, no matter what Wall Street says.

As Phil Koopman, an expert about AV safety, wrote: 

"This is an important first step for Tesla on the road to robotaxis. But as other companies have learned, this is the end of the beginning, and there is a very long road ahead to scale up to a viable product."

Cheers,

Paul 

The one-quarter mark of the 21st century offers us a chance to reflect on the past – assessing the technology that has transformed insurance, recognizing how our industry has changed in the digital age, and striving to understand what the rise of insurtech, AI, and automation have taught us.

It's also an opportunity to look forward, using the insights of the past to better understand where the industry is headed in the future.

Over the past decade, digital transformation has redefined nearly every facet of insurance, from policy underwriting and distribution to the way claims are processed and managed. This rapid digitalization has primed insurers to embrace deeper cross-industry collaboration, adopt smarter and more streamlined claims processes, and balance innovation with compliance and customer trust.

As insurers look to predict the state of their industry by 2030, they're now well equipped to ride the next wave of disruption – whatever it may be. Those that invest in intelligent automation, personalized engagement, and flexible operating models will be best positioned to thrive in a dynamic and data-driven marketplace.

AI: A New Operating Model

By leveraging AI's ability to automate tasks, personalize offerings, and streamline workflows, insurers now operate in a reality where underwriting is up to 36% more efficient, customer service teams are over 30% more productive, and claims are processed up to 50% faster. Consider that 70% of simple claims are now resolved in real time – a rate of service that might have seemed unimaginable merely a decade ago.

On the sales side, AI is helping carriers boost conversions and cut acquisition costs, while freeing agents and brokers from time-consuming admin so they can focus more on providing the human touch so many clients desire.

That's just the beginning. AI and machine learning are maturing rapidly, poised to become even more foundational to underwriting, pricing, fraud detection, and service. These tools can process large amounts of data, assess risk, and design products, allowing insurers to optimize workflows across the value chain, enable seamless claims processing, automate policy issuance, and reduce administrative overhead. Personal AI assistants are also gaining momentum, guiding customers through purchasing decisions, answering policy questions, and offering recommendations based on life events or behavioral data, all in real-time.

These technologies won't just improve efficiency: They'll reduce costs, improve accuracy, and allow human teams to focus on higher-value interactions.

Customer-Centricity and Personalized Offerings

By 2030, successful insurers will be able to anticipate customer needs and deliver frictionless, individualized, omnichannel experiences. Leading insurance carriers are already prioritizing customer-centric approaches, developing innovative operating models that fundamentally drive value for insurers while constantly improving the way they interact with policyholders.

When insurers incorporate AI into claims processes, they not only enhance efficiency but infuse greater empathy across the customer journey. AI-powered systems can now analyze tone and sentiment in real time – such as detecting distress in phone callers – allowing insurers to ensure more responsive and compassionate customer interactions. Additionally, by freeing human agents from menial tasks, AI allows them to spend more time serving customers with the care and attention they deserve.

Insurers will also continue using data to offer more tailored policies. Consider the hyper-personalized underwriting possibilities of auto insurance based on driving scores. A 2024 survey from CMT, for example, found that 92% of respondents either agreed or strongly agreed with the statement that "Every driver in the U.S. should pay insurance based on how safely they drive."

The result isn't just greater customer satisfaction. It amounts to deeper, more profitable relationships across a lifetime of engagement.

Unlocking Speed and Resilience

Data-driven decision-making, supported by cloud-native infrastructure and AI, are allowing insurers to reduce the process of testing and launching products from months to mere weeks. This level of automation will streamline service, reduce cycle times, and ensure dynamic compliance with evolving regulatory standards.

It can also enhance workforce capabilities, re-skilling employees to work alongside AI tools and take on more consultative roles. A dual investment in both technology and the people who know how to use it will be key to fostering resilience, innovation, and long-term growth.

Insurance will also continue to be deeply interwoven with adjacent sectors, from banking and healthcare to energy and transportation. By integrating and partnering with insurtechs, digital platforms, mobility providers, and health services to expand value beyond the policy itself, carriers will deepen their ecosystem partnerships and expand what they can offer to customers. As enablers of long-term resilience against risks of any kind, insurers can reinforce trust and stability in an era of seemingly endless disruption.

Ensuring a Safer Future

Incremental adjustments won't be enough to keep carriers competitive.

The winners will be those that embrace change not as a threat, but as an opportunity to redefine their role – from reactive risk managers to proactive partners, dedicated to adopting whatever technologies and strategies are necessary to improve the lives of their customers.

In January 2025, fires in Los Angeles killed at least 29 people and destroyed over 18,000 homes and buildings, marking one of California's worst disasters.

Insurance offers one mechanism to support individuals who have suffered catastrophic losses, but insurance companies are now more cautious than ever when covering wildfires and other weather-related risks. Last year, insurers worldwide paid out more than $140 billion in claims relating to natural catastrophes, the fifth consecutive year with losses exceeding $100 billion.

Major natural disasters that cause substantial insurance claims, including storms, wildfire and flooding, are anticipated to become more severe with climate change.

But even for those perils for which recent spikes in payouts have been down to factors such as economic inflation and population growth, climate change still provides an unwelcome boost to risk and, therefore, premiums.

Typically, insurers used past claims to predict future losses from the same perils. So long as there had been no major unexpected disasters or big shifts in risk or exposure in a specific area, insurers could be confident that the premiums paid by the many would be enough to finance the claims of the few.

But in 1992, destruction from Hurricane Andrew in the Gulf of Mexico exposed the fragility of that approach by causing insured losses three times higher than expected by industry insiders in Florida alone. Insurers switched to sophisticated catastrophe models — tools that combine the physics of specific natural hazards with details about building construction and insurance information — to estimate possible financial losses. And in the past few years, as the imprint of global warming on natural hazards has become more obvious, the insurance industry has been recruiting climate and Earth scientists to reduce the likelihood of being taken by surprise.

Today, insurers have more-realistic views of their customers' exposure to weather- and climate-related risks and the scale of potential claims. Inside the limits set by government regulators, insurers decide how much risk they can tolerate across their portfolio, raise premiums for owners of more-exposed homes and purchase reinsurance to prepare for losses larger than they could normally afford.

Bankrupt insurers can't pay claims. Policyholders therefore benefit from risks being estimated correctly so losses can be paid by charged premiums. But insurers also use the latest catastrophe modeling and climate science to justify higher premiums, which are fast becoming unaffordable. Too many people are forced to choose between paying more for the same insurance, accepting lesser coverage to keep premiums manageable or letting their insurance lapse.

Continuing to ratchet up insurance premiums to keep up with mounting losses from hurricanes, wildfires and other perils might make financial sense — but it is also indicative of a growing crisis facing the industry. In many countries, including the U.K. and the U.S., the average cost of home insurance is more than twice what it was a generation ago. Yet, paying higher premiums doesn't make homeowners any more prepared for disaster. And despite charging higher prices, many insurers are still losing money through home insurance because they are paying out record-setting claims relating to natural catastrophes.

When insurers can no longer afford payouts to high-risk properties in a region, or even entire jurisdictions, they often withdraw coverage. In doing so, they make individuals more vulnerable, less able to recover from a disaster.

And they also create a diminished market for insurance products and higher geographical concentrations of risk. This upward spiral needs to be reversed. Insurance has typically served to transfer the financial costs incurred by disasters away from homeowners, not reduce risks of them happening. Now, approaches to push down both risks and costs are needed.

U.K.-government-backed Flood Re has been designed to make flood insurance more affordable for high-risk areas, with some insurers offering discounts on insurance premiums for homes with flood gates in flood-prone areas. Meanwhile in the U.S., Strengthen Alabama Homes pays people to stormproof their homes and has been shown to lower insurance premiums, and some insurers offer discounts on insurance premiums for homes with impact-resistant roofs. A number of states in the U.S. are also considering legislation that would require insurers to tell their customers how they can reduce weather-related risks to their properties and offer discounts for individual-, community- or state-level mitigation efforts.

All of these actions are helpful. Yet they still treat the symptom, not the disease. Adaptation without mitigation is not enough. Ultimately, we need to reach net zero, to prevent any further increases in global temperature and return risks of catastrophes to a lower level.

Falling short of that, many individuals and communities in the U.K. and the U.S. and worldwide might soon be priced out of home insurance and therefore financial protection against the consequences of natural disasters, leaving them unable to recover.

For many of us working in the property & casualty (P&C) insurance world, the very ground beneath our feet feels like it's shifting. Here in Canada, we've seen it firsthand: the terrifying roar of wildfires sweeping through communities like those in Fort McMurray, the relentless deluge of atmospheric rivers causing historic flooding in British Columbia, or the icy grip of eastern Canadian winter storms that cripple infrastructure. 

These aren't just isolated incidents any more; they're increasingly common and often more intense. And if you look beyond our borders, the story echoes globally – from super-typhoons devastating coastal cities in Asia to prolonged droughts turning landscapes into tinderboxes in Europe. These aren't mere headlines; they're direct, escalating challenges to the very promise of protection we make to our clients.

Catastrophe modeling has long been our steadfast companion, our indispensable compass in this turbulent environment. It's the science that helps us quantify the seemingly unquantifiable, to make sense of the immense, unpredictable forces of nature so we can price risk fairly and manage our portfolios responsibly.

But let's be honest: Even with our most sophisticated models, we've always been running a sprint against time, trying to extract insights from an ever-growing ocean of data. How do you keep pace when the climate itself seems to be "innovating" faster than ever? How do you make strategic decisions when crucial information is scattered across countless unstructured sources – every tweet from a storm-chaser, every drone image of a damaged rooftop, every local news report, every scientific paper on permafrost melt? Trying to sift through all that manually is like trying to catch water with a sieve.

This is where the story truly gets exciting. Large language models (LLMs) aren't just another buzzword or a fleeting tech trend. They represent a groundbreaking, fundamental shift. These powerful AI tools promise to profoundly change how we, as P&C insurers in Canada and across the globe, gain foresight, operate with efficiency, and build true resilience in an increasingly volatile world.

Yesterday's Playbook Won't Win Tomorrow's Game

"Business as usual" is a dangerous illusion, especially when it comes to managing natural hazards. The climate's relentless evolution means that historical data alone is no longer a perfect predictor. Urbanization pushes more valuable assets into vulnerable zones, from Toronto's burgeoning high-rises in potential flood plains to sprawling suburban developments near wildfire-prone areas. And in our connected global economy, a localized event – like a hailstorm over the Alberta prairies or an earthquake in the Pacific Northwest – can have immediate, cascading economic repercussions far beyond its initial footprint.

Our traditional cat models are incredibly powerful, providing precise probabilistic assessments from the structured data they ingest. But their "blind spot" has always been the sheer volume of unstructured, real-time information. Imagine a prairie hailstorm: the frantic social media updates from affected towns, the rapid updates from emergency services, the nuanced observations in a local building inspector's report. Or an Atlantic hurricane brewing: the subtle shifts in sentiment on local news channels, the firsthand accounts of coastal erosion. This constant torrent of informal yet vital information often gets missed or takes days of painstaking, manual effort to properly analyze and integrate. The real challenge for us now is to evolve beyond static risk assessments to dynamic, living insights that guide our decisions, minute by minute, whether we're managing a local Canadian community's exposure or a sprawling global portfolio.

Catastrophe Modeling: Ready for an Upgrade

Cat models aren't going anywhere. They will remain the analytical bedrock that allows us to manage immense risk. They elegantly break down complex natural perils into understandable components:

• Hazard: What's the storm doing? How hard is the ground shaking? What are the projected paths and intensities of wildfires across our vast Canadian forests?
• Exposure: What exactly is insured in this specific area? Every single building, every vehicle, every piece of infrastructure – from the gleaming high-rises in Toronto and Vancouver to the rural properties across the Prairies or the remote communities in the North.
• Vulnerability: How will these specific types of assets fare against those specific perils? What's the structural integrity of a building in Calgary against a severe hailstorm, or a coastal property in Nova Scotia against rising sea levels and storm surges?

By running countless simulations, these models provide us with the probabilities of loss. This crucial information informs everything from how we accurately price policies and manage our portfolio aggregates, to how much reinsurance we judiciously buy, and how we intelligently allocate our capital. They truly are the unsung heroes of our financial stability, both here in Canada and across our international operations.

Giving Our Cat Models a Voice and a Memory

Here's where the transformation gets truly exciting. LLMs, the same technology powering much of the generative AI revolution, are masters at understanding, interpreting, and generating human language. Their superpower is processing the unstructured data – the very information stream that has historically challenged our traditional analytical tools. Imagine the possibilities when we bring this extraordinary capability to our catastrophe modeling and the wider insurance ecosystem:

1. Unlocking the Deluge of Unstructured Data: From Noise to Insight

• Beyond the Numbers: Picture this: LLMs can instantly sift through mountains of news articles, scientific breakthroughs on the thawing permafrost in Canada's North, real-time social media chatter during a widespread power outage in Ontario, emergency dispatches from a flood-stricken town in Quebec, even subtle contextual notes attached to drone inspections of wildfire-damaged forests in Alberta or B.C. They pull out crucial, real-time nuggets about evolving dangers, localized damage, or hidden community vulnerabilities. No more missed signals, no more slogging through disparate reports.
• Policy Language, Instantly Understood: We all know how dense and regionally specific policy documents can be. LLMs can quickly read through vast libraries of policies, identify complex peril exclusions or specific coverage triggers – perhaps related to overland water coverage, wildfire smoke, or even specific seismic zone endorsements. This ensures our exposure data feeding into models is precise and accurate, minimizing ambiguity.
• Adding Richness to Properties: Imagine automatically enriching property data with nuanced details like "a heritage brick house, built in the 1920s, with a newly reinforced basement, surrounded by mature trees in a high-wind zone" – all extracted from casual notes, historical records, or even public property descriptions. This adds crucial layers to our vulnerability assessments within the cat models, whether for a home in Vancouver's urban interface or a sprawling farm in rural Saskatchewan.

2. Models That Learn and Adapt in Real-Time: Dynamic Foresight

• Dynamic Intelligence: LLMs can constantly scan the horizon for the latest scientific research on climate patterns affecting Canadian winters or global monsoon seasons, building code changes in coastal regions, or even subtle demographic shifts affecting urban exposure. This means our cat models can be updated almost continuously, reflecting the very latest understanding of risk, rather than waiting for scheduled, often less frequent, refreshes.
• "What If" Scenarios, on Demand: Actuaries and risk managers won't be limited to pre-packaged scenarios. They can simply ask in plain language: "What's the projected loss if a Category 4 hurricane makes landfall north of Halifax, combined with a 100-year flood in the surrounding areas, taking into account recent infrastructure upgrades and projected sea-level rise?" LLMs help translate these complex, nuanced questions into precise model inputs, allowing for tailored risk analyses for specific Canadian regions or global markets.

3. Clearer Conversations, Faster Decisions: Bridging the Understanding Gap

• Breaking Down the Jargon Barrier: Cat models produce incredibly complex, statistical outputs. LLMs can act as our expert interpreters, translating these highly technical results into clear, concise insights for everyone – from the CEO in the boardroom needing strategic context, to the underwriter assessing a new policy's risk, and the claims adjuster on the ground needing immediate, practical guidance. This vastly improves understanding and accelerates critical decision-making across the organization.
• Reports That Practically Write Themselves: Imagine automating the generation of post-event loss estimates, comprehensive portfolio analyses for both Canadian and international exposures, and even complex regulatory reports. LLMs can pull data from various sources and weave it into coherent, professional narratives, ensuring consistency and remarkable speed.
• Empathetic Communication, At Scale: In the chaotic aftermath of a catastrophe, LLMs can power highly empathetic and personalized communications to policyholders. Imagine instantly providing relevant instructions, real-time claim updates, and tailored support based on their specific policy and reported situation, whether they're affected by an ice storm in Quebec, a flood in the Prairies, or a distant earthquake.

4. Agile Claims Management Post-Event: Swift Response When It Matters Most

First Notice of Loss, Redefined: LLMs can instantly process unstructured inquiries from phone calls, chatbots, emails, and social media, extracting crucial claim details and automatically kicking off the claims process much faster. This is particularly vital in large-scale events that affect thousands across a broad geographic area.

• Instant Damage Triage: When combined with computer vision (analyzing drone footage or policyholder photos from a disaster zone), LLMs can provide immediate preliminary damage assessments, prioritizing and triaging claims for human adjusters, ensuring help gets to those who need it most, faster.
• Smarter Fraud Detection: By analyzing textual patterns in claims narratives, cross-referencing data, and spotting inconsistencies, LLMs can significantly bolster our defenses against fraudulent claims, protecting our integrity and resources across all markets.

5. Sharper Underwriting and Portfolio Strategy: Precision Risk-Taking

• Real-time Risk Pricing: Underwriters can leverage LLM-derived insights from real-time data to refine risk assessments and adjust pricing dynamically, ensuring our rates truly reflect evolving perils and exposures across different geographies, from Canada's unique risk zones to global territories.
• Optimized Reinsurance: With more granular, living risk insights, we can make smarter decisions about our reinsurance purchasing and capital allocation, strengthening our financial position against both Canadian-specific and global catastrophic events.

6. Navigating Regulations with Ease: Global Compliance, Local Insight

• Always Compliant: LLMs can continuously monitor regulatory updates across various Canadian provinces and diverse international jurisdictions, automatically flagging potential compliance gaps related to our cat model usage or data reporting.
• Effortless Reporting: Automating the heavy lifting of compiling data and narrative for complex regulatory filings, significantly reducing manual effort and potential for error, ensuring we meet all obligations efficiently.

Navigating This New Frontier Together

This isn't to say it's easy. Bringing LLMs into our world comes with real, tangible considerations:

• Data Integrity & Trust: We need absolute certainty that the vast amounts of data feeding these LLMs – often sensitive and proprietary – are accurate, secure, and used ethically. Robust data governance is paramount, ensuring compliance with Canadian privacy laws and international standards.
• Bias and Transparency: LLMs can inadvertently pick up biases from their training data. We must ensure model transparency and keep our expert "human-in-the-loop" to ensure fair, explainable, and equitable outcomes, adhering to the highest ethical standards.
• The Integration Puzzle: Seamlessly connecting LLMs with our existing cat models, core insurance systems, and diverse data sources will require thoughtful architectural design and skilled engineering. It's about building bridges, not just adding new components.
• Building New Skills: We'll need to invest in our people, developing new skills in data science, AI engineering, and even "prompt engineering" – the art of asking LLMs the right questions. This upskilling will be crucial for Canadian talent to lead not just at home but on the global stage.
• Evolving Regulations: As with any transformative technology, clear regulatory guidelines will need to evolve across jurisdictions – from Ottawa to Brussels, Washington D.C. to London – to ensure responsible adoption across the industry.

The path forward is about intelligent experimentation and collaboration. It's about launching strategic pilot programs, fostering true collaboration between our cat modelers, data scientists, IT teams, and business leaders. And crucially, it's about partnering with technology leaders who genuinely understand both the intricate world of insurance and the nuances of AI.

Conclusion: Amplifying Our Foresight, Fortifying Our Promise

The convergence of catastrophe modeling and large language models marks a pivotal moment for the P&C insurance industry. 

LLMs are not here to replace the profound expertise of our cat modelers, or the critical human judgment that guides our underwriters and claims professionals. Instead, they are here to be powerful augmentations – tools that amplify our human capabilities, allowing us to see further, respond faster, and operate with unprecedented precision.

By strategically embracing this LLM revolution, we can move beyond merely reacting to risk to truly anticipating it. We can gain deeper, more dynamic insights into peril, exposure, and vulnerability, leading to more accurate pricing, faster claims processing, and more resilient capital management. 

In an increasingly uncertain world, the P&C insurers – particularly those navigating Canada's unique and evolving risk landscape and contributing to the global insurance market – that boldly and thoughtfully adopt this new paradigm will not only strengthen their own foundations but also provide greater stability, security, and peace of mind to their policyholders. 

We will solidify our role as essential pillars of societal resilience. The time to explore this transformative frontier is now – with authenticity, courage, and a clear vision for a more secure future.

Generation Z is rapidly entering the workforce, and as a result, it is important for employers to understand their perspective to attract and retain this talent. To understand Gen Z at a deeper level, Tugman Consulting completed in-depth interviews with a sample of Generation Z participants ranging in age from 19-24, with 22 being the most common age within the study.

Generation Z, born between 1997 and 2012, makes up 27% of the global workforce as of 2025, bringing unique perspectives and expectations to the workplace. Their experiences, particularly during COVID-19, have shaped their values around work-life balance, mental health, and community.

The study findings:

Importance of Community in the Workplace

Gen Z places a high value on connection and community among coworkers, which significantly influences their job satisfaction.

• Participants expressed a desire for friendships with coworkers outside of work.
• Employers should facilitate opportunities for candidates to meet potential coworkers during the hiring process.
• A welcoming environment that allows for open communication is essential for Gen Z employees.

Compensation and Financial Empowerment

Financial security is a top priority for Gen Z, who expect livable wages and compensation that allows for a fulfilling life.

• Participants emphasized the importance of being paid fairly for their work.
• Many expressed anxiety about achieving the same financial stability as previous generations.
• Compensation should enable experiences, such as travel, rather than just covering basic living expenses.

Normalization of Mental Health Conversations

Mental health discussions are commonplace for Gen Z, who expect workplaces to support their mental well-being.

• Participants indicated that time for therapy is crucial, and not allowing it is a deal-breaker.
• Employers are expected to take an active role in preserving the mental health of their workforce.
• Open conversations about mental health should be encouraged in the workplace.
• Managers should take time to "know" employees so they can differentiate between an emerging mental health issue and a performance issue.

Diversity, Equity, and Inclusion as Core Values

Gen Z views diversity, equity, and inclusion (DEI) as essential components of a workplace culture, not just as programs.

• Participants said visible diversity is crucial; lack of it could lead them to seek other job opportunities.
• DEI should be integrated into business practices rather than treated as a checkbox exercise.
• Inclusion means valuing contributions from people of all ages and experiences.

Flexibility in Work Arrangements

Flexibility is a key expectation for Gen Z, who seek a balance between work and personal life.

• Participants want to work hard but do not want to be defined by their jobs.
• They prefer environments that recognize personal needs and allow for flexibility in work hours.
• Hybrid work arrangements are seen as ideal for maintaining work-life balance.

Career Growth and Development Opportunities

Opportunities for career advancement are critical for Gen Z, who desire transparency in growth paths.

• Participants expressed a need for clear communication about development and learning opportunities.
• They expect to be considered for advancement despite their age and level of experience.
• Many believe that moving between roles is essential for growth and learning.

Impact of COVID-19 on Work Perspectives

The COVID-19 pandemic has significantly influenced Gen Z's views on work, emphasizing the importance of meaningful employment.

• Participants want to work for companies that align with their values and contribute to society.
• The pandemic has heightened their appreciation for in-person connections and the need for balance in work.
• They resist returning to pre-pandemic work norms that do not prioritize well-being.

Generation Z is bringing new perspectives into the workplace. They are open about their mental health and where they stand on social issues. They will bring their whole selves to work, and they expect their managers to be comfortable with that. Employers should beware of stereotypes. 

While this generation does not work to live, they are hardworking and will give their all for the right work environment and financial compensation. Understanding this generation's perspective will not only help employers attract and retain them, but it will also help gain their trust and loyalty, which will lead to their optimal productivity.

At Westfield Insurance, I'm seeing firsthand how data is reshaping our industry. The industry possesses a significant amount of data, but is it the right data? Previously, this volume of data was enough for an underwriter to assess risk. Today, to write sustainable business, underwriters have to cut through the noise and extract actionable insights, which raises the question: Is the insurance industry moving into the era of Underwriting 2.0?

During the fourth episode of insurtech Send's "Infuse" webinar, I had the chance to discuss how data is shaping the underwriting process with several other industry leaders. I pointed out that professionals across many disciplines at Westfield are embracing AI and learning how to appropriately integrate it into their underwriting workflows. However, fellow panelist Kelly Cusik, MD, of Deloitte Consulting, was candid about how the industry lacked maturity in tapping into generative AI. Kate Enright, head of data at Miller Insurance Services, was optimistic about unlocking the full potential of data by adopting modern technologies. 

The consensus was clear: Underwriting has evolved, and data is now at the heart of how we assess risk and price policies.

Here are four ways to leverage data as usable, insightful, actionable, and truly embrace Underwriting 2.0:

1. Evolve from process-driven to data-driven

Although underwriting is still rooted in process, we are now supplementing it with advanced data sources and AI-powered tools to get a deeper picture. As an example, at Westfield, we are integrating third-party data that leverages AI to look at risk characteristics across the underwriting cycle from ingestion to submission. These tools will not replace underwriters but provide them the right data at the right time to speed decisions and improve consistency, ultimately influencing sustainability. And it's not just us. My peers across the industry from those like Miller Insurance Services to consulting firms like Deloitte are all leaning into modernization. APIs, portfolio-level exposure tracking, and digital contract tools, are all playing a role as underwriting goes through a transformation.

2. Be AI-ready

You may be surprised to learn that AI is not new to the insurance industry. We've been exploring machine learning and predictive analytics for years as a way to support risk modeling, pricing and triage.

Gen AI is just the next phase in the AI evolution. During the webinar, Kelly indicated that many companies have the tools but lack the trust and training in adopting gen AI. She suggests that with the right culture and leadership, the industry will see more smart technology helping underwriters do what they do best.

3. Make change interesting

Earlier this year, Westfield hosted an AI Day across the company. We had more than 200 employees participate. We brought in a speaker from a large AI research and deployment company, hosted breakout sessions, and gave employees hands-on access to AI tools. The energy was incredible.

People weren't afraid. They were curious. They saw how AI could make their workday easier, not harder. And they walked away with real ideas that could help solve business problems and achieve business goals. That's what happens when you approach innovation as a team sport with the right guardrails.

4. Remember: Insurance is still an art and not all science

During the webinar, Kelly reminded the audience not to lose sight of the human side of insurance. She suggested that industry leaders must invest in training, cross-functional collaboration, and creating space for experimentation. Kate Enright from Miller Insurance Services explained that real transformation happens when data, people, and process evolve together. She suggests insurers need to spend 50% of the time on the value story, 25% on data, and the rest on making change engaging.

Final Thoughts

Underwriting continues to change. It's no longer just about process and cost control. It's about how effectively you use data to make better decisions. And the future is collaborative, data-driven, and human at its core. We're not just underwriting risk any more. We're underwriting the future—and I couldn't be more excited about where we're headed.