Key Takeaways

• Recent outbreaks of hantavirus and Ebola demonstrate that serious infectious disease events can create significant public health challenges without necessarily posing a high risk of becoming global pandemics.
• Strengthening public health systems and maintaining informed vigilance are essential to managing future infectious disease risks.
• The next global health threat may emerge from an unexpected source, making sustained investment in preparedness, surveillance, and response capabilities more important than focusing on any single pathogen.

Executive summary

Recent outbreaks of Andes hantavirus and Bundibugyo Ebola have renewed attention on pandemic risk, but neither currently appears likely to cause sustained global spread. Their main significance lies in what they reveal about pandemic preparedness: the importance of rapid detection, effective contact tracing, international coordination, and timely countermeasures.

While disease outbreak severity warrants attention, pandemic potential depends more on efficient human-to-human transmission, pre-symptomatic or respiratory spread, population immunity, and the ability of health systems to respond quickly. Therefore, these recent outbreaks are best understood not as signals of an imminent pandemic, but as important reminders to remain vigilant, balanced, and guided by the full body of evidence.

The hantavirus outbreak: A local exposure becomes an international public health exercise

The hantavirus outbreak, linked to expedition cruise ship M/V Hondius, illustrates how a rare infection can quickly become a major international containment challenge. In this event, Andes virus infection was first identified more than three weeks after the first death. Epidemiological investigation and viral sequencing suggested that the index patient likely acquired the infection during travel in South America, including areas of Argentina and Chile where the Andes virus is endemic. Because the case was detected in the context of cruise ship travel, passengers and high-risk contacts had already dispersed across multiple countries before the virus was identified.

The cruise ship setting inevitably recalled memories of the early COVID-19 outbreaks at sea, when ships became symbols of uncertainty, containment challenges, and global spread. Biologically, Andes virus is very different from SARS-CoV-2, but the echoes were familiar: delayed recognition, international passenger dispersal, and the need for rapid cross-border coordination.

This does not make Andes hantavirus a likely pandemic pathogen. Its human-to-human transmission remains limited and requires close contact with respiratory secretions. This outbreak resulted in 13 infections and three deaths, yielding an outbreak-specific case fatality rate (CFR) of approximately 23%. The 42-day follow-up period for contacts was completed with no additional secondary cases detected, and the World Health Organization (WHO) officially declared the outbreak over on July 2, 2026.

The event illustrates how global mobility can transform a small outbreak into a complex public health operation. Even when the biological risk of sustained spread is low, delayed recognition, international travel, and long monitoring periods can place substantial demands on surveillance, communication, and coordination systems.

The Ebola disease outbreak: A reminder that surveillance and rapid detection matter

The significance of the continuing Bundibugyo ebolavirus outbreak in the Democratic Republic of the Congo, with associated cases and transmission previously reported in Uganda, should not be underestimated. This outbreak has expanded at an exceptional pace and is now the second-largest on record, with a CFR of approximately 44% as of July 30, 2026. The WHO has designated the outbreak a public health emergency of international concern (PHEIC), reflecting both its severity and the need for coordinated international action. Unlike the Andes hantavirus outbreak, however, it highlights a different set of containment challenges: the consequences of limited surveillance, contact tracing, and diagnostic capacity.

The outbreak response is further complicated because it is caused by a rare species of Ebola without an approved vaccine and the epicenter is an area affected by conflict. Bundibugyo virus is a less common Ebola virus species, and many existing diagnostic tests, vaccines, and treatment strategies historically have been developed with other Ebola virus types in mind. As a result, the outbreak likely went undetected for months, delaying recognition and response.

Ebola outbreaks demand urgent attention because of their potential for substantial mortality and capacity to disrupt and overwhelm health systems. But disease severity alone does not determine pandemic potential. Ebola generally spreads through direct contact with infectious bodily fluids and is most contagious once symptoms are present, especially during severe illness. These characteristics make sustained global spread much less likely than with a respiratory virus that transmits efficiently before symptoms develop.

The primary concern is not that Ebola will become the next global pandemic but that outbreaks can expand rapidly when they occur in settings where conflict, humanitarian crises, strained healthcare infrastructure, or gaps in diagnostics, contact tracing, and countermeasures hinder early detection and control.

What these outbreaks remind us about pandemic risk

Pandemic risk is not driven by a single feature. The greatest risk arises when several factors merge: a novel pathogen, meaningful disease severity, efficient human-to-human transmission, and low population immunity. Transmissibility is especially important. A virus that spreads through respiratory droplets or aerosols, particularly before symptoms appear, is generally more difficult to contain than one that requires close contact after symptom onset.

This is why the pandemic potential of Ebola and Andes hantavirus remains low in the current context, despite their seriousness. In contrast, influenza viruses remain a more persistent pandemic concern because they can spread efficiently between people, transmit before symptoms are recognized, and evolve or reassort into new strains. Avian influenza remains a risk to monitor closely, not because it currently spreads efficiently between humans, but because adaptation could change that risk profile.

Recent evidence adds nuance to the avian influenza risk discussion. Windborne spread of H5N1 between poultry farms via virus-contaminated material from infected bird droppings cannot be ruled out, although it remains difficult to prove and does not necessarily imply efficient airborne transmission between humans. Separately, asymptomatic human H5N1 infections have been reported, albeit infrequently and often identified through enhanced surveillance or household contact investigations. These findings do not change the current assessment that sustained human-to-human transmission remains limited, but they reinforce why avian influenza requires careful surveillance, especially for changes in transmission dynamics.

https://public.flourish.studio/visualisation/29931603/

This comparison highlights why severity alone is not enough to determine pandemic potential.

The recent hantavirus and Ebola outbreaks particularly draw attention to the system-level drivers of pandemic risk, rather than viral characteristics: global connectivity, weak early detection and response, and limited countermeasures.

Global travel can move exposed individuals across borders before an outbreak is recognized. Delays in diagnosis and containment increase uncertainty and response complexity. Limited availability of pathogen-specific tests, vaccines, and treatments increases vulnerability, even when the pathogen itself has limited pandemic potential.

Three key lessons

1. Travel and ecological change continue to reshape infectious disease risk. Global travel, trade, ecotourism, agricultural expansion, and increasing human contact with wildlife all create more opportunities for zoonotic spillover and cross-border spread. Many spillover events remain dead ends, but repeated opportunities for exposure – sometimes described as viral chatter – give pathogens more chances to adapt.
2. Pathogen disease severity should not be confused with pandemic potential. A pathogen can be highly lethal but poorly suited for sustained global spread. Conversely, a pathogen with a lower CFR but efficient respiratory transmission can have far greater population impact. For risk assessment, the key question is not simply “How dangerous is this virus?” but “How easily can it spread, when does transmission occur, and what level of immunity and countermeasures exist?”
3. Preparedness is shaped by speed. The rate of detection, sequencing, reporting, contact tracing, clinical response, and countermeasure deployment can determine whether a localized outbreak remains contained or becomes a wider crisis. Even when no pandemic emerges, delayed recognition can still cause avoidable morbidity, mortality, operational strain, and public anxiety.

Conclusion: A test of preparedness, not a prelude to a pandemic

The Ebola outbreak in the Democratic Republic of the Congo and Uganda and the recent Andes hantavirus outbreak are serious public health events that deserve close attention. Yet despite their severity, neither currently exhibits the transmission characteristics typically associated with sustained global pandemic spread. While both pathogens can cause severe disease and significant mortality, their epidemiological profiles make widespread worldwide transmission unlikely in the current context.

That does not diminish their importance. These outbreaks are reminders that emerging infectious threats continue to test health systems, surveillance capacity, diagnostic readiness, and international coordination. The key takeaway is one of informed vigilance rather than alarm. Pandemic preparedness is not only about identifying the pathogen most likely to cause the next global crisis but also about strengthening the systems that allow us to detect, interpret, contain, and respond to infectious threats before they escalate.

References

1. https://www.ecdc.europa.eu/en/infectious-disease-topics/hantavirus-infection/surveillance-and-updates/questions-answers-outbreak
2. https://www.sciencedirect.com/science/article/pii/S2590170226001822
3. https://www.who.int/emergencies/disease-outbreak-news/item/2026-DON614
4. https://www.nejm.org/doi/full/10.1056/NEJMra2607216
5. https://www.sciencedirect.com/science/article/pii/S1755436523000403
6. https://www.cidrap.umn.edu/avian-influenza-bird-flu/can-avian-flu-spread-wind-cant-be-ruled-out-experts-say
7. https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2840680
8. https://www.who.int/publications/m/item/covid-19-global-risk-assessment--version-10

A humanoid robot ran 100 meters in 9.39 seconds at the World Humanoid Robot Games on Saturday, bettering the human record of 9.58 seconds set by Usain Bolt. Another ran the 400 meters in 38.16 seconds, almost five seconds faster than the human best. Still another jumped almost nine-and-a-half feet into the air — nine-and-a-half-feet! The human high jump record is a hair over eight feet.

But the winner of the 100m race crashed into the padding at the end of the track, reeled backward for maybe 10 meters, then fell and snapped in the middle, sending sparks everywhere. 

That image, on top of the impressive speed, the jumping and the other abilities demonstrated at the games in Beijing, strikes me as a pretty good summary of where humanoid robots stand today. Once you sort through all the hype (from Elon Musk, among many others), you see tremendous progress in specialized functions, but you also see that humanoid robots won't show up broadly in factories or homes — or the insurance policies that cover them — for many years.

Let's have a look.

Frankly, I've never understood the need for robots to be humanoid. I mean, if you want to really go fast for 100 meters, you should put a little electric car in the lane on the track. One of those babies will get to 60mph in the first 50 meters  — forget Usain Bolt's top speed of a bit under 28mph. The whole race would take maybe three seconds. If you want a robot to set a high jump record, there are these things called rockets.

And specialized robots are already doing great work in a host of settings, as this New York Times article describes in detail:

"Autonomous carts have largely replaced forklifts for moving materials around [a Hyundai] plant. Four-legged robots that look like dogs search for flaws by peering under car bodies. Flat, wheeled robots the size of queen beds slide under Ioniq 5 or Ioniq 9 electric vehicles as they come off the assembly line, lift them from below and transport them to testing stations."

I'd also remind you of the videos we've all seen of the robot forklifts whizzing around Amazon warehouses that grab products from bins so high it would take a person a bit to climb or be lifted there, or that simply pick up bins and take them to people for easy picking and packing. Robots that look nothing like people are being developed that would flip burgers and do other routine kitchen work in fast-food restaurants. 

The Times article says auto makers have high hopes for humanoid robots that will complement the massive, one-armed machines they already use to pick up and place car parts and weld them into place, but look at the main use the Times describes: 

Hyundai plans to use humanoid robots "initially for arranging components in a specified order for humans to install in vehicles. That is the same basic task that the BMW robot, made by Figure AI, demonstrated in Spartanburg. Mercedes-Benz, which is testing robots made by Apptronik in Austin, Texas, also plans to use them for handling and sorting materials. Carmakers say they have trouble retaining workers to perform such basic tasks."

That sort of use could be very helpful, but it's hardly an example of a versatile robot that could perform an unlimited variety of tasks outside of a controlled environment like a factory.

There are two major problems to overcome first, and neither looks to be solvable any time soon, no matter how much a humanoid robot outpaced Usain Bolt. 

The first is that human hands are really, really hard to imitate. Not only are fingertips sensitive in ways that machines have trouble replicating but they manage a combination of flexibility and strength that perplexes robot designers. If they want the strength of the human hand, they need to put a powerful motor in the wrist. If designers want the dexterity, they have to put a host of tiny motors in all the joints where a hand can flex. But if you do both, you wind up with a large, clunky hand. You have to make tradeoffs, and you wind up with a hand that doesn't measure up to the human combination.

The second problem is that, to be as versatile as proponents claim humanoid robots will be, they have to be able to operate in messy environments that robots simply can't navigate at the moment. A factory is one thing, but imagine a robot in your home having to step over the toys left in the family room, not bump into the dog, sort through a whole family's laundry, handling a special item precisely as your teenager demands, and so on. Not gonna happen. 

The Wall Street Journal reports that investors are piling into companies working on so-called world models, which use video games and simulations to develop more robust "action" models about how the real world operates, to go beyond the sort of top-down, rule-by-rule, prescriptive models that robots rely on today. But even the enthusiasts say the world models are currently at about the stage where the large language models used for generative AI were with ChatGPT2. That model was released seven years ago.

You'll surely continue to hear lots about the imminent arrival of armies of AI-powered, humanoid robots. They're an enticing idea. Everybody who remembers "The Jetsons" wants their Rosie the maid. There's even a cool term for the trend: "embodied AI." Hypemeister Elon Musk will, alone, make sure that humanoid robots stay in the news, given that delivering 1 million Optimus robots (at maybe $30,000 apiece) is one of the goals he has to meet to qualify for his $1 trillion pay package. And lots of sci-fi fans will continue to resonate with images of robot servants handling our daily drudgery. 

But you can ignore the hype for at least a few more years, both in personal terms and in terms of providing workers comp coverage and insuring homes. Specialized robots will provide great value in industrial settings, but they'll continue on their current trajectory. There won't be some radical change because of a new class of robots. And nobody but the earliest adopters will be tripping over a humanoid in their kitchen any time soon. 

Cheers,

Paul

Over the past several years, I’ve spent time with business owners, brokers, and community leaders in cities like New Orleans, New York, Cleveland, and Oakland. The conversations differ by market, but they converge on the same theme: not just affordability, but uncertainty. Business owners describe paying for losses out of pocket rather than risk losing coverage. Agents describe shrinking availability and carrier appetite that shifts with little warning. The message is consistent: insurance remains essential to economic resilience, yet it is often experienced as unpredictable and hard to navigate.

That reality shaped District Cover’s Urban Risk Report. The goal wasn’t just to document what’s happening in urban insurance markets, but to understand why — and what it means for the businesses and communities that depend on stable coverage.

The Stakes Are Higher Than We Often Acknowledge

More than 85% of small businesses operate in metropolitan areas. They are not just economic units — they are anchors that create jobs, provide essential services, and define neighborhood character. When those businesses can’t access insurance, the effects ripple outward: vacant storefronts, residents forced to travel farther for basic goods, and a gradual erosion of the commercial diversity that gives a city its identity. Insurance alone can’t solve the structural challenges facing urban communities, but it remains a critical part of the solution — enabling businesses to secure leases, obtain financing, and recover after losses.

A Market That Feels Constrained—But Isn’t Always

There’s a growing perception that urban risk is becoming harder to insure, and in some ways that’s understandable. Catastrophe exposure, shifting capacity, and evolving liability trends have tightened underwriting and pushed more placements into E&S markets. Even as pricing softens elsewhere in commercial lines, that relief hasn’t reached urban corridors to the same degree — density means claims are more correlated, and social inflation has hit liability lines especially hard, so urban risk is still being priced against a liability curve that hasn’t softened the way property has.

But what we found in the Urban Risk Report is more nuanced: the constraint isn’t a lack of insurable risk. It’s a gap between how risk is distributed in urban environments and how it’s evaluated. Two businesses on the same block can carry materially different exposure based on construction, occupancy, and maintenance — yet underwriting still leans on broad geographic assumptions, often at the ZIP code level, that erase that variation. Rebuild costs compound the problem: urban labor, permitting, and older building stock routinely push replacement values 15–35% above suburban benchmarks, and traditional valuations often understate that gap.

The result is a market that feels tighter than it actually is, where viable risks get declined or mispriced not because they’re uninsurable, but because they don’t fit the model. When admitted carriers pull back rather than re-price at this level of detail, the risk doesn’t vanish — it moves. California’s FAIR Plan alone grew from roughly $153 billion in exposure in 2020 to more than $450 billion by 2024. That’s not evidence these risks are uninsurable. It’s evidence they’re being routed around rather than evaluated.

The Opportunity: Precision, Not Avoidance

The answer, we believe, is precision. Insurers now have the tools — property-level data, permit and renovation records, corridor-level foot traffic, and AI-structured local data — to underwrite at the resolution urban risk actually requires, instead of a ZIP-code average. Carriers that use them can identify well-run, well-maintained businesses inside neighborhoods labeled high-risk wholesale: viable businesses that others overlook simply because they haven’t looked closely enough.

That’s also, candidly, a business opportunity. Underserved urban corridors represent real unmet demand — business that can be profitable for carriers willing to underwrite on merit rather than geography. Timing matters here too: precision underwriting is easier to build before pricing hardens further, not after. Catastrophe costs are rising industry-wide, and the carriers positioned to expand into underserved corridors profitably are the ones doing it now, on their own terms, rather than being forced into it later by a harder market.

Reframing the Conversation

The most important takeaway from this work is simple: urban businesses aren’t becoming uninsurable. They’re being misunderstood. That distinction matters — not just for underwriting outcomes, but for the long-term resilience of the communities these businesses support. At District Cover, we believe participating in urban markets takes more than showing up during favorable cycles. It takes a long-term commitment to building stability for both businesses and the communities they serve.

This report is one contribution to that effort — an invitation to insurers, brokers, policymakers, and community leaders to think harder about how insurance markets can better support the economic resilience of America’s cities. Because insurance isn’t just risk transfer. It’s what makes it possible for businesses to open their doors, invest in their future, and stay part of the communities that depend on them.

Download the full Urban Risk Report to explore the data, trends, and insights shaping the future of urban insurance.

Running an independent agency means constantly triaging what gets your attention. Renewals, service calls, new business, and the next carrier appointment. With all of that in motion, it's easy to lose track of opportunities that could help you grow your business.

If your agency's book is built mainly on home and auto, commercial offers a real chance to diversify while picking up higher premiums along the way. It's an easy opportunity to overlook, though. Commercial quoting has traditionally been complex, and without the right technology, spotting where those opportunities live has been hard to do.

That's starting to change. Here's what's shifted, where the commercial opportunity actually lies, and why independent agents are especially well-suited to act on it.

From Three Workflows to One

Commercial quoting has traditionally meant juggling three separate systems: the application, the rater, and the carrier portal. Each handoff meant re-entering the same client information. Each one added time and room for error. A client would ask about a BOP, and the quote might sometimes take until the following week, once the risk worked its way through submission to quote. 

That's changed. Modern rating and quoting technology has consolidated those steps into a single workflow. Client information is entered once, and carrier quotes surface directly inside the submission workflow. A process that used to take the better part of an afternoon can now take a few minutes. That means commercial opportunities sitting in your book are much easier to act on.

The Cross-Sell Opportunity Already in Your Book

A good number of your personal lines clients probably also own a business. More of them than you'd guess are likely relying on you for their home and auto. For commercial coverage, though, they're going elsewhere, or nowhere at all. That's not a knock on how you've built your book. It's just what happens when there's more relationship there than there's been time to explore.

The upside is that you don't have to go find these clients. You already have them. A good management system can flag commercial cross-sell and upsell opportunities directly inside your existing client records. It connects the dots you might not have time to connect on your own. A client with a general liability policy but no commercial property coverage is a signal worth a second look. 

It's the kind of detail that's easy to miss manually scanning a file, but easy for the right system to surface automatically. Leaning on technology this way is a strong place to start with commercial. The opportunity is already sitting in your book, and the more policies a client holds with your agency, the more likely they are to stay.

Why the Independent Agent Still Closes the Deal

Finding commercial opportunities is only half the equation, though. Technology can surface them, but it won't close them. That part still comes down to relationships, and that's where independent agents have an advantage no software or national carrier can replicate. You know your clients deeply: not just their premium history, but their business, their plans, what keeps them up at night. You know the local market they operate in and the pressures specific to it. That kind of trust is exactly what turns a flagged opportunity into a signed policy.

The benefits of commercial go far beyond any individual sale. Expanding into commercial helps you build a diversified business. Personal lines and commercial lines don't always move in sync. Rate cycles, carrier appetite, and underwriting conditions in one don't necessarily track the other, so agencies with a foot in both aren't relying on a single market's conditions to carry the whole book. Writing more commercial accounts also gives you room to build real depth in a niche — restaurants, contractors, professional services, whatever fits your market. That kind of specialized knowledge makes your agency an obvious call for local business owners.

From Opportunity to Action

Taking advantage of the commercial opportunity doesn't require a big shift in how you run your agency. The friction that used to make commercial too much work to chase down has largely disappeared. Software built to spot these opportunities is now doing more of that work for you. What's left is simpler: the accounts, the trust, and the expertise you've already built are finally easy to put to work.

The Problem

Most IoT loss-prevention studies don’t survive actuarial scrutiny — underpowered samples, uncorrected selection bias, and hidden program costs have left carriers unable to trust the ROI case for smart-home safety programs, or to justify scaling one past a pilot.

Why It Happens

Proving a sensor prevented a claim runs into three structural problems: there’s no dated invoice for a loss that never happened, early results are skewed by anti-selection bias and poor test design, and the real benefit only emerges once a program has matured well beyond its first few months. Most vendor studies are sized and timed to miss all three.

The Solution

Back solutions that have already cleared the actuarial bar — tens of thousands of device-exposure years, cohort-level (not anecdotal) evidence, and documented, auditable repair records — rather than programs still running on pilot-stage promises.

Results

The synthesized results from four home insurance partners with mature deployments and statistically relevant underwriting years of data are as follows:

• 60% of non-weather water losses occur on the pressurized side of the water system (Mains water)
• 40% of non-weather water losses occur on the drain side of the water system
• 60% reduction in the frequency of non-weather water losses coming from the pressurized side of the water system
• ROI of 40% in mass market homes, 163% in mass affluent homes, and 980% in high-net-worth homes

It's easy to assume insurance silos stem from communication problems. When teams don't share enough information, systems become disconnected, and business units operate in disparate ways.

But silos begin with organizational structure. Teams prioritize work according to the operating model around them. If organized vertically, technology follows the same pattern. Each business unit may build its own systems and data practices to address its immediate needs, creating fragmentation that's difficult to unwind.

Breaking down silos in insurance requires an operating model that connects people, data, systems, and accountability. You can accomplish this by preserving each business unit's deep expertise while building a shared infrastructure that helps them work more effectively across your organization.

Insurance Needs Depth and Connection

Each line of business has its own risk appetite, customer needs, broker relationships, underwriting considerations, and claims complexity. That depth is one of the reasons clients and brokers turn to specialty carriers in the first place.

However, specialization can create the perception that every business unit is entirely unique, making it harder to identify common threads across the entire organization. Teams may assume there's little opportunity to share processes or insights because their work feels highly specific to their line of business.

Many of the core processes behind each segment are similar across the entire organization. You need experts who understand their markets in detail, but you also need the infrastructure to help those experts share their insights and serve customers with greater consistency.

Disconnection Creates Real Costs

When teams and systems aren't connected, you can miss opportunities that should be visible across your organization. A single customer may have relationships across multiple business units, but your organization may not have a unified view of that customer. Even when you know that connection, the customer or broker experience may vary.

Disconnected systems can also create operational inefficiencies. If each business unit builds its own technological solution for a similar issue, your organization may end up solving the same problem several times. That creates higher implementation and long-term maintenance costs. It also limits the ability to cross-skill employees or create business continuity across related functions.

From a technology perspective, the better model is to build once and deploy many times. You can identify what's common and manage the differences intentionally.

The more connected your organization becomes, the easier it is to create consistent workflows and more predictable delivery. But reaching that point requires you to think differently about the operating model behind your technology.

Prioritize Product Over Project

Many insurers still approach technology through a project-based model. A team is assembled to complete a specific project. Once they deliver the project, that team disbands and moves on to the next initiative. This approach can create inconsistency. Each new project may require a new team and a new understanding of the business problem.

However, a product-based model offers a different approach. Instead of bringing teams to the work, you bring the work to your teams. Those teams own a product, platform, or business capability over time. They understand the business context, the technology environment, the backlog, the capacity of the team, and the outcomes they're responsible for delivering.

This model can be especially valuable when teams are built around shared capabilities, such as underwriting or claims, that support multiple business units rather than a single line of business.

Underwriting offers a clear example. While each specialty line is unique, many parts of the underwriting process are similar across the business. Rather than having each business unit build its own separate underwriting tools or workflows, you can create a shared team focused on underwriting as an organizational capability. That team can include experts from different specialty lines, allowing them to build common solutions while still accounting for the unique needs of each business unit.

This approach enhances specialization. Business unit experts contribute to a broader capability that benefits from shared knowledge and accountability. It also creates the foundation for stronger data practices, because connected teams are better positioned to collect and use that data consistently.

Shared Data Is the Foundation for Scalable Technology

Data plays a central role in how you prepare for automation and more advanced digital capabilities.

For these technologies to create value, you need strong data foundations. If data is fragmented across business units or disconnected from the teams building technology, something like AI can become another siloed tool rather than a source of your organization's value. A common data set across business units can help you build technology that supports multiple parts of the organization.

Data engineering should be embedded into the teams responsible for business capabilities. When data and business expertise come together within a product team, you're better positioned to build tools that solve real problems and can be scaled across similar use cases.

But you still need human judgment to harness that data and technology.

Keep Humans in the Loop

Brokers and clients need people who understand the risks and the context behind every decision. Don't believe efficiency means removing the human element from your business.

Technology should help teams focus more time on the work where human judgment matters most. For underwriters, that may mean using automation tools to reduce more routine tasks so they can spend more time on risk selection and pricing sophistication. For business development teams, it may mean using data to identify trends that warrant outreach, such as a change in business flow from a particular broker or market segment.

But a human still needs to understand what's happening and determine the best next step. The human decision then creates feedback that can improve the model over time.

Use technology to make your experts more focused and informed. To do that well, avoid some common modernization traps.

Tie Modernization to Your Business Value

One of the most common risks in technology modernization is chasing solutions before clearly defining the business problem. New tools can be appealing, but technology should never be a solution looking for a problem.

Be selective about what you build versus what you buy from vendors. If a capability helps differentiate your business, building it provides more control and long-term value. If it supports a standard business function, buying an existing solution may be the more efficient choice.

Avoid creating new silos through disconnected tools or over-customized solutions. Make sure technology teams have the structure and visibility required to deliver consistently. That's what turns modernization from a series of projects into a sustainable operating capability.

Building Your Connected Insurance Enterprise

As the insurance industry continues to modernize, treat technology as part of a broader evolution of how work gets done. Brokers and clients are looking for insurers who can combine their expertise and insights with the infrastructure of the future. Balance efficiency with delivering the clarity and confidence they need in an ever-changing, complex market.

Insurers have become very good at preparing for the big moments. Hurricanes, hailstorms, wildfires, floods and tornadoes are analyzed in remarkable detail, helping carriers estimate losses and prepare for large-scale events.

But some of the biggest drivers of property risk develop in silence, long before a named storm appears on the forecast.

Properties are constantly responding to the environments around them. Day after day, ordinary weather conditions gradually change the condition of roofs and exterior materials. Over time, they change the condition of a property in ways that aren't always visible until the next severe weather event turns gradual deterioration into a claim.

Catastrophe models are essential to helping us understand what could happen. What's becoming just as important is understanding what's already happened to the property before that event occurs.

Property condition tells part of the story

Rarely does a single storm cause a roof to fail; catastrophic damage is almost always the culmination of long-term environmental stress. Years of exposure to thermal fluctuations, moisture penetration, UV radiation and heavy rainfall systematically degrade roofing systems. By the time severe weather arrives, the property may already be more vulnerable than anyone realizes. The storm gets the blame, but the conditions leading up to it often tell an equally important part of the story.

These gradual changes have historically been difficult to observe consistently across large portfolios. That's beginning to change. Better property-level data is giving insurers a much clearer picture of how long-term climate exposure affects the condition of homes over time.

A recent analysis of more than 2.8 billion AI-derived roof observations across nearly 2,100 U.S. counties pointed to consistent relationships between chronic climate exposure and roof longevity. Counties experiencing the largest daily temperature swings showed roof aging about 23% faster than those with more stable climates. Homes in hotter, more humid regions tended to have shorter roof lifespans than comparable homes in cooler, drier environments.

None of those findings should be viewed in isolation. Combined, they suggest that long-term environmental exposure can influence property condition in ways that deserve greater consideration in underwriting.

A better question for underwriters

Property underwriting has traditionally focused on hazards surrounding a home. Is it exposed to hail? Flood? Wildfire? Wind? Those questions still matter.

The question that is becoming just as important is, what condition is this property actually in today?

Two homes built in the same year with similar construction can age very differently depending on the conditions they've experienced over time. One may have spent years exposed to repeated temperature swings. Another may have seen persistent humidity or heavier rainfall. Looking only at a property's age or location doesn't always explain those differences.

The industry has spent years getting better at predicting what a storm might do to a property. We're now much better positioned to understand what's happened to that property before the storm ever arrives.

Current aerial imagery and AI analysis make it possible to observe how properties change between quote, renewal, and claim. It doesn't replace traditional underwriting, but it provides another layer of context and data when evaluating current risk.

What this means for carriers

Catastrophe models are still essential. They're strongest when paired with a current understanding of how the insured property is changing between major weather events.

Hazard models explain the environment around a property. Current observations help explain how that environment may already be affecting the property itself. Looking at both together gives insurers a more complete view of risk than either can provide on its own.

That has practical implications across the business. Property condition can become part of renewal decisions instead of something evaluated only after a loss. Portfolio-level trends can help identify neighborhoods where homes appear to be aging faster than expected, even outside traditional catastrophe zones. Claims teams gain additional context about pre-loss conditions, while policyholders have more opportunities to address maintenance issues before relatively small problems become larger claims.

These aren't new objectives. Insurers have always wanted better information. What's changed is the ability to observe property condition consistently across large portfolios instead of relying solely on snapshots captured months or years earlier.

Looking ahead

One of the more interesting findings from the roof analysis wasn't simply that roofs age differently across the country. It was that the geography of climate exposure itself is changing.

The report found that U.S. land area in the highest rainfall-intensity band expanded from roughly 35,000 square miles during 1980 to 1984 to about 300,000 square miles during 2020 to 2024, an increase of approximately 750%. That doesn't mean every community faces the same level of risk. It does suggest that millions of properties are now experiencing environmental conditions that differ from what they were originally built to withstand. This reinforces the value of pairing historical experience with a current understanding of property condition.

Property insurance has always been about understanding uncertainty. Catastrophe models will continue to play a central role in that work. As the industry gains better visibility into property condition over time, insurers also have an opportunity to make more informed decisions between the next major weather event, not just after it.

Climate risk doesn't begin when a hurricane makes landfall. In many cases, it starts years earlier, one season, one temperature swing and one roof at a time.

The carriers that recognize those changes sooner will be better positioned to price risk, strengthen portfolios, and help policyholders address vulnerabilities before they become costly claims.

Sometimes, when a headline hits, you just know the story is going to drag on for months, even years. The rushed sale of the Los Angeles Lakers for $12.5 billion is one of those headlines. 

It has all the elements. There are big names -- the buyers are Bob Iger, the former CEO of Disney, and Josh Kushner... yes, the brother of Jared and, thus, by extension, tied to the president. The Lakers franchise is iconic. The price is the highest for any team in the history of sports. And the sale looks like just the beginning. Mark Walter, who is selling the Lakers, may have to unwind his whole sports empire, and we'll see a new wave of stories any time he sells one of the other iconic names in his portfolio: the Los Angeles Dodgers, the Premier League's Chelsea, the F1 Cadillac team, the WNBA's LA Sparks, and more.

This headline also, I'm sorry to say, includes the insurance industry. That's because the impetus for Walter's yard sale is a federal investigation into the at least $20 billion that he pulled secretly from insurance companies he controls so he could finance his sports empire. He hasn't been charged with any crimes, and the investigation could, of course, lead nowhere. There's also no indication at this point that others in the insurance industry are using their companies as banks, beyond what's allowed by law and routinely reported. But you can be sure that there will be lots of scrutiny both for insurers and, perhaps, for rating agencies.

Let's have a look.

The Wall Street Journal does its usual, thorough job of reporting all the complexities of the investigation into Walter and his financial services firm, Guggenheim Partners, so I won't recount them in detail here. I'll just note that the WSJ says there is about $1 trillion in private credit that insurers have disclosed they are providing to related entities, as Walter did, without disclosing the extent of those loans until recently. That's a lot of money.

As far as I know at this point, much of that stems from an open and seemingly smart approach that private equity firms are taking in life insurance. PE firms are buying life insurers and using their vast investment portfolios both to increase the PE firms' assets under management and to increase the yield that the insurers get on their investments. If PE firms produce better returns than the firms had been generating, then everybody wins.

Walter may, in fact, be able to offer that sort of defense -- what basketball players would describe as, "No harm, no foul." He led a group that bought the Lakers for $10 billion in June 2025 and sold the team just 14 months later for 25% more. He led the group that bought the Dodgers in 2012 for $2.15 billion, and the team is now valued at perhaps $8 billion. But not all the investments were winners. The WSJ says Walter used funds from his insurance companies to, for instance, make an early investment in Beyond Meat, whose stock price has fallen from $4,700 to $11. Walter also leveraged his sports empire to buy personal properties, including mansions.

We'll have to see what happens when Walter unwinds the $20 billion of loans that his firm now acknowledges it didn't disclose.. and any additional ones that come to light now that the Feds are investigating his businesses. 

But don't expect the scrutiny of him, or the insurance industry's investment practices, to go away any time soon.

Cheers,

Paul 

Paul Carroll

Your latest economic outlook struck me as reasonably optimistic, despite the U.S. tariffs and our war against Iran, but the environment changes awfully fast these days. Where do you stand now?  


Michel Léonard

I'm sorry to disappoint. We put out our Outlook a few weeks ago, and I've already turned more pessimistic. You know me — I lean optimistic by default. But here's what happened, and it all comes down to the Iran conflict:

When we put out our last quarterly estimates, everyone was talking ceasefire — and what it would mean for the P&C industry, replacement costs especially, and for the broader U.S. economy, inflation especially. Then the ceasefire evaporated, and the consensus swung back to the war grinding on. So we started repricing growth and inflation across the board — systemic, structural, economy-wide.

I say more pessimistic because I'm not ready to call a trend yet. You've heard me say it takes two quarters to make a trend — and we're not there. We won't have two clean quarters of data until late this year or early next, on what the Iran conflict has done to the economy, or is still doing by then. But we've clearly shifted from worrying mostly about GDP growth to worrying about price stability and jobs, too.

Paul Carroll

Oil prices haven’t risen as much as many of us expected, because various countries, especially China, have cut way back on consumption or dipped into strategic petroleum reserves. As a result, the oil shock hasn’t spread throughout the economy as much as it might have. But with the resumption of at least occasional fighting in Iran and the depletion of petroleum reserves and stockpiles for fertilizer and other oil-dependent products, are we reaching a breaking point where contagion can no longer be contained?



Michel Léonard

Yes, absolutely. We've been measuring how oil-price contagion spreads into the wider economy two ways: first, the damage already done, through existing PPI [produce price index] data; second, how it could unfold from here, through stress tests.

On what's already happening: higher oil prices have jumped the fence from energy into everything else — the contagion is real. Aluminum and petrochemicals, machinery and equipment, electronics and accessories, fertilizer manufacturing, plastic resins and materials — all up an average of 34% year-over-year over the last six months.

Paul Carroll

As the conflict drags on without a clear path to negotiations or a long-term solution, what do your stress tests say might happen to replacement costs in P&C insurance?



Michel Léonard

Our P/C replacement-cost stress tests ran three conflict durations — through end-2026, into 2027, into 2028 — and zeroed in on the three lines the Iran conflict is already hitting hardest on replacement costs: farm owners, commercial property, and personal and commercial auto. For those lines, we estimate replacement costs climb an average of 13% year-over-year if the war runs into 2027, and another 20% if it drags into 2028.

Paul Carroll

How should people use scenarios like this to plan?

Michel Léonard

In our last outlook, we said the whole point of stress tests is to build consensus around managing the severity of tail events — not their likelihood. Scenarios have to be credible, but not the most likely — and almost never are. 

Let a stress-test conversation slide into political debate about how likely each scenario is, and it's dead on arrival. Pick credible scenarios, skip the likelihood argument, and go straight to impact severity — that's how you actually get people to consensus on a risk-mitigation strategy.

For P/C, that brings us to actuarial analysis — backward-looking by definition. Here, that's a feature, not a bug. You anchor to the data in the rearview mirror, then work with regulators on where reserving and rates should go. The stress tests are the headlights — you use them after, mostly to set reserving or policyholder surplus above what regulators require.

Now, if you ask which of the three is most likely, I'll say this may be the most impossible conflict to forecast I've seen in decades. Put a crystal ball in my hand instead of structured analysis, and I'd still bet on the one-year scenario — into 2027.

But I'll say it again: each scenario's exact probability is beside the point, as long as it's credible. What matters for our audience is that nobody can argue the war couldn't drag on another year or two. So my advice: reserve adequately for all three scenarios — in a way that's economically viable for your own book and how it actually behaves.

Paul Carroll

No matter what happens in terms of costs, the outlook for growth seems fairly positive. Is that correct?



Michel Léonard

The pressure point is what the Fed does. Real tell: the new chairman's very first communiqué ended with a standalone line — "The Committee will deliver price stability." That's just not how these things are written. I've never seen anything like it. It signals rate hikes for an inflationary environment, it locks the Fed in. And markets didn't like that line one bit when there was no rate increase at the next meeting.

I won't get into the dynamics between the new chairman and the president who appointed him — or what the president wants, which everyone knows. But if the Fed hikes based on today's inflation drivers, that's a basic policy mistake. Rate hikes work best on demand-driven inflation; this is supply-side, so raising rates buys you almost nothing.

As Greenspan used to say, rate hikes would be the wrong medicine for the disease. here This inflation is the White House's to cure — not the Fed's.

In my view, a 25 basis point hike in this geopolitical climate would land with several times the force it would in normal times. It would slam the brakes on growth — tipping GDP toward a contraction and pushing unemployment toward 5%.

Paul Carroll

A fair amount could change on tariff policy and other issues based on what happens in the November midterm elections. How do you see these political dynamics affecting the situation?


Michel Léonard

I'm so glad you brought that up, because the tariffs are back. It's August, and out of nowhere a few weeks ago we've got tariffs on Canada and others again. I didn't see that coming — I thought we'd closed that chapter. The only way they make sense right now is to change the subject, from Iran-conflict inflation to protectionism. But it's not all performative — tariffs are increasing government revenues, real money coming in, more than expected back in 2022. 

Tariffs are inflationary by design. They won't do affordability any favors — and insurance affordability least of all.

The midterms and the rest of the DC noise you're pointing to — that's what tips the one- or two-year scenario.

The one-year scenario gets more likely if Democrats lead the polls into the midterms and take at least one chamber of Congress — this could cause a replay of 1980, when Reagan was already offering a better deal to the Iranians before he'd even won the presidency.

The two-year scenario gets more likely if the Democrats don't win back a chamber. You'd have a lame-duck president who'd probably want to end the war he started on his own watch —  or, again, we could have a rerun of 1980.

Paul Carroll


What happens if the Iranians show for the long term that they can keep the Strait of Hormuz mostly closed and that the U.S. can’t protect our longtime allies in the Middle East as much as we and they thought we could?

Michel Léonard

That changes the whole ballgame — the repercussions would run in every direction. But now I'd be trading structured analysis for the crystal ball again, so I'll stop there.

Paul Carroll

We’ve covered a lot. How would you summarize?



Michel Léonard



First, prices are already spiking — and not evenly; some costs aren't behaving the way you'd expect. Check the actual BLS and BEA data. That part matters. Second, P/C replacement costs are climbing, too, hardest in farm owners, homeowners, and commercial property.

Finally — and I'll end on optimism, like I promised — I think this slowdown is short-lived. As long as the Fed keeps its hands off rates.

And if you’re a Triple-I member, don’t forget that you can reach out to us directly. Membership has its privileges. 

Paul Carroll

Thanks, Michel. As insightful as always.

Insurance carriers rely on third-party providers for nearly every part of their operations, from underwriting and vehicle verification to claims processing and fraud detection. These partnerships can help insurers move faster, access specialized technology and improve the customer experience. However, they also create more opportunities for sensitive information to be accessed, transferred and stored outside the carrier's direct environment.

While an insurer can outsource a process, it cannot outsource accountability for the customer data involved. If that information is exposed, mishandled or used to support an inaccurate decision, policyholders are unlikely to distinguish between the carrier and the provider working behind the scenes. They simply see a failure by the company they trusted with their information.

That shift comes as recent research from The State of Incident Response Readiness 2026, found that 73% of organizations say they are not fully prepared for a major cyberattack, despite having incident response plans and security tools in place. The findings highlight that readiness depends not only on technology, but on governance, coordination, and operational discipline. That is why vendor security can no longer be treated as a narrow IT concern or a compliance requirement that is addressed once during procurement. It has become a core operational risk that can directly affect claims outcomes, fraud exposure, regulatory scrutiny, and customer trust.

Third-Party Risk Is Embedded in Insurance Operations

Outside providers are often deeply involved in the workflows that shape important insurance decisions. In auto insurance, for example, carriers may rely on third parties to verify title status, registration, ownership, vehicle identity, and lien information. These records can determine whether a claim moves forward, whether a payment reaches the correct party or whether a file should receive additional review.

Consider a total-loss claim. Before settlement, an adjuster may need to confirm that the claimant owns the vehicle, identify any active lienholders, and verify that the title record supports the proposed payment. When that information comes from an outside provider, the carrier is relying on more than the accuracy of the record itself. It is also relying on the provider's ability to collect, store, and deliver the information securely.

If the provider lacks strong access controls, reliable audit trails, or clear data-governance practices, the carrier may have difficulty determining who viewed the information, whether it was altered, or how it was used. Those gaps can create delays during the claim, weaken fraud investigations, and make it harder to explain or defend a decision later.

Although the relationship may be managed through a vendor agreement, the consequences of a failure ultimately belong to the carrier, not the contract.

Vendor Reviews Must Go Beyond the Checklist

Most carriers already have a process for evaluating third-party providers. That process may include security questionnaires, contractual requirements, insurance documentation, and annual compliance reviews. These steps are important, but they are not enough on their own.

A vendor can meet a basic procurement requirement while still creating risk in its everyday operations. A written policy does not necessarily show that employees follow it consistently, that access is properly restricted, or that vulnerabilities are addressed before they affect customers.

Carriers should look closely at how a provider's controls work in practice. Who can access sensitive information, and why? How is that access monitored? Is the underlying data pulled directly from the authoritative source, or aggregated and refreshed on a batch cycle? Can the provider show when a record was retrieved or changed? How does it respond when a vulnerability, outage or data discrepancy is identified?

Independent assessments and certifications can help carriers evaluate whether a provider has established repeatable security practices and whether those practices have been tested over time. Their real value is in helping an insurer gauge whether a vendor has the discipline and maturity required to support an important business process.

Security Also Includes Data Integrity

Security also means making sure the information behind underwriting and claims decisions is accurate, current, and traceable to an authoritative source. A record may be securely stored and transmitted but still create real risk if it is incomplete, outdated, or disconnected from the authoritative source. That matters in claims and fraud workflows, where a title discrepancy, undisclosed lien, or ownership inconsistency can change how a file should be handled.

Take a vehicle that was re-registered in a new state last month. If a vendor's registration feed lags behind the state's actual record, an adjuster pulling ownership data may still see the previous owner attached, delay payment to the current policyholder, or route a routine claim into manual review for no real reason. Nothing was breached. The record was just behind, and being behind creates the same downstream cost as a security failure.

According to Point Predictive's 2026 Auto Lending Fraud Trends Report, fraud exposure reached $10.4 billion in 2025, driven in part by synthetic identities, AI-generated documents, and title-related fraud. Fraudulent title documents, fake lien releases, and manipulated ownership records may appear legitimate during a routine review. As more underwriting and claims workflows lean on AI, the quality of the data feeding those systems matters even more. A governance gap at one vendor can now surface as errors across far more decisions, and much faster, than it would have a few years ago. Claims and fraud teams need to understand where the information came from, when it was last updated, and whether there have been changes that require further investigation.

This traceability gives adjusters greater confidence in their decisions and allows fraud teams to identify potential issues earlier in the process. It also creates a clearer record if a payment, settlement, or verification decision is questioned after the claim has closed.

For insurers, secure data needs to do more than resist outside access. It has to be reliable enough to support the decisions being made with it.

Operational Maturity Becomes Clear During a Disruption

The quality of a vendor relationship is often most visible when something goes wrong. A system outage, data discrepancy or security incident can interrupt claims processing and delay communication with policyholders. What matters most at that point is whether the provider can identify the problem fast, explain which systems and records are affected and communicate clearly through the response. A provider that can't answer basic questions about the scope of an incident becomes part of the disruption instead of the fix.

Carriers should settle escalation procedures before a problem occurs: who contacts the vendor, what information the provider has to produce and how affected workflows keep running while the issue gets resolved. That oversight shouldn't stop once the vendor is onboarded. A provider that starts out supporting one verification task can end up embedded in several claims, underwriting and fraud workflows within a year, and oversight needs to grow with the relationship instead of lagging behind it.

Customers Still Hold the Carrier Responsible

Most policyholders will never know which third-party systems are used to support their claim, but they will notice when those systems create a problem. A delayed verification turns a routine claim into a frustrating wait. A security failure exposes exactly the information a customer trusted the carrier to protect.

From the customer's perspective, it's all part of the same insurance experience.

That is what puts a stalled claim or an exposed record on the carrier's reputation, not the vendor's. The controls operating behind the scenes shape how quickly a claim moves, how confidently an adjuster communicates, and whether a policyholder keeps trusting the carrier once the process is done.

Insurers should evaluate vendors with the same seriousness they apply to speed, accuracy, and service. Strong providers build security and governance into how they modernize, so carriers keep control of the data behind their decisions instead of trading it away for speed.

In practice, that starts with two questions worth asking before the next disruption: what happens to a file when a vendor's system goes down, and how fast does a stale or disputed record get flagged before it reaches a customer? If those answers aren't clear today, that's the gap to close first.

Third-party partnerships will continue to play an important role in insurance. As that reliance grows, the real competitive advantage lies in choosing partners that can be trusted with the data behind every underwriting decision, claim, and customer interaction. Processes can be outsourced. Accountability cannot.