Recent announcements of two new robotaxi services in the U.S., together with a robotaxi-related surge in Tesla stock, seem to be raising lots of questions about how quickly autonomous vehicles (AVs) will become a major factor on our streets and about what they will mean for auto insurers.

Once you understand the "stack," to use the Silicon Valley term for the layers of hardware, software and operational service that make up a robotaxi service, the implications start to become clear.

So let's have a look. 

The reason for the recent surge of interest in robotaxis stems in part from last week's announcements by Amazon and May Mobility. Amazon said its toasterlike Zoox AVs will start operating in Las Vegas, while May Mobility said it will start offering rides in Atlanta in its autonomous vans, which will be offered to those who hail a ride through Lyft. 

In addition, Tesla fans are out in force again, boosting the stock with claims such as that the company will be able to remove safety drivers from its robotaxis imminently and can thus have thousands of the vehicles operating fully autonomously by the end of the year. 

As backdrop, there was a highly publicized report from Goldman Sachs in June that says fully autonomous robotaxis will be generating $5 billion of revenue in the U.S. by 2030. That's a lot of robotaxi rides.

I'd urge some caution on the pace. Zoox is just offering rides between five points in Las Vegas. May Mobility is operating only in a section of Atlanta, with what the CEO calls “a few hands’ worth of vehicles” and with safety drivers behind the wheel. Tesla is operating perhaps three dozen cars in Austin, Texas, all with safety drivers, and, while CEO Elon Musk said earlier this year that millions of autonomous Teslas will be on the road by the second half of next year, his projections have been wildly overoptimistic for a decade now. 

Still, robotaxis seem to finally be on a clear glide path toward widespread adoption. 

Google's Waymo says it has 2,000 AVs on the road in Phoenix, San Francisco, Los Angeles, Austin, and Atlanta, completing hundreds of thousands of paid rides per week. Waymo plans to keep steadily increasing the number of cities it serves. Wired magazine says that, "in China, WeRide, Baidu’s Apollo Go, and Pony.ai are all running robotaxis in multiple cities; WeRide has started operations in Abu Dhabi, too." Regulators in Beijing and Shenzhen are allowing the robotaxis to operate without safety drivers, and Abu Dhabi encourages AVs in special zones. So, while Europe is moving somewhat slowly, there is plenty of international competition to keep up the pressure for improvements in the technology and for deployment. 

What does this all mean for auto insurance?

Let's look at the stack. It's dividing up into the hardware (the car), the software (the artificial intelligence that operates the car) and the operator (the company that will keep the cars recharged, positioned so they can get to passengers quickly, cleaned, and so on). 

Insurers could have a role at any of those tiers but will likely have little to do with the software. In any case, any insurance will be commercial. Personal insurance won't be a factor for the simple reason that people don't need to insure their driving if they aren't driving. 

The hardware

For a time, General Motors tried to take on all three layers in the stack. It made the cars that used the AI developed by its Cruise subsidiary, which also operated a robotaxi service. But GM shut Cruise down last year and is focusing on incorporating its AI into existing lines of vehicles. Tesla is also trying to handle all three layers, while offering insurance, to boot, but as I've said several times now (including here), I don't think Tesla can pull it off. Musk is just using cameras as sensors, while others are using radar and LIDAR, as well. Musk does have better access to certain kinds of data than other robotaxi companies because he has cameras in every car and has so many Teslas on the road, but I don't see how having better maps can cover for the lack of real-time data from sensors that are more sophisticated than cameras. Tesla may eventually have usable technology, but Musk is years behind Waymo, and I think he'll stay there. 

If Musk does succeed with his grandest vision, he would introduce an opportunity for personal auto insurance in the robotaxi world. That's because he has said individual owners will be able to upgrade their software with Tesla's latest AI capabilities and make their cars available as part of a robotaxi fleet that Tesla would coordinate. Owners would presumably be responsible for the upkeep of their cars, so they might buy insurance to cover their liability.

But if I'm right that Tesla will be a minor player in robotaxis for the foreseeable future, then the hardware is a separate layer. It splits into two pieces: the manufacturing and the ownership. 

The role of insurance in manufacturing will be what it's always been. Car makers will have to worry about product liability and could purchase insurance, but the behemoths that do that sort of work will likely self-insure. 

Ownership could be a different question. Just about nobody wants to own assets these days. Everybody wants to be an asset-light company like AirBnB and not have to commit the hundreds of billions of dollars that robotaxis will cost. So it's not clear yet who will own the vehicles. If, as I suspect, huge companies — perhaps formed just for the purpose — own the cars, they'll likely self-insure. If not, there could be opportunities for commercial insurers.

The software

The AI is where the magic happens, and the work is so expensive that it is either being done by huge companies now or by startups that will surely be acquired by massive companies. So, yes, this layer will surely be self-insured. There's room for commercial insurance on, say, theft of business secrets but not for personal auto insurers.

The operators

Who will own this layer isn't quite clear yet. Waymo, for instance, is operating its own network at the moment, and that makes total sense at this stage of the technology. Waymo needs to have people in-house available when one of its vehicles runs into a problem, both to smooth over issues for paying customers and to learn where problems are, so the underlying AI can continually be improved. But Waymo and others won't always have to own the whole operating process.

Yes, Waymo will need to always be running the AI in its cars, but that doesn't mean it will have to handle the dispatching, the recharging, the cleaning and so on, and there could be opportunities for insurance there. 

I suspect the dispatching will be handled by big companies. Uber and Lyft have been doing a nice job of positioning themselves as partners to AV companies. Google could also muscle its way into dispatching through its maps — if you're asking about a destination, Google could easily offer you a ride. So there could be an interesting battle here (perhaps with antitrust implications for Google), but I don't think it will matter to insurers because the winners will self-insure against any customer issues.

The rest of the operations, though, could create opportunities for commercial insurers. The cleaning and recharging could well be handled by smaller companies, perhaps different ones in different cities, and they could well want to lay off some of the risk of dealing with customers, who can have any number of hard-to-predict problems. 

But, again, no luck for personal auto insurance.

As I've said, there are plenty of reasons to think that some of the claims about robotaxis are hyperbole, but if Goldman Sachs is even close to right that robotaxis will be generating $5 billion in revenue by 2030, vs. my back-of-the-envelope calculation of maybe $100 million this year, then we're on an exponential curve. And as all those companies that didn't come to grips with the pace of Moore's law have learned over the past few decades, exponential change can sneak up on you really fast. It's always wise to be thinking ahead.

Cheers,

Paul 

P.S. If you really want to think ahead, imagine what cities will look like if there is the sort of takeover by robotaxis envisioned in this article at Vox: "A self-driving car traffic jam is coming for US cities." 

The underlying changes will be deceptively simple. Parking garages, which take up as much as 40% of the space in some cities, will pretty much disappear. So will curbside parking. But the number of cars on the road will increase drastically. 

When you go from there, though, there could be all sorts of effects. Maybe apartment buildings replace the parking garages, and the additional housing makes cities less expensive and draws more people. Maybe having even more people in cities facilitates time in the office and leads to more office buildings, too. Maybe restaurants and other small businesses benefit greatly from increased foot traffic... or maybe they don't, because robotaxis are taking people straight to their destinations, so they aren't walking and window shopping. And maybe robotaxis make it easier to live in suburbs and exurbs, so city populations actually diminish and people work at home more.  

The good news is that cities won't change nearly as fast as the car fleet does. It takes a lot longer to tear down and replace a parking garage or rip up city streets than it does to beam a software update to a car. 

But I still find this kind of thing interesting to ponder and thought you might, too. 

The insurance sector is at a turning point. Once defined by legacy systems, complex actuarial models and decades-old policy structures, the industry now sits on the cusp of transformation powered by artificial intelligence (AI), including its subsets, generative AI (GenAI) and agentic AI.

According to EY, nearly 99% of insurers are either already investing in GenAI or exploring it due to its expected productivity, cost and revenue benefits, while KPMG highlights that 81% of insurance CEOs now list GenAI as a top investment priority despite economic uncertainty.

In this reality, we're seeing insurers move beyond proofs of concept into enterprise-scale adoption, unlocking outcomes across cost optimization, customer engagement and productivity. However, to truly embrace AI and its benefits, insurers need to rethink their approach to the operating model.

In this article, we'll explore why a product-aligned operating model is essential for scaling AI, where AI delivers tangible outcomes and the reinvention of the software development lifecycle (SDLC) with the ultimate goal of building long-term agility and growth.

From projects to products: How operating models are changing

Historically, change in insurance was delivered through projects. Teams formed temporarily around a scope and budget, handed off work across functions and disbanded at "go-live." Ownership was fragmented: Business wrote requirements, IT built, operations supported and data sat apart. That model optimized for completion, not continuous outcomes, and every new initiative restarted the learning curve.

Today, leading insurers organize around enduring products, including claims intake, quoting, billing, fraud detection and agent experience, which are each owned by a cross-functional team spanning business, data, engineering, design and risk. These product teams run on backlogs and objectives and key results (OKRs), ship frequently and treat AI, data and controls as integral. The shift concentrates accountability, shortens decision time and turns change into a repeatable capability.

The benefits are material. Product-aligned models reduce handoffs, embed governance where work happens and scale AI consistently across lines of business. They improve cycle time and quality, make investment transparent and help talent focus on customer and agent outcomes instead of internal coordination. For AI specifically, this model unites infrastructure, data and process expertise under clear ownership, giving organizations the trust, agility and repeatability required to move beyond pilots to production at scale.

The lesson here is that technology transformation must be matched by operating model transformation. Traditional structures, designed for incremental change, can't fully harness the potential of AI. That is why HCLTech's research found that 88% of surveyed businesses are moving toward product-aligned operating models.

Culture plays a decisive role. Those who embrace AI along with an operating model and cultural transformation will emerge as winners.

Where AI is delivering tangible outcomes

Insurance is inherently data-driven. From decades-long life policies to property and casualty (P&C) lines dependent on climate, location and risk data, the industry generates vast amounts of structured and unstructured information. Historically underused, this data is now being unlocked by GenAI, which can connect directly to disparate sources and derive insights without extensive re-engineering. What was once too expensive to modernize has suddenly become viable, enabling insurers to transform legacy systems, streamline claims and fraud detection and create new growth opportunities.

In this environment, there are three areas that stand out where insurers are realizing measurable value today:

1. Driving productivity and reducing costs

AI-powered platforms are streamlining IT operations, the software development lifecycle, QA and testing. Productivity improvements range from 12–15% up to 40–45%. For example, AI-assisted testing and code generation have cut cycle times significantly.

2. Enhancing customer and agent experiences

Whether in contact centers, claims processing or agent interactions, AI is reimagining engagement. Automation is not just about efficiency; it's about building more intuitive, personalized journeys.

3. Empowering the workforce with AI assistants

Digital assistants for underwriters, claims analysts and agents are emerging as powerful tools. Rather than replacing human expertise, these AI co-pilots augment decision-making with real-time insights and recommendations.

These outcomes are why 65% of insurers expect AI to deliver revenue lifts of over 10%, while 52% anticipate cost savings.

Moving from experimentation to scale

For several years, insurers explored AI through proofs of concept. That period of over-experimentation is now giving way to a new phase: implementing AI at scale to deliver enterprise-wide impact.

Scaling AI, however, is not just a technical challenge; it is an organizational one. Insurers must start by establishing a clear value realization framework. Without a baseline, it is impossible to track benefits such as cost savings, productivity gains or customer experience improvements.

Equally important is organizational change management. AI alters workflows, including how underwriters assess risk, how claims are processed and how customer service agents interact with policyholders.

In underwriting, for instance, AI is already enabling faster, more accurate risk assessment and reducing time-to-quote. Similarly, in group insurance, AI-driven automation is streamlining the quoting process, cutting cycle times and improving pricing accuracy. Unless employees are engaged and supported through such changes, adoption falters.

Responsible AI must also be embedded from the outset. Governance frameworks, regulatory monitoring, bias mitigation and continuing risk assessment are critical in a sector where trust is paramount.

Success will hinge on culture. Organizations that treat AI as an isolated initiative risk marginalizing its potential. By contrast, those that democratize AI by placing tools in the hands of underwriters, claims handlers and IT engineers foster adoption at scale.

Redefining the software development and IT operations lifecycle

One of the less visible but highly important areas where AI is transforming insurance is the end-to-end software development lifecycle (SDLC). While many organizations deploy point solutions for specific stages, the real opportunity lies in orchestrating AI across the entire lifecycle.

Consider the chain reaction: Inaccurate requirements gathering leads to flawed code; flawed code creates more defects in testing; weak testing allows problems into production. From demand capture and code generation through QA and release, embedding AI throughout the lifecycle enables insurers to improve quality, reduce cycle times and lower costs.

Similar benefits extend into IT operations, where insurers are moving away from traditional machine learning models toward agent-based automation. These adaptive systems empower administrators to build agents that can "skill themselves on the fly," creating resilience in run environments.

Building long-term agility and growth

AI is no longer a futuristic ambition. Instead, it is a present-day competitive differentiator. It enables insurers to cut costs, accelerate modernization, elevate customer and agent experiences and empower employees with intelligent tools.

But success will depend on more than technology. It requires clear value frameworks, responsible governance, cultural adoption and new operating models. With KPMG finding that 62% of insurance CEOs citing talent gaps as a barrier to growth, investing in people is also crucial. Here, AI should be seen as a partner to human expertise, not a replacement.

The winners in insurance will be those who seize this turning point to not only re-engineer processes but also reimagine possibilities. AI is not just reshaping the industry; it is redefining its future.

The idea of property and a neighbor is easy.

The idea of digital and cyber and a neighbor is hard.

The first set is visible. The second is invisible. One exists in real space. One is ethereal.

Property - it’s houses, cars, trucks, machines, buildings, businesses, infrastructure, products, and more. Real things with real people and real-world locations - a very physical world with names, addresses, and contact information. Things you can see. Things you can touch. Very relatable parts of any real neighborhood things. (Save intellectual property for another time).

Digital and cyber are not that. They exist in an e-world where everything is e-real. 

Every memory storage location and every processing chip can be thought of as having their own names, addresses, and contact information, but only in an e-real way. On the internet, there are IP addresses. The physical device could literally be anywhere, but in compute logic, it’s all just a bunch of slashes and dots away from any other device any-e-where as one address can route and link to others.  

A home or building sits on land or a lot all uniquely assigned mutually exclusive coordinates. Data scattered among redundant arrays of independent disks do have their own addresses with 1’s and 0’s, but these can be overwritten or even erased as well as copied and stored in multiple places which may move around. Similarly other storage mechanisms, including cloud storage, are also in play. Everywhere your data move may be considered another neighborhood, and if their prior instance is not scrubbed, that ghost trace is a latent neighbor that you didn’t know you didn’t know about.  

These addressable endpoints also include situational features - like operating system, software version, patch sequence, and other options that reflect what is going on at each endpoint. As with an open window on a rainy day, you would have wanted it closed if you knew what was going to come through it. The stakes are higher when an interloper is looking for open doors and windows, or as digital/cyber relates, ports and vulnerabilities. 

Just as a house has doors and windows that can be locked or left open, each digital address (IP address) has services and software that can be secured or exposed. But even locked doors can have weak locks or hidden flaws—some locks are easy to pick, and some windows can be forced open. Similarly, even protected digital services can have vulnerabilities that skilled attackers can exploit.  Sometimes an indirect approach is easier, like posing as technician to a call center representative to open a door unwittingly.

We are becoming more comfortable with the concept that digital equals information, digital equals data, and digital channels are ways of interacting with these. 

We are in a transition to the mindset that everything now is data…. Desktop and remote is how we imagine and represent the people, places, and things in any real neighborhood.

But we are just at the threshold of understanding that these representations stored in the ether of the internet are living in invisible cyber neighborhoods.

We can think of a cyber neighborhood where every computer core or memory storage device in a chip, circuit board slot, machine, server, rack, network, sub-network, datacenter, platform, cloud, cloud region, etc. is like a real-world rooftop address geolocation or even a rally point like a pin drop or a WhatThreeWords Earth pixel. 

The programmers, administrators, hackers, programs, bots, code, communications, protocols, APIs, and AI agents are neighbors under those rooftops and around those locations. 

Some compute environments are like an owned and occupied home by the same person for decades, others are like a rent-by-time-slice hoteling office, and some are like a dark alley or underpass with shady dealings and no identity required. 

(Read also: “No one can hear an AI scream in cyberspace…” from ITL.)

The reality of “bad neighbors” in the real world and “bad neighbors” in cyberspace is stirring the insurance world and the risk marketplaces. 

There has been a sector rotation in cyber criminal appetite turned toward P&C this year, and an unfortunate horizontal weakness is currently in active exploit with a popular CRM system product. Whether targeted or opportunistic, the e-safety of the insurance neighborhood cannot be taken for granted. 

The idea of a safe neighborhood or a dangerous one can transfer between real and e-real constructs.  Safety as an index can be ephemeral when exposed to a threat and quickly remediated, or it can be structural and lie undetected while exploited at scale, a false presumption of safety. When, not if, hidden exploits are uncovered, the assessment and remediation processes cycle anew.

The risk of the e-world is constant and global. This is unlike real world perils like watching the track of a hurricane, which is seasonal and geospatially proximate.

Primacy and recency of cyber threats are the constant reminders of what is less imagined - our digital neighbors in our digital neighborhoods are in a continuous state of invisible digital churn. Any time we share any digital resource, there are others sharing it, too.

While there may be some examples of isolated computing with no connections, communications, hosting, integrations, or application programming interfaces, the most common enterprise IT situation is multiple core systems interacting on premise and intra/inter cloud resources with vendors, third parties, and partners.

It is difficult to delve into the wildness of internet cyber situations; some are inherent, while others are sporadic. Some are software- or hardware-related that appear accidental with incidental vulnerabilities, and others are thoughtfully crafted exploits by human ingenuity, now adding AI capabilities.

Regardless of the nature of the cyber risks, the level of connectedness and the risk across connections may vary user by user, company by company, machine by machine, software by software, interface by interface, network by network, platform by platform and cloud by cloud.

Like people and businesses occupying houses and buildings in the physical world at literal addresses using a variety of names and aliases, the digital world can be seen in a similar fashion.

Company computer infrastructures and their cyber vulnerabilities span a spectrum of more fully controlled with more uniform homogenous cyber risk (walled garden and locked down with dedicated security and engineering) to widely distributed with dynamic heterogeneous cyber risk (hosted on multiple platforms with multiple networks with different management systems and software and haphazard oversight of many participant digital neighborhoods and denizens (people, businesses, robots, and AI agents, etc.).

From a moated castle to a flea bag hotel the risk of both the infrastructure and the neighboring occupancy is an analogy of the consistency or inconsistency of cyber risk, which will vary over time. A bad actor can get into a castle but then be confronted and mitigated. But a bad digital neighborhood leaves more at risk more of the time.

What is invisible to the eye is the infrastructure connectedness of extended digital networks. Many castles working together may tunnel to each other. Many discount motels may do the same. Throw in a crime-ridden abandoned building drug den and you get a deteriorating sense of what could be out there - invisibly except for digital means.

So... a long wind-up.

Extending the analogy just a bit further, some digital means look at all the doors and windows of all the spaces known to belong to a company or to be transacted by the company and another. But these approaches don’t include all the adjacent and proximate spaces to those. These are “glass partly full” covering approaches that combine strength and efficiency but lack comprehensiveness. 

The concept of watching and recording hundreds of millions of internet domains and billions of interactions between them and archiving those observations across a decade and more seem too large for assessing any single company’s risk. But someone has done it, for a different business reason than cyber assessment. Now comes the serendipitous epoch of cross purpose innovation - re-purposing an existing asset for a new use case.

The fabric of a connected, internet-wide data infrastructure permits the rollup of sub-networks, networks, domains, and “ultimate domain,” which tie information across the digital world into a form where it can be linked to legal entities. This is where cyber risk at each digital rooftop can be assessed and aggregated to a building, block, tract, region, and so on to score the whole of the risk as an algorithm of consistency over each of its parts. These parts can be associated logically to the legal entity level and a new understanding of cyber risk can be attributed, aggregated, and accumulated like never before.

This capability to assess organizational risk across complex and otherwise invisible connections is novel and useful. As cyber threats change over time, and legal/digital entities also change over time, the continuing dynamic assessment adapts and creates information to act on.

Turning data into decisions and actions makes this process valuable. And that value can be achieved by incorporating these data, analytics, or both, into modern digital and cyber analyses and risk management and monitoring solutions. Using multi-level risk scoring that can count and analyze the number and severity of vulnerabilities at each level will let you see not just where the problems are but how serious they are.

Ensembles of data and analytics most always deliver more robust solutions.

My old friend and colleague Chunka Mui recently posted a thoughtful essay on how companies should start thinking about the next phases of the generative AI revolution — which is where the profound changes will happen. 

By now, just about every company is experimenting with gen AI, and many even have gone beyond pilots and into production. But, drawing on Doug Engelbart's classic thinking about businesses' "A, B and C processes," Chunka lays out the need to go beyond AI's usage in what Englelbart would call A processes — those that a company uses to operate every day. Companies need to institute B processes, which are designed to improve those processes that run the business. And — here's the real payoff — companies need to design C processes to improve the B processes.

I know that can sound rather theoretical, but Chunka shows how it all gets practical very fast — and the approach has worked before. Drawing on Engelbart, among others, Chunka was the co-author of "Unleashing the Killer App: Digital Strategies for Market Dominance," which was a huge best-seller and a sort of bible for innovators in the internet boom of the late 1990s. Even after the bubble burst, the publisher of the Wall Street Journal lauded the insights in 2005 and labeled "Killer App" one of the five best books on business and the internet.

I'll give you the short version of Chunka's thinking and apply it to insurance, leading up to his seven questions that will help ensure that you're seeing the full potential for generative AI in your business.

To put insurance terms to the ideas in Chunka's piece, A-level processes are how agents and brokers sell, how underwriters price risk, how claims are settled and how customer service centers operate. We've all seen stories, and probably even personally experienced, how AI is being deployed at this level.

B-level processes improve those processes, and it's easy to see how gen AI can make the improvements happen even faster. The AI can, for instance, instantly spot patterns in responses to sales pitches to see what works and what doesn't work, including nuances such as time of day, day of the week, number of weeks or months before a renewal, proximity to a life event, etc. The AI can detect emotions that humans may miss as potential customers talk; the AI can then pass the information to agents, helping shape the conversations. And so on. The AI can also speed the process by which learnings are gathered, distilled and fed back to agents so the A-level processes improve and agents can be more effective next time. The same sorts of B-level improvements can happen in claims, underwriting, customer service and other parts of insurers' businesses. 

From what I've seen, many companies are at least starting to think about B processes. I've published quite a few pieces about, for instance, using AI to catch fraud, to have results from claims fed back to underwriters to improve their appraisal of risks, and to help underwriters both gather data more efficiently and to highlight changes in the policyholder's situation since the last policy was issued.

But I have yet to hear about much in the way of using AI to get to the next level, to the C processes. They're a bit harder to characterize but are crucial. As Chunka writes, "C-level work isn't merely about scaling incremental improvement—it enables organizations to question and redefine their very purpose. It allows not just better performance, but different futures. C-level improvements accelerate the rate and type of change—unlocking exponential leverage."

From the initial internet boom, I'd say Amazon is the best example of C-level thinking. It started out selling books and continually worked to sell books more efficiently — showing A-level and B-level processes — but was always driven by a C-level vision that founder Jeff Bezos referred to as "The Everything Store." He wanted to sell everything to everybody, even as he founded the company more than 30 years ago.

Amazon Prime was a direct outgrowth of that vision. Once Bezos started to host enough other businesses on the Amazon site, he saw he could lock in customers by offering them fast delivery based on an annual fee — getting them out of the habit of factoring shipping costs into every purchase. That lock-in then helped him attract more merchants, feeding a virtuous circle that continues to this day. 

AWS wasn't foreseeable back in the early days of Amazon but is the sort of happy accident that can happen when you set out for a C-level reinvention rather than just a B-level continual improvement. Bezos saw that many merchants needed help operating their sites, so he started a cloud service — and being in the business early let him see the huge demand before potential competitors and get a massive head start that has translated into a business that generated $108 billion of revenue last year, with an operating margin north of 35%.

For insurers, I could see a C-level approach to gen AI facilitating the move toward a Predict & Prevent model, beyond today's repair-and-replace approach to risk and losses. Gen AI can gather information — even across the silos that bedevil insurers — and analyze it instantly, then send it to whomever needs to have it, in time to perhaps prevent a loss.  

A well-meaning recent attempt to get bad drivers to improve was based on a single communication to individuals with multiple moving violations, whose behavior was then monitored for the next six months. It won't shock you that driver behavior changed not at all. What we need is the sort of instant information that Nauto provides to truck fleet drivers about speeding, about tailgating, about drowsiness, about road conditions and accidents ahead, etc., based on AI analysis of images from cameras: one facing the road, one facing the driver. A C-level approach to innovation with gen AI can facilitate that sort of timely feedback — and not just for drivers. It can also help, for instance, the timely provision of information to utilities about faults in electric lines, as detected by Whisker Labs' Ting sensors in people's homes. A C-level use of gen AI could help communities monitor and encourage homeowners to harden their properties against wildfire, reducing the risks for everyone. And so on.

More generally, gen AI can be used to flesh out the sort of what-if scenarios that business leaders use to stretch their thinking and prepare for challenges and opportunities. Instead of just briefly entertaining the thought of a recession, of war spreading from Ukraine to other parts of Europe, or of even more remote possibilities, leaders can use gen AI to develop more elaborate scenarios and explore the complex interactions that may matter to a business but that are hard to see in a quick look. Even at huge companies that have planning departments, gen AI can help flesh out scenarios faster — gen AI could look at today's weak jobs numbers in the U.S. and speculate in detail on what it means for workers' comp enrollment, for employee-sponsored healthcare programs, for general economic growth, for Fed rate cuts and more.

"Killer App" explained the power of what-if analysis, in one of the many parts of the book that have stuck with me. Chunka said the invention of the electronic spreadsheet in the late 1970s led directly to the wave of mergers and acquisitions in the 1980s and 1990s. Why? While smart young financial analysts could always crunch numbers, they previously had to manually update every cell in a spreadsheet if an assumption changed. With the electronic spreadsheet, they could let their imaginations run wild.  They could just set an interest rate or a sales figure or cost savings or whatever and have the assumption ripple through a spreadsheet until the analysts got the sort of result from a potential merger or breakup that they wanted. Their bosses would then sell the idea to companies or aggressive investors — and watch the fees roll in. 

Chunka, boiling down his thoughts on the A, B, C processes, suggests these seven questions that you should ask yourself to make sure you get the full benefit from generative AI:

1. Are we using AI only to do the same work faster, or are we also using it to design entirely new ways of working?
2. What systems and processes do we have to spread AI-driven learning and improvement across the organization?
3. How are we actively identifying and challenging the assumptions baked into our current workflows, products, and business models?
4. Where could AI help us fundamentally reimagine our business model—not just optimize existing operations?
5. Who is accountable for leading and sustaining C-level improvement—and do they have the authority and resources to act?
6. How are we ensuring that AI adoption does not quietly encode and scale harmful biases, flawed assumptions, or misleading correlations?
7. Do we have the culture, skills, and adaptability to continually improve how we improve?

He writes, "The real prize is using AI to redesign the road itself—not just drive faster on the old one."

Cheers,

Paul

P.S. I've told my Engelbart story before, so I'll just reprise it briefly here.

In the late '90s, I attended a cocktail party at a friend's house in Silicon Valley and struck up a conversation with an older man, who expressed interest when I told him I edited a magazine for Diamond Management & Technology Consultants that focused on innovation through digital strategy. When he asked for an example of the sort of article I published, I told him I had just edited a piece on A, B and C processes. 

"But that's my idea," he said.

"That's Doug Engelbart's idea," I replied.

"And I'm Doug Engelbart," he said.

He was, too. Engelbart, one of the most celebrated of the pioneers of personal computing, lived next door to my friend. 

AI is revolutionizing the way insurers deal with underwriting and claims management. However, adoption still faces barriers that go beyond implementation. 

The most frequent blockers in adopting AI are not technological. Though insurers start AI projects with trusted vendors and a clear understanding of why the work is necessary, many stall. Our team has seen underwriters across commercial and specialty lines hesitate to rely on scores generated by AI. At the same time, claims teams worry about the possibility of using models to affect settlement decisions.

A Pilot That Stalled — And What Changed

I find it challenging to persuade underwriters to trust AI-generated recommendations. In a commercial P&C insurer pilot project, the AI model was ready in four weeks, but rollout stalled for several months because underwriters didn't trust scores without context. Adoption only took off after we explained how our AI advisor worked in real life. For this, we asked our partner to provide 4,000 historical data points, which we then used to train the AI model. Also, we did the following:

• Showed the top factors influencing AI scores
• Allowed underwriters to override AI outputs
• Offered to keep an audit trail of all recommendations and decisions
• Embedded AI results in the tools they already use

As a result, we've got a data-rich advisor that calculated triage, appetite, and winnability scores in a matter of minutes, but more importantly, a solution that underwriters trusted enough to start using. Such trust turned a pilot into a full-fledged software product, taking underwriting processing to a new level.

Transforming Manual Workflows Into Digital Journeys

In the case described, AI helped underwriters transform traditional, often outdated and manual, processes into an automated digital journey. Triage scores are calculated more accurately as the platform ensures that data is complete. Appetite matches submissions against preferred segments and considers the company's guidelines and rules. Winnability predicts the likelihood of winning the deal. All scores are calculated automatically, saving underwriters' time for final decision-making.

Overcoming Fears of AI Replacing Professionals

Another challenge is the fear that AI could replace underwriting and claims specialists. The key is to convince underwriters that AI is a helper rather than a rival. On a project that required a certain level of automation in claims, our solution was to integrate natural language processing to extract data from documents supporting claim submissions from customers. As a result, claim managers have 25% more time for complex cases requiring more attention and direct communication with clients.

Asking for feedback is also crucial. It allows you to discover when AI predictions and recommendations were right or wrong and use that information to refine the models. And when people see their feedback improve models, trust accelerates.

In measuring the impact of AI in underwriting and claims, it's not about providing ROI to leadership. It's more about building credibility, so the people who use AI believe it works.

From our experience, here's what works for measuring the impact of AI:

1. We measured the current state before AI was introduced (average triage time, claim cycle time, loss ratios, etc.)
2. Together with customers, we tracked the usage rate and override frequency
3. Our experts looked for early wins during one quarter to scale further

Success doesn't mean integrating complex algorithms only. It comes from addressing AI adoption challenges, delivering measurable results, and building solutions that insurers trust.

62% of HR professionals who participated in SHRM's 2025 study stated they operate beyond their normal capacity. Such heavy workloads can lead to burnout, hindering HR teams' ability to manage human resources efficiently, which can damage entire organizations.

Optimizing HR tasks with intelligent process automation (IPA) can help alleviate excessive workloads, prevent burnout, and improve efficacy. Unlike traditional process automation, which solely focuses on mechanizing structured, rule-based processes, AI-enabled IPA can be applied to a broader range of HR activities, including those requiring intelligence. Forward-thinking business leaders have already recognized the potential of IPA for enhancing various HR processes, with streamlining HR workflows with automation and AI being their top priority for the next one to two years.

Based on our experience in HRMS software development, here are key concepts of intelligent automation in HR, its common use cases, and implementation best practices.

Key technologies for intelligent automation in HR

● Robotic process automation (RPA)

Traditional RPA bots are already widely used in HR to automate repetitive, rule-based tasks, but they can't handle more complex and non-linear processes. Augmenting RPA bots with AI models enables them to process both structured and unstructured information, make informed decisions, and learn from data, which allows companies to streamline more time-consuming HR tasks.

● Smart assistants

Smart assistants use artificial intelligence capabilities to understand user queries formulated in natural language via voice or text and respond to them accordingly, from providing information upon request to performing actions across corporate systems, saving HR specialists' time.

● Personalization

Nearly 20% of employees surveyed by McKinsey in 2025 reported dissatisfaction with their employer, while 7% expressed a desire to leave their jobs, which can pose a risk of quiet quitting. Tailoring employee experiences to their unique needs and preferences is one way to make staff feel more valued and satisfied, which can improve retention.

However, providing personalized support services, career development opportunities, and wellness programs can be exhausting for HR teams already operating under heavy workload. Leveraging AI-enabled tools equipped with experience personalization capabilities is a way to address this challenge.

● HR data analytics

To identify employee skill gaps, detect turnover risks early, and make informed workforce management decisions, HR teams need to analyze large amounts of workforce-related data, which can be challenging due to its ever-increasing volumes. Automated AI-enabled data analysis tools can support HR teams in processing relevant information and generating data-based insights to accelerate and enhance HR decision-making.

Common uses for intelligent automation in HR

● Recruiting

Recruitment is often considered the most critical yet complex and time-consuming aspect of human resource management. Luckily, many recurring recruiting tasks can be streamlined with the help of intelligent process automation tools.

For example, Majid Al Futtaim, a Dubai-based retail and leisure company, leveraged a set of IPA technologies, including experience personalization and HR data analytics tools, to build a more efficient and smooth hiring process. Now, they use AI to automate candidate scheduling, personalize candidate communication, assess candidate fit for the company's culture, and even predict their likelihood of success in different roles. "We've reduced our time to hire by 30% with AI. AI has also driven a significant improvement in the quality of hires, ensuring that every new team member aligns with our culture and contributes to our vision," said Mai Elhosseiny, vice president of talent at Majid Al Futtaim.

● Onboarding employees

Onboarding newcomers is another time-consuming activity that IPA can optimize. A prime example is Santander, an Argentina-based financial services company that hires between 50 and 100 employees per month. Onboarding used to be performed manually and sequentially and required an average of six weeks per person. Santander automated this process with intelligent RPA bots, which can automatically inform relevant departments about new team members, set up employee accounts, and perform the necessary compliance checks for each new hire. As a result, onboarding was reduced to just two days.

● Supporting employee talent development

Talent development can create excessive workloads for HR teams. But AI-enabled IPA tools have already proven efficient for optimizing diverse aspects of talent development, from performance appraisal to internal talent acquisition and training personalization.

Kuehne+Nagel, a Switzerland-based logistics provider, intended to enhance visibility into career development opportunities for nearly 78,000 employees across 1,400 locations, streamline internal recruiting processes, and eventually stimulate internal mobility. To achieve these goals, the company decided to implement an AI-enabled internal talent marketplace. After employees fill in their profiles within the new system, AI algorithms automatically analyze their data, match it with open learning and job opportunities, and provide recommendations. The AI system also generates analytical insights for recruiters, which helps assess the company's current talent needs, evaluate talent gaps and strengths of employees, and identify internal candidates best-suited for specific jobs. The tool already helped the company increase conversion rate for internal candidates by nearly two times, while decreasing the time required to fill for internal requisitions by 20%.

● Handling employee queries

Companies can apply AI-enabled automation tools to handle various types of employee queries, including questions about benefits and training programs, time-off requests, and medical document submissions. 

Covestro, a German manufacturer of high-tech polymers, was looking to expedite the processing of sick leave certificates submitted by employees. Manual processing took an average of seven minutes, which was too time-consuming, given that HR teams received more than 500 certificates per week. The company deployed AI-powered RPA bots, which can classify submitted documents as sick leave certificates, extract necessary data, and then input it into employee profiles in the ERP system. As a result, Covestro saved 85% of the time HR teams previously spent on manual sick leave submissions processing.

Useful practices for implementing intelligent automation in HR

● Implement process intelligence tools

Automating the right processes is crucial, and AI-powered process intelligence tools can identify HR activities most suitable for automation.

These tools can provide process mining and task mining capabilities helping identify bottlenecks within workflows, visualize business process data for stakeholders involved in a project, and even predict the impact of automation on specific tasks.

● Start with a pilot automation project

Conducting a pilot IPA project allows companies to validate the feasibility of IPA without incurring expenses associated with a full-scale implementation, detect hidden automation pitfalls and hurdles early on, and lay a strong foundation for larger IPA initiatives.

For a pilot project, companies should select one or two HR-related processes and establish clear KPIs to measure the impact of automation. Companies need to carefully evaluate the project's success, collecting stakeholder feedback and analyzing lessons learned.

● Communicate IPA benefits to employees

According to Deloitte's 2025 report, companies are six times more likely to achieve a financial advantage from AI when employees feel they personally derive value. However, the same report reveals that 77% of companies do nothing to share the improvements AI can bring.

Managers spearheading IPA adoption should work closely with HR teams from the start of a project, articulating the improvements IPA can bring to their work and keeping them informed about the progress and impact of IPA.

Modern HR teams regularly struggle with excessive workload, which hurts their productivity. Implementing intelligent process automation to optimize a range of recurring HR tasks is a way to alleviate pressure and enhance the performance of human resources departments. 

As of 2025, more than 21 million applications for new small businesses have been filed in the U.S. But behind this growth lies a serious vulnerability: underinsurance. Just 13% of small business owners with insurance coverage feel fully prepared for risk.

This protection gap should be a wake-up call for insurers. Small and midsize enterprises (SMEs) are a rapidly growing segment, yet many remain underprotected, leaving them both vulnerable and underserved. For carriers, this is an opportunity to support a critical part of the economy while capturing growth.

But closing this gap demands a shift in strategy. Insurers must rethink how they engage small businesses, delivering solutions that are seamless, timely, and integrated into the systems SMEs already use to run their operations.

Insurers that act now can turn this unmet need into a growth engine and position themselves as trusted partners in an expanding market. Wait too long, and you risk a generation of business owners moving forward without you.

Why traditional models are failing SMEs

Consider a family-run coffee shop. For years, business has been steady and incident-free, so insurance feels optional — until a small electrical fire forces the café to close for three weeks. Without coverage, the owners pay out of pocket for repairs while losing revenue, and what once seemed like a "low-risk" business suddenly faces a financial crisis.

This story is not uncommon. Many SMEs put off purchasing insurance beyond the minimum required because the process feels inaccessible. Policy language is dense and full of jargon, leaving owners unsure of the difference between general liability, professional liability, and workers' comp. Nearly 70% of small businesses report struggling to understand coverage limits, leading to insufficient protection from the start.

Traditional distribution models compound the issue. Legacy carriers have established reputations that garner SME trust, but often rely on outdated, paper-heavy processes that feel inaccessible to busy business owners. Newer digital-first carriers offer sleek self-service platforms, yet many lack the credibility of heritage names. The result? SMEs are left without coverage that feels both reliable and convenient.

This disconnect doesn't stem from disinterest. In fact, 82% of small business owners say insurance coverage for their business is extremely or very important for their operations. The demand is real, but current products and channels don't meet the realities of SME size, budget, and needs.

Insurers have an opportunity to redesign coverage to reach businesses that traditional models have left behind. A digital-forward, personalized strategy will meet SMEs where they are by simplifying the path to purchase while building trust and long-term loyalty.

Three ways carriers can close the SME protection gap

SMEs remain underinsured because the insurance buying process often works against them. Policies are hard to compare, language is overly complex, and support isn't built for fast-moving businesses with limited resources.

Carriers that remove these friction points are better positioned to meet SME expectations and capture a largely underserved market. Here are three strategic moves that can help make it happen.

1. Get in on the ground floor of embedded insurance

Embedded insurance meets business owners at the exact moment they need it, delivered through the tools they already rely on, like accounting software, e-commerce checkouts, payroll platforms, and registration sites.

Rather than requiring a separate search or offline process, coverage options appear contextually, right where decisions are being made. This reduces friction and reframes insurance as a natural part of operations, making SMEs more likely to see its value and take action.

The opportunity is significant: Embedded insurance is projected to generate over $70 billion in gross written premiums by 2030. Investing now will position you at the center of how SMEs evaluate and manage risk.

2. Tailor communications to real business needs

Even when insurance is embedded at the right moment, the message still needs to resonate. Many SMEs don't know where to start when it comes to shopping for policies, and broad, generic messaging doesn't help. Businesses face industry-specific risks, so one-size-fits-all offerings leave owners unsure whether coverage really applies to them.

AI and data analytics are helping insurers change that. When connected to platforms that small businesses already use to manage finances, payroll, or HR, insurers can access real-time signals to tailor outreach based on how each business actually operates. A freelance graphic designer may benefit from professional liability coverage, while a growing food truck fleet is more concerned with commercial auto and workers' comp.

Personalization also helps business owners understand why coverage matters. When SMEs see that you understand their unique risks, insurance becomes less of a generic add-on and more of a practical safeguard for the business they've worked hard to build.

3. Balance digital tools with human connection

Even with embedded distribution making it easier to access coverage, trust is still earned through human connection. SMEs need the flexibility to start online and pivot to an advisor when questions or concerns arise.

Routine tasks like requesting certificates of insurance or updating information should be fast and self-service. But when it comes to claims or complex purchases, SMEs should have easy, immediate access to licensed advisors who can provide personalized, empathetic support.

Making human interaction a built-in feature strengthens SME confidence and drives long-term loyalty.

How to own the SME protection opportunity

Closing the protection gap is a win-win: Small businesses get the protection they need to weather setbacks, while insurers tap into one of the most dynamic and underserved markets.

To seize the moment, carriers must focus on embedded offerings that deliver personalized service. This ensures coverage aligns with the real, current needs of small businesses while remaining accessible and trustworthy.

SME growth shows no signs of slowing. Now is the time to rethink how protection is delivered, move beyond outdated models, and earn lasting trust from the entrepreneurs powering the next wave of economic expansion.

In an industry obsessed with innovation, one of our most under-leveraged opportunities lies in something as old-fashioned as human conversation.

Take long-term care planning (LTC). Despite decades of sobering statistics, consumer education, and product development, LTC Talk Avoidance Syndrome remains alive and well—not among consumers but among the very professionals meant to guide them.

It's a systemic issue. And for insurance executives and innovators, it's also a strategic opportunity.

Long-term care risk isn't a niche issue. According to the U.S. Administration for Community Living, nearly 70% of Americans over age 65 will need some form of long-term care, yet only a fraction are financially prepared. According to a LIMRA summary from late 2024, just 3% to 4% of adults over age 50 have some sort of insurance to mitigate the LTC expenses.

The need is obvious. So why aren't more clients protected?

Because far too often, the conversation never happens.

Financial professionals avoid the topic for fear of upsetting clients, getting bogged down in emotional resistance, or simply not feeling equipped. Clients avoid the topic because the implications are uncomfortable, the costs are intimidating, and the future is always "later."

This conversational gap isn't just bad for families. It's bad for business. Every missed LTC planning conversation is a missed opportunity to build trust, create loyalty, and provide meaningful risk management.

While many in our industry focus on performance, pricing, and product features, the most powerful differentiator may be something more human: emotional security.

Emotional security is what clients feel when they know they're protected, not just financially but personally. It's the trust that's built when a financial professional helps them face tough realities—and guides them through.

In today's commoditized landscape, emotional security has become the key to unlocking client loyalty, intergenerational planning continuity, and resilience in advisor-client relationships. It's what drives referrals, repeat business, and retention during volatile markets.

Yet most financial professionals aren't trained to offer it.

This is where insurance executives have an urgent and valuable role to play.

When we talk about innovation, we often default to digital tools, AI, or frictionless platforms. These are essential, of course. But we can't innovate our way around human fear, aging parents, or adult children caught off guard by caregiving.

We need to rethink our product development priorities.

Innovation must also mean designing products that make it easier for professionals to have difficult conversations, and easier for clients to say "yes" to planning. That's not just about simplicity. It's about psychological accessibility. It's about creating solutions that align with how people actually think, feel, and make decisions.

This kind of emotionally intelligent product design bridges the gap between protection and peace of mind. It's not just solving a financial problem—it's solving a behavioral one.

If emotional security is the goal, then LTC planning is its crucible. It's where we as an industry prove whether we're willing to lead people through life's most difficult transitions or let them face the issues alone.

This isn't just a distribution problem. It's a leadership opportunity.

• Are we equipping financial professionals with the tools and training to handle emotional resistance?
• Are we creating incentives that reward meaningful planning over quick wins?
• Are our products and messaging designed with emotional behavior in mind—or just actuarial logic?

LTC Talk Avoidance Syndrome doesn't just cost consumers. It costs us trust. It limits our growth. And it undermines the promise that our industry makes: to help people live with security, dignity, and confidence—no matter what life throws at them.

For those leading the insurance and annuity space, this is a moment to ask: How do we define innovation?

Yes, it's about technology. Yes, it's about efficiency. But it's also about empathy.

The future belongs to firms that recognize emotional fluency as a strategic asset, and emotional security as a deliverable, not just a byproduct.

This means:

• Creating solutions that address behavioral obstacles, not just financial gaps.
• Supporting financial professionals in building trust through emotionally intelligent planning.
• Embracing products that offer clients flexibility, security, and peace of mind in one package.

The LTC crisis is growing. The need for solutions is clear. What's missing isn't capability—it's courage.

Let's lead with both.

Nearly every article addressing medical inflation and its effect on property and casualty (P&C) insurance claims begins with a reference to the broader trend of lower overall inflation. While the cooling effect of inflation has notably benefited the prices of goods and commodities, healthcare services remain significantly affected. According to Peterson-KFF, in June 2024, the Consumer Price Index (CPI) for all urban consumers rose by 3.0% compared with the previous year, while medical care costs increased by 3.3%. When excluding healthcare services, the overall CPI growth was limited to 2.9%. This marked June 2024 as the first month since early 2021 where medical care prices had risen faster than the general inflation rate.

Additionally, over the span of the past two decades, the price of medical care and its subcomponents has escalated by 121%, compared with an 86% increase in the prices of all consumer goods and services. This disparity amounts to 35 percentage points. Notably, healthcare expenditures accounted for 18% of U.S. GDP in 2023, underscoring the growing significance of healthcare costs within the broader economic landscape. As such, the acceleration of healthcare inflation presents significant implications for workers' compensation and casualty bodily injury claims.

Here is a view of national health expenditures by year through 2023 as prepared by Peterson-KFF:

Factors Contributing to Medical Inflation

Several key factors contribute to the persistent rise in medical costs, including:

1. Aging Healthcare Workforce: The Association of American Medical Colleges projects that within the next decade, 40% of the U.S. physician workforce will be aged 65 or older. Simultaneously, a shortage of up to 3.2 million healthcare workers is expected by 2026. The reduced availability of healthcare professionals drives up wages for medical staff, in accordance with the principles of supply and demand. These increased provider costs inevitably flow through to patients, including those involved in P&C claims.

2. Increased Costs for Medical Equipment: The prices of medical equipment, parts, repairs and services are on the rise. Hospitals and medical facilities that rely on critical equipment face growing expenses. As with other cost increases, these higher expenses are ultimately passed down the line to payers, including insurance carriers.

3. Hospital Care Price Fluctuations: According to a report published by the National Council on Compensation Insurance (NCCI) in April 2025, although medical prices softened meaningfully in the first quarter on the combination of several trends, with physician care inflation price changes smaller than in 2024, over time, higher supply costs for equipment and supplies may also lead to higher prices in physician services, facilities, and long-term care. Hospital outpatient care prices saw a moderate growth of about 4% in the last quarter of 2024, followed by an even more moderate 3% increase in the first quarter of 2025. Inpatient care prices rose by approximately 3% in 2024.

4. Third-Party Bodily Injury Claims and Billing Irregularities: Third-party bodily injury claims are often reported late, with the demand package sometimes serving as the initial notice of a claim. These claims are frequently submitted using non-standardized forms, such as UB-04 or UB-92, which may lack necessary billing codes. This raises concerns about whether the charges submitted are inflated or exceed what is considered usual and customary for services rendered in a specific geographic area. In some cases, the answer to this question is affirmative.

5. Increase in Medical Providers and Services per Visit: According to the Workers’ Compensation Research Institute, the number of workers' compensation claims involving multiple healthcare providers has increased by 19% over the past five years. Additionally, the average number of services provided per medical visit has risen by 13% since 2017. The increased number of providers per claim, coupled with more services rendered per visit, results in higher costs per claim.

Implications for Workers' Compensation and Casualty Claims

The rise in medical inflation presents significant challenges for P&C carriers. To manage these challenges effectively, insurance providers must leverage insights from claims data. By using data-driven strategies, claims organizations can identify the most effective medical providers for each case, reduce medical costs, and improve claim outcomes for injured parties.

The adoption of advanced analytics tools will enable P&C carriers to navigate the complexities of rising healthcare costs. As healthcare inflation continues to outpace general inflation, carriers that use data-driven solutions will be better equipped to manage the financial pressures associated with rising medical expenses while ensuring that quality care is maintained for claimants.

By taking a data-informed approach, P&C insurance carriers can better position themselves to mitigate the financial impact of medical inflation, improve operational efficiency, and enhance overall claim outcomes.

As first published in WorkCompWire.

Renovation remains a defining trend in the U.S. housing market. In fact, nearly half of homeowners (48%) plan to renovate in 2025. Median budgets are climbing to around $24,000, while high-end projects often top $150,000. For high-net-worth homeowners, those numbers multiply — expansions, specialty rooms and luxury finishes are increasingly common.

Yet amid design plans and contractor negotiations, insurance is often overlooked. Homeowners should always notify their insurer before any renovation project begins. Failing to do so can result in higher deductibles, denied claims or policies that no longer fit the new risk profile. For affluent households, the stakes are especially high: The wrong coverage approach could mean hundreds of thousands in uncovered loss.

1. Unreported Exposure

Risk: Projects that cost more than 10% of a home's insured value, extend beyond a year or require moving out temporarily alter the home's risk. If insurers aren't informed, claims could be contested.

Best Practice: Notify your broker early. A simple litmus test: If you're moving out, disabling security systems or investing more than 10% of insured value, call your advisor. This allows for adjustments before the risk materializes.

2. Policy Reclassification and Deductible Shifts

Risk: Large-scale renovations can require a shift from a standard homeowner's policy to a builder's risk or course of construction policy. If overlooked, deductible surprises can surface. Some carriers apply a construction-related deductible many times larger than a typical homeowner's deductible.

Best Practice: Confirm with the insurer whether builder's risk coverage is required. These policies are designed for homes "in transition." Establishing them early prevents costly disputes if a fire, water loss or theft occurs mid-project.

3. Contractor and Subcontractor Liability

Risk: Renovations introduce third-party exposures. Hiring a contractor with inadequate general liability (GL) or workers' compensation (WC) coverage creates liability exposure. If a subcontractor is injured or damages property without proper insurance, a carrier may be left without recourse.

Best Practice: Require certificates of insurance from all contractors and subcontractors. For high-value properties, ensure GL limits are consistent with the replacement value of the home. Carriers frequently request this documentation and can help validate that coverage is adequate.

4. Underinsurance During and After Renovation

Risk: Renovations increase replacement costs. Without a coverage adjustment, reimbursement may only be for the pre-renovation value. Replacement costs surged more than 55% between 2020 and 2022, driven by inflation and supply chain challenges. If the homeowner's coverage hasn't kept pace, a catastrophic loss could leave the homeowner significantly underinsured.

Best Practice: Request periodic revaluation during and after construction. Policies with extended replacement-cost features or inflation guards can help, but they aren't substitutes for accurate dwelling limits. Insuring your home to value is critical after a renovation project.

5. Vacancy, Theft, and Fire Hazards

Risk: Many renovations involve temporary vacancy or disabled security systems, which dramatically change exposure. Standard homeowner's insurance often excludes theft or vandalism after 30 or 60 days of vacancy. Fire hazards from activities like sanding floors or rewiring electrical systems elevate risk.

Best Practice: Inform the carrier if living elsewhere during a project. Confirm that belongings in storage remain covered and that valuables such as artworks, if moved off premises, are stored in approved environments. Ask whether endorsements for theft of building materials, or a course of construction policy should be added while work is underway.

Closing Perspective

The numbers are clear: 98% of homeowner's insurance claims involve property damage, with average claim severity approaching $24,000 in higher-risk areas. For wealthy homeowners undertaking renovations, those costs can climb into six figures, so you need to make sure the proper coverages are in place prior to starting your project.

Renovation is a fundamental change to a home's risk profile. Treat it accordingly. By contacting the broker early, validating contractor coverage, adjusting limits during construction, and re-evaluating after completion, the insured is protecting both their property and their investment.