International insurance broking operates across multi-actor systems without a structured method for reading the connectivity between them. Complexity becomes concrete when renewals stall, when claims escalate without warning, when regulation forces last-minute adjustments. Pressure concentrates in certain places, travels along some pathways, and dissipates in others. 

The geometry of these movements is what I call adjacency: the measure of how tightly actors are bound to one another, and how their ties carry or absorb pressure. The concept draws on network theory's insight that structure shapes behavior, and on systems thinking's recognition that interdependence produces non-linear effects. What adjacency mapping adds is an operational instrument calibrated to the specific architecture of international insurance programs, one that translates structural insight into practitioner decisions.

An international program is not a set of bilateral relationships. It is a system in which master clients, local clients, brokers, and insurers connect continuously, and in which a shift in one part alters conditions across the rest. A disputed claim at the local level can reverberate upward until it unsettles the master layer. A regulatory delay in one jurisdiction will delay the entire renewal cycle. When negotiations falter between a master broker and a local insurer, expectations unsettle across several markets simultaneously. The system propagates pressure because its ties differ in weight, consequence, and resilience.

The structure begins with the system's elements. Six actors form the state vector of any program:

Here, Smc denotes the master client, Slc the local clients, Smb the master broker, Slb the local brokers, Smi the master insurer, and Sli the local insurers. The notation names the nodes that matter. The model captures structural connectivity. It measures the presence, intensity, and resilience of operational ties, not the informal influence, cultural distance, or reputational history that also shape relationships. Understanding how the system functions requires capturing how strongly these actors are tied to one another.

The adjacency matrix A fulfils this function. It represents the interaction weights between stakeholders: each element wij indicates the presence and intensity of the relationship between stakeholder i and stakeholder j. The matrix is first constructed in abstract form, mapping the position of each interaction within the system:

The abstract form locates each relationship within the system. The subscripts identify the two stakeholders involved; the element wij denotes the weight of their tie. The purpose of this construction is to formalize the network so that the system can be analyzed as a structure rather than through accumulated observation. Once defined, weights are assigned on a 0 to 1 scale. On this scale, 0 denotes the absence of adjacency; 0.3 indicates a weak tie with limited interactivity; 0.6 represents strong adjacency with effective coordination; and 1 signals optimal alignment. High adjacency is a marker of capability: two stakeholders are tightly coupled, mutually responsive, and able to sustain efficient workflows. Low adjacency signals fragmentation and the structural risk of disconnection. The weights are practitioner judgements. Their value lies in making an assessment explicit that experience tends to leave implicit. A broker who has managed the same program for a decade carries a mental map of its connectivity. The adjacency matrix makes that map visible, comparable, and open to revision.

Construction begins with a structured assessment across all active relationships in the program. The broker assigns an initial weight to each tie by asking three questions: how often do these actors interact operationally, how reliably does information move between them, and how quickly does the tie transmit pressure when the program is under strain. These criteria are observable without measurement instruments. They are the qualities experienced brokers already assess informally. The matrix makes that assessment formal, consistent, and transferable across programs and teams.

A populated matrix takes the following form:

The matrix is a map of the system's connective capacity. A weight of 0.6 between master and local clients reflects strong alignment: headquarters and subsidiaries adjust to one another with speed. A 0.3 between master clients and master brokers indicates a weaker tie, where coordination exists but is less intensive and more susceptible to friction. A 0.2 between master clients and master insurers signals low adjacency: limited interactivity risks disconnection unless brokers actively mediate. A 0.6 between master brokers and local insurers, by contrast, marks a high-value link, one where workflow is active and system coordination is at its strongest. High adjacency marks the ties through which decisions travel, alignment is secured, and operations proceed without friction. Low adjacency marks the fracture lines where interactivity is minimal, silos form, and misalignment compounds.

Adjacency mapping derives its analytical value from the fact that connectivity is never static. Strong ties allow programs to move with speed and coherence. When master and local brokers hold a 0.6 adjacency, coordination is tight and workflow advances without resistance. When a claim escalates across a 0.6 link between local and master insurers, the system responds rapidly. Weak ties do the opposite: they isolate segments of the program, delay decisions, and erode effectiveness.

The architect's objective is to sustain ties at 0.6, the threshold at which alignment holds, coordination costs nothing, and the program moves with structural coherence.

Three patterns govern how pressure moves through the system. Concentration forms where multiple strong ties converge, typically around master brokers holding 0.6+ adjacencies with both local brokers and master insurers. These nodes become coordination hubs, capable of synchronizing decisions across jurisdictional boundaries. Propagation measures the efficiency with which decisions travel. The difference between a 0.6 and a 0.3 tie is the difference between transmission and friction. A 0.6 link between master and local insurers ensures a claim escalates without delay; a 0.3 tie ensures it stalls, and the broker must compensate manually for what the tie fails to carry. Absorption occurs at weak adjacencies of 0.3 or below, where pressure dissipates rather than transmits. Occasionally this buffers noise; more often it marks a structural disconnection that prevents system-wide coordination. These patterns do not operate independently. A weak tie between master broker and local insurer becomes more consequential when the master client to master broker tie is also degraded. Compound weakness across adjacent nodes accelerates fragmentation in ways that no single tie, read in isolation, would predict.

Because ties shift, the map must be kept current. A static diagram decays. A weak link can be reinforced into a strong adjacency by deliberate effort; a strong tie will weaken if neglected. Four events should prompt a reassessment. First, personnel change at any node, meaning the tie shifts with the person. Second, a regulatory change in any jurisdiction covered by the program. Third, a claims event that escalated beyond its expected path. Fourth, the approach of renewal, which is always a structural stress test. Each signals that the weight of at least one tie may have moved without the broker noticing. Adjacency maps are instruments that require periodic review and active maintenance. Brokers who update them see the system. Those who rely on experience alone see only what the system once was.

During renewals, adjacency maps identify which ties sustain workflow and which must be reinforced before they become bottlenecks. In claims, they reveal which relationships enable rapid escalation and which will stall it. Consider a master broker to local insurer tie that registers 0.6 in stable conditions but drops to 0.3 during renewal following personnel turnover at the local level. The map makes this degradation visible in advance. The broker can then rebuild the tie through intensified communication, workflow realignment, or deliberate relationship investment before claims season converts a weak link into a coordination failure. The same logic applies during a major claims event. A local insurer holding a 0.6 adjacency with the master insurer will escalate rapidly and with precision. One holding a 0.3 will delay, misframe, or absorb the claim at the local level, forcing the master broker to intervene manually at precisely the moment when speed matters most. The map identifies this vulnerability before the claim arrives. In regulatory matters, the map shows where connectivity must be strengthened to secure compliance. In each case, the broker acts before disruption, reinforcing the ties the system depends on rather than repairing them under pressure.

The broker who monitors adjacency, reassesses ties under pressure, and rebuilds degraded links before they become failures is sustaining program coherence. That is what rigorous servicing looks like in practice.

The central proposition of adjacency mapping is that program performance correlates with the aggregate strength of ties between its actors. The broker whose counterpart is responsive, informed, and quick to act is not simply lucky in his relationships. He is operating across a tie with high adjacency. When that tie degrades, the program follows, regardless of how well the individuals involved know each other. This is a testable claim. Brokers who map their programs over time will find that degradation in tie strength precedes operational failure, and that deliberate investment in adjacency produces measurable improvements in renewal speed, claims resolution, and regulatory compliance. Together they provide foresight into where the system is strong, where it is fragile, and where investment in interactivity will deliver the greatest return. The program, read this way, becomes a structure with legible geometry.

International insurance broking will always be exposed to uncertainty. Renewals will clash with shifting regulation, claims will appear at awkward times, and timelines will compress under pressure. But complexity is not chaos. By treating programs as systems and adjacency maps as diagnostic instruments, brokers can anticipate rather than endure, and reinforce rather than repair. Pressure still moves through the system. Adjacency maps tell you in advance where it will concentrate, where it will stall, and where it will dissipate unnoticed. In a system this complex, that is the only form of control that holds.

What Part(s) of the Insurance Industry Can Benefit From Our Product: 

(You may select any or all of these bullet points.  Just let us know which ones to remove.  IF you select Other, please provide us with that Category.  Maximum of 1 Other Category can be added.) 

• Claims
• Customer Experience
• Customer Service
• Distribution
• Operational Efficiency
• Other: 

Three Main Benefits of the Product: 

• Performance:  Higson executes rules with an average latency of 0.23 ms and supports up to 9,000 requests per second in a standard configuration. Thanks to the built-in ONNX runtime, trained ML models can run directly inside decision tables without requiring a separate inference layer.
• Scalability:  Our CPU-based licensing model means you pay for actual production usage rather than user counts. A proof of concept can run on AWS at approximately $0.63 per hour, with infrastructure costs scaling predictably alongside transaction volume.
• Business-Owned Decision Logic:  Pricing analysts, underwriters, and compliance teams can create and deploy rule changes directly using decision tables, Groovy and Python functions, as well as visual flows without relying on IT support or opening development tickets. The platform includes full version control, auditability, and simulation capabilities before deployment to production.

Why We Are the Right Solution For Your Needs

Most P&C insurers manage pricing, underwriting, claims, and compliance rules across separate systems - each governed by different teams and operating on different release cycles. When regulatory changes occur, organisations often end up running multiple parallel workstreams while hoping all outputs remain aligned.

Higson consolidates decision logic within a single rules engine, giving business users direct control over processes that are traditionally dependent on IT delivery queues.

Pricing analysts, underwriters, and compliance teams can author and deploy changes directly using decision tables, visual flows, and embedded scripting capabilities with full version control and auditability built in.

From a technical perspective, Higson executes rules with an average latency of 0.23 ms and supports up to 9,000 requests per second. A proof of concept can run on AWS at approximately $0.63 per hour, while CPU-based licensing ensures infrastructure costs scale with actual usage rather than user counts.

 Watch Our Demo: 

Learn More

Are you interested in learning more about Higson-Business Rules Engine powered by Decerto and about this product?  Please visit these sites to learn more or to schedule a personal demo:

• www.higson.io
• https://pl.linkedin.com/company/decerto
• https://calendly.com/lukas-niedospial-1/60min

As long as everyone has been telling their Ted Turner stories in the wake of his recent death, I thought I'd tell mine, before getting on to this week's business: what I see as a growing backlash among younger generations toward AI that business leaders need to contend with.

My story comes from my friend Marc (a former managing director at Marsh McLennan, as it happens). He was at the helm in a multi-day sailboat race around Long Island in the 1980s and timed the start almost perfectly. In the chaotic way that these races start, you don't know when the horn will blow, so you circle as you try to be at full speed with a clear path to the starting line when the horn sounds. Marc had succeeded — but Ted Turner was bearing down on him, aiming for the same spot on the line that Marc was going to cross. 

Marc had the right of way, but this was Ted Turner, recent winner of the America's Cup, in a much bigger, faster boat, with a world class, steely glare as he steered his boat on a collision course with Marc. 

Marc never wavered, and at the last possible moment Turner bore off, did a 360, and crossed the starting line a minute or so later. Turner won the class among the biggest boats, while Marc and his crew just did well in his class of smaller ones. Finishing late at night, he and his crewmates headed to a bar to decompress. At 1am, they were getting ready to call it a night, when the bartender set a round of drinks in front of them and said they were sent with the compliments of the gentleman at the door. The gentleman was Ted Turner. He nodded respectfully in their direction. Then he gave them the finger with both hands and stormed out.

The bartender told Marc that Turner said he'd been scouring every bar on the waterfront in search of Marc and his friends. Whatever else you want to say about Turner, the man had style.

Now on to the backlash against AI that we all should be watching. 

I use my daughters, aged 32 and 29, as my antennae about attitudes among Millennials and Gen Z, and they started bristling about AI months ago. Initially, they complained about the huge amounts of water required for cooling. If I ever mentioned using an AI for something, one of them might make a snide remark — they're given to snide remarks with their father — like, “I guess the real prompt is: ‘Hey ChatGPT, could you please drain another reservoir for me?’”

Hyperscalers' wild need for electricity for their gen AI data centers led to concerns about what AI was doing to the environment. That my daughters' electric bills were climbing didn't help matters.

More recently, they've resonated with the concerns of those facing the prospect of having data centers built near them, each spanning perhaps tens of thousands of acres. To top it all off, my older daughter lost her writing job to an AI, as I mentioned last week. The girls have told me to turn off the AI summary that Google Search now offers.

A recent New York Times article reports on a Gallup survey that found Gen Z's attitude toward AI souring, and for reasons that go well beyond the sorts of environmental concerns that initially triggered my daughters. 

"Many respondents did acknowledge that A.I. might make them more efficient in school and the workplace," the article said. "But they were concerned about how the technology would affect their creativity and critical thinking skills.

"Young adults in the work force were especially skeptical. Close to half of those surveyed said the risks of artificial intelligence outweighed its potential benefits in the workplace, an 11-point jump from the previous year. Only 15 percent said they saw A.I. as a net benefit."

The Times also reported on a viral video (that my daughters had already made sure I saw) of a woman giving a commencement speech in which she declared that "the rise of artificial intelligence is the next Industrial Revolution" — only to be roundly booed by the students. 

“'What happened?' [she] stammered, looking over her shoulder, as if searching for an escape hatch," the Times reported.

She continued: 

"'Only a few years ago, A.I. was not a factor in our lives.

"The crowd erupted in cheers.

“'And now, A.I. capabilities are in the palm of our hands.' Boooooooooo.

"One might call it a 'read the room' moment."

Eric Schmidt, former CEO of Google, got booed even harder when talking about AI in his commencement address at the University of Arizona on Friday.

I'm not saying dissatisfaction among younger generations will stop the adoption of generative AI, any more than concerns by earlier generations could stop the internet or the smartphone. I'm also not saying Millennials and Gen Z are Luddites; they're extremely sophisticated about technology. 

What I'm saying is that younger generations seem to be taking a warier approach than those of us of a certain age, who've not only been through a few technology revolutions and have accepted their inevitability but whose views are perhaps softened by what all the AI investments are doing for our retirement accounts. 

And those younger generations get a vote. The discussions among business leaders may be about use cases for AI, about how to implement AI most effectively, about how to demonstrate ROI to shareholders, and so on, but your employees are going to be doing that implementing. If a big chunk of your work force dislikes or distrusts AI, they can provide a lot of silent resistance that may surprise you if you haven't made the effort to understand their concerns and to work with your employees to address them.

Cheers,

Paul

P.S. After writing this commentary last night, I wake up today to find that I'm not the only one thinking about the AI backlash. A New York Times columnist wrote: Why College Grads Are Booing Their Commencement Speakers. The Wall Street Journal led its website with: The American Rebellion Against AI Is Gaining Steam. Their reporting/reasoning differs a bit from mine, but my conclusion remains the same: Proceed with caution. 

P.P.S. It is with great sadness that I note the passing of Stephen Applebaum at age 81. Stephen was one of the earliest and dearest friends of ITL and was generous not just with me but with everyone he met in his decades of work in the insurance industry. I looked back through the 80-some articles Stephen wrote or co-wrote for us over the years to see if I might single out a few, but there are just too many sharp insights. I will point to one, which he wrote a year ago with his business partner, Alan Demers, because it's not only very smart but because Stephen always struck me as an empathetic man: "Re(Defining Empathy in Insurance." 

Here is a link to a brief obituary, to the funeral arrangements and to a way to donate to the Dragonfly Foundation, a favorite of Stephen's that focuses on pediatric cancer care.

May his memory be a blessing.

What Part(s) of the Insurance Industry Can Benefit From Our Product: 

(You may select any or all of these bullet points.  Just let us know which ones to remove.  IF you select Other, please provide us with that Category.  Maximum of 1 Other Category can be added.) 

• Claims
• Customer Experience
• Customer Service
• Distribution
• Operational Efficiency
• Other: 

Three Main Benefits of the Product: 

• Point 1 (Max 160 characters, including spaces and punctuation)
• Point 2 (Max 160 characters, including spaces and punctuation)
• Point 3 (Max 160 characters, including spaces and punctuation)

Why We Are the Right Solution For Your Needs

General Description or Sales Pitch.  We recommend limiting this to 1 to 2 paragraphs, so the reader can quickly get to watching your Demo Video.  

 Watch Our Demo: 

Demo VIdeo Placed Here.  Video must be hosted by ITL, so please provide us with a method to download the video.  

Learn More

Are you interested in learning more about <Company> and about this product?  Please visit these sites to learn more: 

(You can provide us with any or all the below links)

• Website Link
• Linkedin Link
• X Link
• Facebook Link

Finding a route through extreme uncertainty from climate tipping points is now urgent for organizations. A pragmatic mindset and proven techniques can uncover your path.

Climate systems are moving toward abrupt, irreversible shifts. These tipping points include the collapse of the Atlantic Meridional Overturning Circulation (AMOC), the system of ocean currents in the Atlantic that plays a crucial role in regulating Earth's climate by transporting heat from the tropics northward.

The AMOC is likely weakening. Should it reach a tipping point, the impacts could challenge the historical wisdom that climate change unfolds gradually. Regional conditions could now flip quickly, bringing severe cooling to northern Europe, forcing storm tracks into new positions, shifting monsoons and altering coastlines.

The most recent Nordic Tipping Week saw researchers and policymakers treat AMOC tipping as a realistic planning case. When science moves from questioning if a tipping point might happen to focusing on when, an organization's planning expectations need to change.

It is critical that organizations and their stakeholders lift themselves out of the climate catastrophizing that tipping points may prompt. That's because doom can shut down action.

Instead, it is important to view tipping points through a constructive lens that focuses on "earthshots," not "moonshots." We're not talking here of hope over expectation but deploying disciplined techniques in which risk professionals are already well-versed.

Tipping‑point‑aware tools and techniques, such as enhanced scenario planning and strategic planning with advanced risk identification, can help you replace passive dread with active preparation. Incorporating tipping points into an organization's risk management and strategic planning will also help maintain credibility with the regulators, investors and insurers we can expect to ask tougher questions on business's readiness for extreme disorder.

The impact of AMOC tipping points

Should the AMOC cross its tipping point, we could see rapid and irreversible climate shifts, including:

• Severe winter cooling across northern Europe
• More intense storms and altered storm track behavior
• Long-term agricultural disruption
• Major changes in water and food availability.

Given measured weakening and converging scientific warnings, the most prudent approach for businesses, particularly those with UK or European exposure, is to incorporate the possibility of AMOC weakening into their risk register and scenario testing. The time has come to view AMOC tipping as a credible tail risk to which businesses need to prepare.

When a system approaches a tipping point, a business should prize preparedness over precision. An organization's role here isn't about beating scientists to pinpoint the exact moment of a shift but strengthening its ability to remain stable when uncertainty accelerates.

Organizations should start by identifying those areas where they are most dependent on climate stability; think agricultural inputs, logistics routes or water availability. How would abrupt cooling, extreme storms or rainfall shifts put pressure on the most climate-reliant nodes of their operations and supply chains?

Enhanced scenario planning for effective adaptation

Organizations can no longer assume risk mitigation will keep climate risk within more familiar limits. Ice melt, freshwater dilution in the North Atlantic, and shifts in rainfall belts are already building momentum, with some impacts locked in for decades. That means businesses should consider tipping-point-aware adaptation as part of their strategic decision‑making, calling on scenarios that extend beyond traditional pathways.

Severe-but-plausible scenarios, such as sudden cooling in Europe or major shifts in precipitation zones, will provide a clearer understanding of the future operating environment. Updated scenarios should be able to test the full chain of consequences, rather than the most familiar ones, investigating how physical risks and supplier reliability might change under tipping‑point conditions.

These scenarios may well feel uncomfortable, but they are also far from improbable.

My older daughter just lost a writing job to an AI (that she had to train to replace her), so I don't currently have the kindest thoughts about where AI is headed, but the technology is going to keep barreling forward whether we like it or not, and we all have to adapt.

So let's take a look at the smartest pieces I've seen recently about where generative AI is headed. We'll look at the "fog of AI," which is making it so very hard to make investment decisions. We'll look at the insurance industry's quandary about how to handle all the data centers being built (maybe). We'll look at lessons learned from early attempts at scaling AI, to see what separates the winners from the losers. 

But let's start with a piece that contradicts the conventional wisdom that AI agents should be treated as employees.

An article in Harvard Business Review says: 

"Leaders assume that anthropomorphizing AI will make the technology feel less foreign to workers or that it will signal the company’s AI ambitions to investors, customers, or internal stakeholders. But it turns out that treating AI as an employee is not so straightforward. 

"In a randomized experiment, we found that humanizing AI can shift accountability away from individuals, increase escalation, reduce review quality, and erode professional identity and trust. What’s more, it doesn’t meaningfully increase people’s intent to adopt the technology and integrate it into workflows—which remain the key obstacle to capturing AI’s enormous value creation promise."

The most striking findings to me were that AIs treated as an employee, rather than a tool, were more likely to lead to humans sloughing off responsibility for any problems that occurred and to more often asking their managers for additional review. The article doesn't argue for slowing down implementation of AI, by any means, but does make a case for changing how many of us think about describing their role.

Another HBR article, titled "The Future Is Shrouded in an AI Fog," offers some comfort for those of us confused about how to proceed with implementing AI. The piece says we pretty much have to be paralyzed by indecision because of the "extreme opacity" about the future of AI:

"Given all the things that might change because of AI, it feels like a fog has descended that occludes our ability to see the future. And right now, that’s its most important—and perhaps most underappreciated—economic effect.... This extreme uncertainty challenges the criteria we use to commit to forward-looking investments."

The opacity doesn't just affect businesses, either. It also hits us as individuals. The article asks, for instance, why smart kids would want to spend a decade training to be a doctor when it's not clear what being a doctor will mean in the age of AI.

Again, self-pity isn't allowed, at least not for very long. The article lays out an approach designed to help us sense change sooner and react with more agility, then tells us to get on it.

Mick Moloney of Oliver Wyman articulates a question I've heard lots of insurance executives pondering lately: How should insurers handle the hundreds of billions of dollars of data centers being built to accommodate the AI rush?

As Mick puts it:

"The six largest AI data center projects currently under construction or formally committed in the United States represent a combined investment of over $120 billion and a combined power capacity target of more than 10 gigawatts — deployed not over decades, as comparable infrastructure has always been, but over three to five years. They are being built by technology companies, AI laboratories, and private equity platforms that have never operated infrastructure at this scale. And they are being financed with instruments that did not exist eighteen months ago."

He doesn't have a silver bullet, but he does offer keen insights into how insurers should think about these six projects based on their power strategy, their financing structures and the risk management capabilities (or, more likely, the lack thereof) of the builders.

The insurance industry will be wrestling with the data center issue for years, but Mick's piece is a good start.

Finally, McKinsey published "The AI Transformation Manifesto," with a dozen observations about what separates the winners from the losers in the age of AI. For instance:

• Technology alone doesn’t create advantage; enduring capabilities do. Who are the early winners at AI? The same companies that have been winning before by building capabilities that allow them to harness any technology effectively.... When these new capabilities are built—and they take time to build—the company accelerates its business transformation with technology and outperforms its peers. The capabilities become the competitive advantage....
• Economic leverage points are your best focal points. Any business model has a few key economic leverage points that provide the biggest impact when improved with AI. In mining, for example, process yield and throughput is a key economic leverage point, and that’s where Freeport-McMoRan achieved game-changing impact. In automotive, supply chain integration is a key leverage point, and that’s where Toyota had its AI breakthrough. Most companies have long lists of use cases. Successful ones focus on achieving deep business transformation in the few areas that matter strategically. That’s where they double down to build AI systems....
• Building the tech and AI muscle of your senior business leaders should be a top priority. We don’t have a single success story where senior business leaders were not in the driver’s seat. IT leaders can support the transformation, of course, but it’s business leaders who need to drive it.

Again, I don't see a silver bullet, but we're learning....

Cheers,

Paul

Budgets have grown, pilots have multiplied, and AI is now a fixture in virtually every P&C strategic plan. Yet 42% of insurers track no AI metrics at all, which means they have no way to validate what works, no playbook to scale it, and no mechanism to stop what doesn't work. Insurers' investment pattern confirms that this is an organizational constraint, rather than a technology one: on average, 72% of AI spending goes to technology and only 28% to change management.

Technology creates capability. But change management determines whether that capability becomes performance. That imbalance is the first signal of what Capgemini identifies in the 19th edition of its 2026 World Property and Casualty Insurance Report as an "architecture mismatch." This is a structural gap that runs deeper than the technology stack, and that no amount of additional AI investment will close on its own.

Three dimensions, one ceiling

The first dimension is a strategy and talent gap. Among the top 20 global P&C insurers, only 35% have explicitly linked their AI strategy to business outcomes beyond efficiency. That narrow framing has consequences: Strategy tends to direct investment toward quick wins rather than the capabilities AI needs to grow over time. In most cases, the result is an incomplete strategy that optimizes the present while leaving the future underbuilt.

The second dimension is technical constraints. Legacy architectures fragment data across functions, making it harder for AI to reason across underwriting judgments, claims assessments, and distribution decisions that depend on context-rich, unstructured information. The barrier is less about the AI itself and more about the environment it must operate in – one that was not designed with AI in mind and does not easily accommodate it.

The third – and arguably most decisive – dimension is organizational. Over half (55%) of insurers cite unclear ownership of AI initiatives as a key constraint. Without clear accountability, programs stay dependent on individual champions rather than building institutional capability. And despite all the work underway, 47% of employees report no meaningful change in their day-to-day work after 18 months of using AI. That points less to a deployment failure than a design flaw.

The problem with fixing one thing at a time

These three dimensions are entangled, which is precisely what makes the conventional response insufficient. Assess, prioritize, sequence: Fix strategy first, then technology, then organization. In practice, addressing one while leaving the others untouched tends to limit progress, rather than unlock it.

Our research identifies the emergence of intelligence trailblazers – the top 10% of P&C insurers – who treat AI as a core operating capability rather than a program to be managed, aligning strategy, technology, and organizational adoption in tandem. Over three years, trailblazers have achieved 21% higher revenue growth and 51% greater share price increases compared with the rest of the industry.

Despite their growth, this group has also not fully solved the problem. AI still largely operates at the task level, workflows remain built for human execution, and the organizational model that closes those gaps – one where human expertise and synthetic execution are deliberately organized around where each creates the most value – is still being built. The opportunity to redesign is real. But it remains an opportunity, not yet an achievement, even for those furthest ahead.

The harder conversation

An uncomfortable question to raise is why this is so difficult to change, even for organizations that understand the problem.

The answer is that the architecture mismatch was not built through bad decisions. Legacy systems were the right investment at the time. Prioritizing technology over change management made sense when AI was unproven, and the organizational implications were unclear. It is not evidence of poor judgment, but the accumulated consequence of individually rational choices made in a different context.

Moving forward requires asking a more challenging question: Do the investments already made, and the ones being considered now, still pay back on the original terms? Most organizations haven't asked that question systematically, because who defines success, who is accountable for outcomes, and how progress is measured beyond deployment were all designed for a time when decisions were quintessentially human. And until that question gets asked, the architecture underneath the pilots stays unchanged – regardless of how many new tools are deployed on top of it.

Trailblazers are not ahead because they have solved the problem or because they run better pilots. They are ahead because they made a different decision earlier: to address the architecture underneath the pilots, not just the pilots in isolation. The next decision is harder: to redesign the organization itself. That decision has not yet been fully made by anyone. But the insurers who make it first will define what competitive advantage looks like in the intelligence era.

For the insurance industry, where decisions have significant consequences, general-purpose safety controls aren't enough to ensure the safe deployment of large language models. Insurance-specific guardrails, which control all aspects of the interaction of artificial intelligence, from input validation to output verification, are a necessity. 

1. The Opportunity: AI Is Reshaping Insurance

AI is already transforming core insurance operations across the value chain. According to ACORD research, 77% of insurers now use AI somewhere in their operations, and early implementations have demonstrated claims processing time reductions of as much as 75% — compressing multi-day workflows into under an hour.¹ The global AI in insurance market, valued at $4.6 billion in 2022, is projected to reach $79.9 billion by 2032.

Core applications already in production include:

• Claims automation and straight-through processing
• Computer vision for property and vehicle damage assessment
• NLP-based document parsing and policy review
• Fraud detection and anomaly identification
• Customer-facing chatbots and virtual agents
• Underwriting analytics and risk scoring

These applications can enhance customer satisfaction, resolving claims faster, and even help employees deal with the sheer volume of policy documents. But the very attributes that make LLMs so appealing to businesses — fluency, speed, and language breadth — also pose the biggest risk to using them in regulated environments like insurance.

2. The Core Problem: Hallucinations in a Regulated Domain

LLM hallucination occurs when a model generates content that is factually incorrect, fabricated, or unsupported by the context provided. In insurance, that could mean:

• Misstating coverage terms or policy limits
• Inventing exclusions or endorsements that do not exist
• Providing inaccurate claims guidance
• Citing non-existent regulations or procedures
• Expressing unwarranted confidence where escalation is required

The scale of this risk is not trivial. Research published in peer-reviewed AI benchmarks has found hallucination rates of 15–30% in general-domain LLMs.² Even in legal AI applications — a domain with similar stakes — clause-review accuracy in the 86–92% range still implies error rates of up to 14% in some contexts.³

For insurance organizations, a single inaccurate coverage explanation or claims instruction can trigger downstream complaints, regulatory disputes, or litigation. Unlike casual consumer applications, insurance AI interacts with financial protection, legal obligations, and sensitive personal information — where errors carry real consequences.

3. Why Generic AI Safety Tools Are Not Enough

Most commercially available AI safety frameworks focus on broad categories such as:

• Toxic content filtering
• Personally identifiable information (PII) detection
• Basic prompt injection defense

These controls are necessary, but they are insufficient for insurance. Standard safety tools do not adequately address insurance-specific factual accuracy, policy compliance, or regulatory conformance. A response can be polite and harmless in tone while still being operationally dangerous if it mischaracterizes a coverage provision or misquotes a policy term.

That is why insurers need domain-specific guardrails rather than generic content filters layered onto general-purpose models.

4. Guardrails as a Business and Compliance Requirement

Guardrails should be understood as a control framework, not a technical add-on. They enforce boundaries across the full AI interaction lifecycle — from what a user inputs to what the system delivers.

Input Guardrails - filter harmful or manipulative requests, detect prompt injection attempts, and prevent users from circumventing policy or compliance constraints.

Dialog Guardrails - manage conversation flow and enforce interaction boundaries, keeping the assistant within approved topics and triggering appropriate escalation pathways.

Retrieval Guardrails - validate external documents and knowledge sources before the model incorporates them into a response, reducing the risk of answers based on outdated or unsupported information.

Execution Guardrails - control external actions and API calls, ensuring that when the AI is connected to claims, policy, or customer systems, operations remain within authorized boundaries.

Output Guardrails - analyze generated responses before delivery, checking for factual grounding, safety, privacy risks, and regulatory alignment.

Together, this architecture transforms AI from a probabilistic text generator into a governed enterprise system — one whose behavior can be monitored, explained, and audited.

5. Why Insurance Requires Specialized Guardrails

Insurance use cases demand a stricter standard because the domain combines four compounding risk factors:

High-Consequence Decisions. Claims settlements, coverage explanations, underwriting support, and fraud workflows directly affect customers' financial rights and legal standing. Errors are not minor UX failures — they are potential compliance events.

Complex Source Material. Policy language, endorsements, exclusions, and jurisdiction-specific requirements are difficult to interpret even for trained professionals. LLMs must be grounded in the actual policy documents, not a generalized approximation.

Regulatory Oversight. The NAIC framework for the "AI Model Bulletin" has five areas of expectations: AI Governance, Transparency, Risk Management, Auditability, and Vendor Oversight.⁴ It is evident from these expectations that insurers need to explain, monitor, and control their AI in production, which is not possible without guardrails.

Sensitive Data Handling. Insurance workflows routinely involve health information, financial records, claim narratives, and other protected personal data. Privacy failures are not just technical issues; they are compliance violations and trust failures with lasting customer impact.

6. A Practical Implementation Approach

Rather than attempting a broad enterprise rollout, insurers should begin with a focused use case that offers high visibility and measurable outcomes. Property and casualty claims processing is a natural starting point: the use case is well-defined, the documents are structured, and accuracy in coverage explanations can be measured against ground-truth policy language.

A phased implementation model should unfold across three stages:

Phase 1 — Foundation (Months 1–3). Establish the guardrail architecture on a single claims workflow. Configure input and output guardrails using the insurer's own policy documents as the knowledge base. Define escalation rules for ambiguous or high-value claims. Instrument logging from day one.

Phase 2 — Validation (Months 4–6). At this phase, human-in-the-loop validation is conducted in conjunction with AI results to verify accuracy, detect hallucination behaviors, and refine retrieval threshold values. Initial bias tests should be performed across various customer types and geography. Compliance and legal should also be involved in validation.

Phase 3 — Expansion (Months 7–12). At this phase, the guardrail methodology is extended to adjacent applications like underwriting support, customer service, and/or document review based on learnings from Phase 1.

The key stakeholders in implementation include claims operations, IT architecture, compliance and legal, data privacy, and a designated AI governance stakeholder responsible for continuing oversight and audit readiness.

7. Ethical AI Must Be Designed In, Not Added Later

One of the most important principles in responsible AI deployment is that ethical safeguards must be built into the architecture from the start — not retrofitted after problems emerge. In insurance, ethics failures can be systemic rather than singular, affecting entire customer segments before they are detected.

The primary ethical considerations for insurance AI are:

Bias Mitigation. Insurers must proactively test AI outputs for differential treatment across customer segments. Research has found that insurance-specific testing can uncover disparate coverage explanations correlated with geography — patterns that generic safety filters are not designed to detect.⁵ Ongoing testing should be built into the governance model, not treated as a one-time validation step.

Transparency. Customers should know when they are interacting with an AI system. The AI should also be able to explain the basis of its response — citing the specific policy document, section, or regulatory reference that underlies its answer.

Human-in-the-Loop Oversight. For complex, ambiguous, or high-stakes interactions — large claim settlements, potential coverage denials, or situations with regulatory implications — the system must escalate to human review. Automation should accelerate decisions, not replace human judgment where judgment is most consequential.

Privacy Protection. PII detection must be robust, particularly in claims workflows involving health information or sensitive personal circumstances. Data minimization practices should be built into the retrieval architecture so that the AI accesses only the information needed to answer the question at hand.

Fairness Auditing. Disparate impact testing across customer segments should be a recurring operational practice, with results informing both model behavior and underlying policy review. Fairness is not a one-time certification — it is a continuing obligation.

8. Conclusion

The case for AI in insurance is compelling. Faster claims resolution, more consistent customer service, and improved operational efficiency are achievable outcomes — and insurers who delay adoption risk falling behind on all three.

But speed without guardrails is not an advantage. LLM deployment introduces real risks of factual inaccuracy, regulatory non-compliance, privacy exposure, and biased decision-making. In a domain where a single miscommunicated coverage term can escalate into a dispute or regulatory inquiry, those risks are not acceptable.

Insurance-specific guardrails are not optional features to be layered on once a system is live. They are the prerequisite that makes responsible deployment possible. Insurers who build control frameworks into the foundation — rather than treating governance as an afterthought — will not only move faster. They will move with the trust, auditability, and regulatory confidence the industry demands.

References

¹ ACORD, "AI in Insurance: State of the Market," 2023; DataGrid, "30 AI in Insurance Statistics," citing ACORD and Risk & Insurance data.

² Ji et al., "Survey of Hallucination in Natural Language Generation," ACM Computing Surveys, 2023.

³ Bommarito & Katz, "GPT Takes the Bar Exam," 2023; see also related empirical work on LLM accuracy in legal clause review, SSRN 2023.

⁴ National Association of Insurance Commissioners, "Model Bulletin on the Use of Artificial Intelligence Systems by Insurers," 2023.

⁵ See emerging literature on algorithmic fairness in P&C insurance, including Casualty Actuarial Society Actuarial Review, 2023–2024.

The insurance industry is one built on history and resilience, but it's also one where the future is facing immense uncertainty.

By the end of 2026, an estimated 400,000 insurance professionals will have retired in the U.S. since the beginning of 2021, according to the Bureau of Labor Statistics. At the same time, nearly one-third of the current global population is Gen Z. And yet, 79% say they've never considered working in insurance due to perceptions of the industry being "boring" or too corporate. Right now, the industry is facing a major workforce shortage that could have the same consequences for our stability as any underwriting cycle or catastrophe trend.

This disconnect is a structural risk to the industry's ability to operate, innovate and respond to crises in the years ahead.

The perception problem is now a workforce problem

For decades, insurance has struggled with an image issue. Despite the passing of the so-called "Great Resignation," the insurance industry continues to face significant workforce challenges.

In an industry survey, Gen Z was asked to identify business sectors that they found the most appealing to work in, and insurance came in last.

Not to mention that this generation is bringing a new meaning to work. One of those is finding a greater purpose in the work itself. However, Gen Z doesn't associate the insurance industry as one that could provide that purpose. Not to mention they want to work in a fun, social environment ... and the perception of the industry is the opposite for most.

It's fair to say that these perceptions are working against the industry.

In reality, insurance is one of the most human industries that exists. It shows up at the most critical moments in people's lives. Most people don't picture it in this way, but insurance is the industry that helps families rebuild after disasters, enables small businesses to survive disruptions and plays a major role in addressing systemic risks like climate change. Yet we continue to present it externally as a series of processes rather than outcomes.

At the same time, there's also an expectation that work environments are dynamic and technology-forward. When those expectations collide with outdated perceptions of insurance, the result is simple. The talent looks elsewhere.

Why talent gaps are becoming operational risks

What makes this moment different is not just the scale of retirements but the nature of the skills leaving the industry.

Insurance has always been a knowledge-driven business. Institutional expertise is the foundation of underwriting decisions, claims handling and client relationships. Still, as experienced professionals exit the workforce, much of that knowledge is at risk of being completely lost or only partially transferred.

The industry is also being asked to evolve faster than ever. Volatility just keeps happening. For example, climate-driven events are increasing in frequency and severity, not to mention cyber risk, supply chain disruption and emerging technologies are introducing new categories of exposure. Meanwhile, customers are expecting faster and more transparent service in real-time.

This is where the talent gap becomes a direct threat to performance.

Without a steady pipeline of new talent, insurers are facing three immediate challenges. First, claims handling capacity becomes strained during surge events, leading to slower response times and diminished customer trust. Second, the adoption of technologies like AI and advanced analytics slows, not because the tools are unavailable, but because the workforce lacks the capacity or skills to implement them effectively. Third, innovation stalls, as fewer cross-disciplinary thinkers enter the industry to challenge legacy approaches.

Because of this, workforce shortages are no longer an HR issue. They become core drivers of operational risk.

Reframing insurance careers for a new generation

If the problem is misalignment between perception and reality, then the solution starts with how we present the industry and how we design the actual employee experience behind that message.

The first shift is reframing the purpose of the work. Insurance organizations need to move beyond describing roles in terms of tasks and instead clearly articulate the impact. Processing a claim is not an administrative function; it's helping someone recover from loss. Underwriting isn't just risk selection; it's enabling economic activity and resilience. If we can't clearly communicate why the work matters, we shouldn't expect younger generations to see its value.

The second shift is making the modern reality of the work visible. Inside many organizations, workflows are already evolving. Automation reduces manual processes. AI is supporting decision-making. Data is becoming central to operations. Yet externally, candidates picture fax machines and cubicles. Bridging this gap requires intentional storytelling and transparency about how the work is actually being done today.

The third is creating clearer and faster paths for growth. Gen Z isn't interested in climbing the ladder and waiting more than 10 years to "become important." They want to understand how they can develop skills and take on new responsibilities within the first one to two years. This requires developing more structured progression frameworks, exposure to different parts of the business and earlier involvement in meaningful decision-making.

Finally, the industry needs to address its reputation directly. There's skepticism from younger generations around complexity, transparency and claims outcomes. Ignoring it will only reinforce distrust. Organizations that acknowledge these concerns and demonstrate how they're improving will be the ones that win over talent, and even customers.

At the end of the day, this all boils down to an alignment problem, not marketing. If your employer brand says "innovative, flexible, purpose-driven," but the actual experience feels slow and transactional, Gen Z will spot that immediately and opt out.

Workforce planning is risk management

All of this leads to a broader point that the industry has not fully embraced. Workforce planning should be treated as a risk management priority.

In insurance, we are disciplined about identifying and managing exposures. We model catastrophe risk. We monitor market volatility. We stress-test portfolios. Yet we've historically approached workforce planning as a functional responsibility rather than a strategic one.

This approach no longer works and cannot continue.

Everything in insurance (i.e., growth targets, service commitments, etc.) ultimately depends on having the right people with the right skills. When that foundation weakens, the impact is immediate. Service levels decline. Innovation slows. Risk exposure increases.

Forward-looking organizations are beginning to recognize this and are integrating workforce considerations into broader risk frameworks. They're mapping critical roles against future business needs. They're identifying where skills gaps are likely to emerge and invest ahead of them. They're rethinking talent models, including how and where work gets done, to ensure resilience at scale.

The bigger picture is ensuring the business can operate reliably under pressure.

The path forward

The insurance industry doesn't have any issues being relevant. What it does have an issue with its perception.

The opportunity in front of us is significant. We have a chance to redefine how a new generation sees this industry and, more importantly, how it experiences working within it. Doing so will not only address current workforce challenges but also position insurers to be more adaptive, innovative and resilient in the face of future risks.

The organizations that move first will have a distinct advantage. They will attract the talent that others struggle to reach. They will build the teams capable of navigating increasing complexity. And they will be better equipped to deliver on the fundamental promise of insurance: stability in an uncertain world.

In 2026, changing the narrative around insurance jobs is not optional. It is a prerequisite for the industry's long-term growth and stability.

The conversation around legal AI usually follows a predictable script about BigLaw billable hours, the democratization of small firms, or whether an LLM can pass the bar. These debates focus on the visible front lines—the lawyers and the courtrooms. But focusing there means you're watching the wrong game. The real transformation is happening deeper within the economic engine of civil litigation, driven by the insurance carriers. And on the other side, the plaintiff bar is arming up just as fast.

McKinsey estimates generative AI could unlock $50–70 billion in insurance industry revenue. Bain found that 78% of P&C insurers are already using generative AI—though only 4% have scaled it meaningfully. The arms race is underway. Most litigators just haven't noticed the battlefield has moved.

From Colossus to LLMs

Carriers have used algorithmic case valuation for 30 years. The best-known tool, Colossus, is a rules-based system with over 10,000 decision rules, relying on structured inputs, including ICD codes, CPT codes, and severity ratings. If something wasn't in a form field, the algorithm was blind to it. What many practitioners don't realize is that Colossus is reportedly still used by over 70% of insurers. If you've negotiated a bodily injury claim in the last decade, your demand was likely run through it or a similar tool on the other side of the table.

Colossus generated over $293 million in class action settlements and sustained NAIC scrutiny over the "black box" problem of algorithmic valuation. That history matters, because the next generation of these tools is far more powerful and far less transparent.

LLMs are the leap. They don't need structured fields. They ingest the entire case file, medical records, deposition transcripts, and police reports, and spot nuance a rules engine never could. The gap between what a carrier knows about a case and what a plaintiff's attorney knows has always been a matter of leverage. That gap is narrowing fast. Vendors such as Shift Technology, CLARA Analytics, DigitalOwl, and Wisedocs are deploying LLM-driven analysis at scale across the carrier ecosystem. Meanwhile, carriers from Allstate to Chubb are building proprietary tools internally.

The Data Moat Is Eroding

The carrier's deepest advantage isn't computing, it's context. Carriers sit on millions of closed claims, private settlements, and internal outcomes that never see a public docket. A carrier AI doesn't just know what a jury in Cook County did last week; it knows what the carrier paid to settle 10,000 similar cases over the last decade without a trial. That training data is unique.

But the moat is narrowing. CLARA Analytics operates a contributory database trained on millions of closed claims across its carrier clients. On the plaintiff side, EvenUp, now valued at over $2 billion, has crowdsourced actual settlement data from over 2,000 plaintiff firms processing roughly 10,000 cases per week. The information asymmetry that defined carrier leverage for decades is real, but both sides are now building proprietary data assets. The gap is closing.

When Models Argue With Models

This isn't hypothetical any more. In January 2026, a startup called Mighty launched a platform that acts as an AI agent negotiating personal injury settlements against carrier AI on behalf of consumers. Its CEO stated plainly: the company gives consumers AI to negotiate with the insurance company's AI. This builds on decades of automated dispute resolution. Cybersettle alone has facilitated roughly 200,000 claims totaling $1.4 billion using algorithmic double-blind settlement since the late 1990s.

Now imagine the next step. A plaintiff firm's AI evaluates a case at $850,000 based on crowdsourced settlement data. The carrier's AI, trained on 40 years of internal claims history, pegs it at $320,000. Does a shared analytical baseline strip away posturing and accelerate resolution? Or does it entrench positions because each side treats its own model as truth? We risk moving from a world of legal judgment to a world of model drift, where outcomes depend less on case facts and more on whose training data runs deeper.

Regulators Are Already on the Case

When a carrier's AI determines a claim is worth zero, how does a plaintiff challenge that logic? Regulators have been working on this since at least 2021. As of early 2026, at least 25 states plus D.C. have adopted the NAIC's Model Bulletin on AI, requiring written governance programs, consumer notice when AI affects decisions, and bias testing. Colorado has gone further, SB 21-169 requires quantitative bias testing for AI used in claims handling, with enforcement tools including civil penalties and license revocation. The black box problem is real, but it's an active regulatory battleground, not an open question. Practitioners who don't understand the compliance landscape their opponents operate under are leaving leverage on the table.

Nuisance Value

If carrier AI gets better at early case triage, the economics of "nuisance value" - paying $5,000 to make a weak claim go away rather than litigating - could shift. Claims that used to settle for small sums may face an automated "no." But let's be honest: there is no published empirical evidence that AI triage is currently eroding nuisance settlement patterns. This is a plausible hypothesis, not an observed trend. And the counter-argument has merit, if AI reduces per-claim evaluation costs, carriers might become more willing to pay small amounts quickly, not less. Conversely, if a model flags a case as high-exposure early, carriers have every incentive to settle fast rather than lowball a claim they're likely to lose at trial.

A New Equilibrium?

Law is an adversarial system. When one side upgrades, the other responds. Carriers are deploying AI across claims processing, litigation prediction, and settlement valuation. The plaintiff bar is responding in kind. Contributory databases are eroding data monopolies. Regulators are imposing transparency requirements that may force carriers to show their work in ways they never have before.

The question for litigators isn't whether AI will change how cases are valued. It already is. The question is whether you understand what's in the black box on the other side of the table—and whether you have your own.