Today, a typical enterprise security team comprises over a dozen specialized functions, such as alert investigation, incident response, threat hunting, vulnerability management, and penetration testing. Even within the application security team, the people doing threat modeling and security assurance are very different from the people doing static application security testing (SAST) and dynamic application security testing (DAST), for example.

As AI agents and assistants increasingly take on the nuts and bolts of many security functions, the role of the security professional transforms into guiding the AI, feeding it the right data and business context, and making more strategic decisions for the business.

In this new world, the most important skill is understanding the shared business context and priorities across the organization, not knowledge of a specific tool or specific alert details. The enterprising employees who do this effectively are the ones who will excel in 2026. They will be positioned to take on any role in the security team, and perhaps all roles at the same time—part-time SOC analyst, part-time AppSec engineer, part-time threat hunter and more. This is because they can simply delegate the domain-specific details to specialized AI agents and assistants.

This is great for CISOs, as they get a highly fungible team of security generalists who can take care of whatever is the top security issue of the day in any domain. Having a single person take care of issues across all domains creates fewer gaps. And it's also great for security team members as it gives them mobility in their careers. It's a win for security across the board. If you're a security generalist, 2026 is your year.

Deepfakes have been a common problem on the Internet pre-2025. In 2025, they entered the workplace, with many incidents of fraud involving adversaries posing as interview candidates or a business partner in a video call.

An orthogonal problem in workplaces has been rogue insiders, employees who hurt their organizations from inside. Sometimes they do this on behalf of an external adversary in return for money, whereas others are lone wolves.

In 2026, these two will converge, with rogue insiders leveraging AI and deepfakes. Employees who have the proclivity to cheat but were previously afraid will be encouraged to cheat, with AI making it easy and deepfakes providing plausible deniability. Any insider has all the business context to customize deepfake attacks to seem much more real than anything we've seen in 2025.

The research on how to detect and defend against deepfakes, as well as techniques to detect and defend against rogue insiders, needs to catch up and tackle this threat effectively. Until it does, there will be a period in 2026 when trust within organizations will be broken. You no longer trust that the email, or the voice call, or even the video call from your teammate is truly from your colleague. During this phase, regular employee training combined with fast adoption of better deployment tools will become critical.

Additionally, the offensive security landscape will change more in the next 24 months than it has in the last 10 years. Traditional pen-testing has remained largely manual and very expensive, while DAST tooling is great at surface-level scanning, weak at context and logic.

In 2026, we'll see new automated approaches to offensive security that understand context, state, and business logic, not just endpoints. Think tools that behave like a creative attacker—chaining vulnerabilities, exploiting misconfigurations, and validating impact the way a human red-teamer would.

That evolution will turn what used to be a quarterly or annual pentest into something continuous and integrated into engineering workflows. Security shifts left to match attacks that are doing the same, into CI/CD, pre-prod validation, and runtime guardrails. Once offensive testing becomes autonomous and contextual, organizations will stop treating pentests as compliance artifacts and start treating them as live safety nets for every software change. 2026 will be the year offensive security becomes just another part of the delivery pipeline.

As AI transforms both security attacks and security tools, this fragmentation hurts agility, restricts scalability, and most importantly creates silos where adversaries hide. We predict this will change in 2026.

AI in insurance has long been discussed in terms of speed and precision; enabling faster underwriting and quicker claims processing, or better risk scoring and fraud detection.

But a new conversation is emerging - one that sees AI not just as a tool for efficiency, speed, or simply automation, but also as a means of creating agility.

It's a shift that couldn't be more timely. The industry is being reshaped by a perfect storm of disruptive forces, all unfolding at once and all demanding faster, more flexible ways of working. Natural disasters have rewritten risk profiles across regions, persistent inflation pressures have pushed carriers to shift focus from premium volume to profitability, and high interest rates have driven withdrawals and non-renewals. Meanwhile, huge advancements in technology and abrupt customer shifts have seen baseline expectations rise even further, especially among the more digitally savvy.

The result is that many agents and carriers struggle to keep up, with traditional legacy systems, compliance processes and slow decision-making cycles slowing them down.

Agility - which, in these terms, is classified as the ability to rapidly scale capacity, instantly apply improvements across an organization, or adapt decision-making in real time to new data - is fast becoming the industry's most valuable currency.

How AI is already bringing agility into insurance

As it stands now, the industry is already seeing early indicators of AI's ability to boost agility where traditional processes are constrained by human resource limitations, training timelines, or system dependencies. Examples include:

• Real-time market adaptation: When carriers shift their risk appetite or market conditions change rapidly, traditional systems rely on API updates or core system changes that can be painfully slow, inconsistent, or lacking in nuance. AI provides remarkable flexibility here - new information from documents, marketing brochures, or support tickets can be ingested directly into the AI engine, enabling immediate updates to underwriting logic and business processes. This means insurers can respond in real time to changing market conditions instead of being constrained by legacy system limitations.
• Strengthening agent-carrier connections: Not only can advanced analytics help agents identify the right carrier fit for their specific needs and risk profiles, but the same technology can work bidirectionally - enabling carriers to identify and connect with agents who align with their business strategies and distribution goals. This approach complements traditional relationship-building with a more strategic, data-driven method.
• Making data analysis easy - and accessible: For agents, AI can make decades of industry data available for easy analysis, helping even new professionals deliver seasoned-level insights to customers. When consulting firm-quality research and analysis is available to decision-makers at every level, complex questions that traditionally took hours or days to resolve can be figured out in minutes.
• Improved speed and operational consistency: AI is known for its speed and consistency, but being able to respond to change in real time enables true agility. For carriers, this extends to claims processing, underwriting, and employee workflow optimization, where AI has demonstrated completion rates that actually exceed human performance while delivering more consistent service across networks.
• Ability to scale capacity overnight: AI eliminates the bottleneck that traditional scaling causes. In phone call management, for example, if a carrier wanted to scale up outreach or avoid missing an influx of incoming queries, they'd need to post job listings, interview, negotiate, hire, onboard, and train - a lengthy process that for entry- to mid-level roles can take anywhere from one to six months before anyone is truly productive. With AI, that capacity can be dialed up overnight.
• Experimentation is far simpler and faster: Easing the process of experimentation allows organizations to respond very quickly to business conditions, needs, and insights. Following on with the phone call management example, traditionally, testing different scripts for handling in- and outbound calls means splitting agents into groups, measuring results, deciding which approach works best, and then retraining people. However, if a new script or approach works better with AI, it can be applied instantly across every interaction, without the slow grind of retraining or overcoming resistance to change.

What AI-driven agility means for the future of insurance

While all of the above is already starting to improve agility across the industry, if we look a bit further down the line, AI's capabilities could have an even bigger impact. It could, for example, act like a virtual sub-agent, capable of finalizing or even binding straightforward policies. A bit like self-service, but with the reassurance that a human agent is still there, making customers feel comfortable while speeding up the process.

It could also affect the traditional quoting experience, such as comparative quoting. In this sense, AI wouldn't just pull APIs and return quotes, deductibles, premiums, and limits - it could also provide deeper insights into the nuances of each coverage. For example, it could draw on customer reviews, past issues, and other relevant data to give a more complete picture. In this way, it could act as a force multiplier, enabling agents to deliver richer, more informed advice to customers. Even a new agent could have decades of experience at their fingertips, helping them provide the same depth of insight as a seasoned professional.

Regardless of which possibility we explore in the future, the main point here is that AI can, and very much should, provide a much quicker, more agile way to respond to change - an ability that's becoming increasingly essential as change itself continues to accelerate across insurance.

The esteemed science-fiction author Arthur C. Clarke believed that "any sufficiently advanced technology is indistinguishable from magic." When it comes to artificial intelligence and its widening effect on the insurance sector, 2026 promises to be a magical year.

In recent years, AI has been tested, examined, admired, and even feared, but not often deployed at scale in the P&C market. That paradigm, I am convinced, will change in the coming months. Look to 2026 as the year when AI progresses from pilot to production across insurance sectors. That's when I expect to see more real-time underwriting, conversational experiences, and dynamic pricing, with many players shifting from the "dipping your toe in the water" phase to the "taking the plunge and swimming confidently" stage. Already, we're seeing underwriting models that can continuously learn, customer interactions that feel like natural conversation, and pricing that adapts to different behavior, context, and risk signals. When I observe how swiftly we've gotten here – I can't help but be bullish on the near future.

The latest numbers buoy my confidence. S&S Insider reports that the worldwide market for artificial intelligence in insurance is set to jump from about $8.6 billion in 2025 to nearly $59.5 billion by 2033. Annual growth rates of approximately 27% suggest an industry quickly implementing AI for claim streamlining, fraud detection, customer service enhancement, and stronger risk management. What's more, early AI adopters are benefiting from cost reductions of 20% to 40% across claims, onboarding, and back office operations, as well as premium growth of up to 15% thanks to improved customer segmentation and more personalized offerings, according to McKinsey data.

AI moves from experimentation to everyday operations

I foresee several transformative trends in 2026 based on advancing technologies and data innovations. One particular breakthrough will come from unifying fragmented data. One of insurance's oldest and stickiest problems has been that policies reside in one system, claims in another, and interactions in a dozen more. But the accelerating push to combine every piece of data into a single intelligent layer that connects all policies, claims, and consumer interactions will pay off this year. Innovations in entity resolution, retrieval-augmented generation, and privacy-safe synthetic data will unlock personalization at scale while safeguarding customers. (It's encouraging that, according to SAS, 79% of carriers are open to using – or are already employing – synthetic data to resolve privacy and data-quality challenges.) Those insurers who check these boxes will win both efficiency and trust from consumers.

I'm also excited about the increasing acceptance of generative AI. Consider that 82% of insurance companies adopting AI are also incorporating GenAI, which demonstrates a rapid progression toward more naturally conversational and content-driven experiences. Additionally, 2026 is poised to be another year of innovation and progress in advanced driver assistance systems (ADAS), which won't necessarily make fully autonomous driving mainstream but could lead to greater clarity involving liability assignments and claims patterns. S&P Global expects that, by 2035, around 40% of new vehicles sold around the world will include advanced driver-assistance features from Level 2-plus to Level 4.

The future belongs to trust builders

What's more, AI is positioned to remove friction from interactions in the coming year, converting what used to be a transaction into a relationship. By analyzing intent, preferences, and behavior in real time, AI-equipped carriers will truly understand – not just price quote – their clients. The companies that stand to benefit most are those that deepen trustworthiness via personalizations that feel effortless and provide real value to the consumer, leading to partnerships that foster long-term loyalty.

Looking ahead, we can also expect insurance technology teams to further evolve. I envision a shift from teams that build isolated services to those that orchestrate intelligent ecosystems. AI systems engineers, data governance leads, and machine learning operations specialists will be among the most indispensable roles. And the best teams won't just be deep technically: they'll blend domain fluency with adaptability, curiosity, and cross-functional collaboration as AI becomes further embedded into every layer of the business.

Obstacles and opportunities

Of course, significant challenges remain. One of the biggest is balancing AI-driven automation with the need for transparency, fairness, responsible governance, and human oversight. After all, automation without transparency erodes trust, and in insurance, trust is the whole ballgame. Although only around 5% of insurers have a fully mature AI governance framework in place right now, Market.US reports that nearly 70% of large enterprises are currently investing in fairness controls, audit trails, and model monitoring. I anticipate that 2026 will be the year when carriers embed explainability into every decision-making model, making "why" as visible as "what." That doesn't mean human oversight will disappear; instead, it will evolve into governance frameworks that ensure fairness, auditability, and consumer confidence.

The new year gives us a lot to look forward to. But it also reminds us that the AI timeline in insurance is unfolding at record speed and is sure to present fresh obstacles and unforeseen X factors that could upset even the most reliable predictions. Still, I've never been as excited about what's beginning to emerge just beyond the horizon.

Group health insurance executives: If your AI is constricted, so are your policies.

In recent years, underwriters and other group health insurance policy designers have leaned in to artificial intelligence tools. In fact, a recent NAIC survey of U.S. health insurers showed that 84% currently use some form of AI and machine learning.

This substantial shift toward AI policy design and management tools is completely understandable – and completely necessary to stay competitive. In a sector where rising costs, unpredictable claim patterns and shifting risk profiles continue to hinder forecasting precision, business-as-usual methods often fall short as teams responsible for assessing and managing risk are charged with making faster, more accurate decisions with fragmented data.

In this increasingly demanding landscape, AI solutions can analyze large datasets quickly, apply consistent methodologies and uncover insights that otherwise would have gone overlooked. In addition to expediency, such tools can yield cost containment, pricing transparency and deeper customization. Unsurprisingly, then, 75% of executives polled in a recent Roots Automation survey deemed AI tools key to premium growth; more than half reported that AI accelerates the quoting process, and nearly half are using AI solutions to help reduce loss ratios.

However, fully realizing the risk management benefits of AI solutions means bringing them out of their single-use silos. The underlying principle is simple: since AI excels at mining and measuring multiple factors with exceptional speed… why limit the data it analyzes? After all, the more factors group health policies can consider, the more accurate, resilient and cost-effective they will inevitably be.

This article explores how group health insurers can optimize AI usage and maximize its game-changing effect. This can only be achieved when AI-powered solutions are integrated into enterprise-level workflows to facilitate consistent, data-driven insights throughout policy lifecycles.

The AI Silo Trap

Anyone of a certain age will remember the big, boxy desktop computers of the 1980s and early 1990s. As the PC revolution took off, so did the ease and speed at which once-onerous tasks could be performed. Everything from word processing to number crunching became a lot easier in a hurry. It was useful, impressive and altogether helpful.

And it was nothing compared with what came next: networking.

Once computers were linked to each other via the World Wide Web, their applications and usefulness were exponentially amplified. Knowledge could be collected more broadly; trends and the opportunities they uncovered could be noticed and acted upon more quickly. The tagline of the day may have been "You've Got Mail," but the force that drove the internet's rapid proliferation was that, suddenly and forevermore, our newly connected computers provided access to far more knowledge than any one PC could offer. Knowledge shared was knowledge gained.

Fast forward to today, and artificial intelligence is emerging from its nascent, newfangled days into the biggest buzzword on the planet, let alone the insurance industry. And like the pre-internet days of PCs, the potential benefits of today's early-stage AI solutions – let's call them AI 1.0 – are being underused largely for lack of connectivity.

In this still-siloed landscape, many organizations that design and manage group health policy lifecycles rely on one set of AI tools and methodologies for assessing risk in new business, another for existing client renewals, and yet another for managing member health risk. Still, the results have been undeniably encouraging: with growing volumes of consumer data available from medical, prescription, and lab sources, even rudimentary AI solutions are making crisper, more confident decisions that go beyond the limits of personal judgment and historical patterns.

Unfortunately, these benefits have led to blind spots. AI solutions have proven so promising that most organizations have overlooked the logical next step: connection.

While AI solutions in and of themselves are exceptional inventions, their effect is limited when constrained to single-set columns. Among other pitfalls, this approach may lead to disconnected data and a potential inability to account for shifts in group risk profiles.

In an environment as multifactored and ever shifting as group health insurance policy lifecycle management, the time has come for AI solutions to take the natural next step in their evolution. Insurance players are well-advised to move from standalone AI processes to enterprise-level, full-cycle workflows that align risk management across new business acquisition, population health management, and existing group renewals.

Better Together: Connected AI Solutions

Much like the dawn of high-speed internet in the late 1990s, group health's fledgling "AI 2.0" era promises unprecedented advantages. Opportunities now exist to transcend segmented AI tools by implementing sweeping AI solutions that provide truly integrated lifecycle risk management. Simply put, such solutions replace several stagnant tools that each examine one aspect of policymaking with one versatile solution that monitors all aspects.

When AI solutions are properly integrated across the myriad datasets inherent in group health, they can maintain tightly controlled continuity across policy lifecycles. First and foremost, connectivity breeds data consistency, which in turn supports enhanced decision-making while providing actionable, member-level insights to power care management solutions.

By rooting decisions in the same risk logic across initial quoting, renewal pricing and continuing population management, this un-siloed, unshackled approach enables end-to-end application of shared data signals and risk methodologies. The result is reduced variability and improved portfolio performance.

Like the internet before it, such solutions thrive on one overarching principle: knowledge shared is knowledge gained. Faster quote turnaround times reduce underwriting lifecycle friction, and unified historical and real-time data inform seamless renewal transitions. With group health's countless footnotes and fine print suddenly on the same, succinct page, the resulting reliable benchmarks optimize underwriting strategies through the newfound ability to measure performance and identify improvement opportunities at each stage of a group's lifecycle.

Of course, any transition can bring challenges – including, for starters, determining precisely how to begin. At the inception of the enterprise-level AI integration journey, insurance organizations should carefully consider their key priorities. At the heart of this introspection is one question: what does interconnectivity-driven success look like?

What challenges are the organization's policy lifecycle management experiencing because, for example, its new business and renewal underwriting tools are separate? Which workflows or decisions would benefit most from shared data and consistent risk scoring? What internal systems or processes will need to connect with the new platform? And of course, how will we train and properly prepare our workforce for this next-generation solution?

In many cases, these considerations mirror patterns seen across the broader market. Let's close with a few examples showcasing the value of AI solution synchronization.

Use Case #1: Detecting New and Emerging Health Risks During a Policy Term

Based on the original census, an employer group appears healthy during initial policy quoting. Of course, several factors can affect this risk assessment, including final member enrollment and the entrance of additional members during the policy's lifecycle. To better account for these factors, an integrated risk scores solution can provide supplemental data that informs pricing and cost containment strategies at renewal.

Integrated risk score solutions can be especially valuable to companies with high member turnover, or that have newer groups with limited experience. The goal is to supplement a group's limited experience-based risk scores with models that mine third-party datasets.

Use Case #2: Consistent Risk Scoring for Refined Pricing Accuracy

Using different tools for new and renewal underwriting can result in inconsistent risk assessments and pricing. An integrated approach uses the same underlying data signals and modeling logic across both business phases.

Such consistency supports fairer and more accurate pricing, reduces volatility in rate changes, and strengthens relationships with employer groups that expect predictability. When the same factors drive decisions from quoting to renewal, underwriting teams can take different actions based on risk scores, explain pricing shifts more clearly, and maintain trust.

Use Case #3: Improving Underwriter Efficiency and Speed

When new and renewal underwriting data resides in separate systems, underwriters may spend extra time reconciling information or duplicating analysis. With an integrated workflow, teams gain access to a consolidated view of group data, historical insights, and predictive signals in one place. This eliminates repetitive work, speeding up both the quoting and renewal processes. As a bonus, faster decisions mean insurers can respond more expediently to broker and employer requests.

Artificial intelligence promises speed, analytics, and cost savings. Many in the insurance industry see it as the "magic pill" that can fix everything.

But that is an illusion. Algorithms do not build trust. In critical moments, clients remember not the dashboards but how they were treated. Our Ukrainian wartime experience has proved it: technology helps companies function, but humanity is what creates loyalty.

The Illusion of Sufficiency

Chatbots. Automated underwriting. Predictive analytics.

They work — until they don't. The first unexplained denial. The first claims glitch. The first case where the system is "technically correct," yet the customer feels betrayed.

Efficiency is about numbers. Trust is about people.

What AI Cannot Replace

AI can optimize processes. But some things remain deeply human:

• Empathy. On the day someone loses a home or a car, they don't need a bot. They need support.
• Ethics. Statistics do not capture humanitarian exceptions. Human judgment does.
• Leadership. In a crisis, employees and clients listen for an honest voice from the top, not another push notification.

AI can assist. But it cannot show humanity.

Lessons from Practice

We have already seen algorithms cause reputational risks. In mature markets, claims denials without transparent explanations triggered public backlash — and losses far greater than the savings.

In Ukraine, the war became a true stress test. Systems operated under constant disruption. Yet loyalty was built by people — managers answering calls in dark days when power was out.

Clients don't remember the speed of a payout. They remember that someone cared.

The Future: Not AI vs. Humans, but AI With Humans

The winning model is not replacement but partnership.

AI should take over routine: risk analysis, fraud detection, data crunching.

Humans should remain where trust is built.

The companies of the future will not be defined by full automation. They will stand out by combining algorithms with a human face.

Conclusion

AI can count.

But only people can build trust.

Four years ago, days after Mark Zuckerberg debuted the Metaverse, I wrote a Six Things commentary that began: "The vision of a metaverse laid out by Mark Zuckerberg last week is bonkers. Nutso on steroids. It won't be realized in my lifetime, yours or his, even if some of the wildest claims about longevity come true and we all live to be 150."

Since then, the Metaverse group within the company Zuckerberg renamed after what I referred to in that commentary as "a fever dream for gamers" has racked up $70 billion in losses, and Bloomberg and the New York Times reported last week that he is planning to cut staff by between 10% and 30%, possibly in January.

So, in retrospect, I'm just sorry I pulled my punches. :)

Trashing the Metaverse on Day One was not a remotely hard call, because it violated one of the cardinal rules of innovation: As much as possible, an innovation has to fit within the existing work environment or lifestyle of the prospective user. Yet the Metaverse required radical changes in how individuals interact — with, as far as I could discern, no appreciable benefits.

It's worth taking a minute to look at where Meta went wrong, because the mistake is awfully tempting for all of us. 

The Metaverse assumes that people want to live online a huge percentage of the time. You have to produce an avatar to act as you and learn all sorts of new behaviors to interact with other avatars and with everything else that populates the online world. (I tried this a couple of years into the Metaverse experiment, courtesy of a consulting firm that was enthusiastic about its prospects, and it was still quite hard just to maneuver, let alone to talk with others' avatars or to conduct a transaction.) 

The rule of thumb in Silicon Valley is that an innovation has to be 10 times better than anything it is intended to replace, yet the Metaverse was far less useful than the Zoom calls and other technologies we already used, while requiring huge changes in people's routines. 

Apple made a similar mistake with its Vision Pro virtual reality device — and yes, I trashed that, too, right after it was announced at the beginning of last year. I wrote: "There's simply no reason to strap a 1 1/2-pound device to your face (nearly the weight of a quart of milk) and put a three-quarter-pound battery in your back pocket so you can type with your two index fingers in mid-air while strangers or officemates gawk at you. Not when some combination of today's laptops, tablets and phones will do just fine."

The Vision Pro has been a dud for precisely the same reasons the Metaverse has flopped. 

By contrast, Metaverse has a budding hit with the AI it has built into Ray-Ban "smart display" sunglasses. The capabilities are still pretty limited but are enough to get started: You can use voice commands to snap photos, record videos, send messages, make calls, and ask questions of Meta's AI. And Meta isn't asking customers to do anything out of the ordinary. Just about everybody wears sunglasses. Besides, Ray-Bans look cool.

When you look at the history of major technology innovations, they almost all replace something similar. Smartphones replaced iPods, which replaced the Walkman, which replaced transistor radios. Smartphones also replaced early cellular phones, which replaced hardwired phones in homes. There was almost no need for changes in behavior; everything just became easier and better.

Note that once you get a new device into people's lives, like a smartphone, you can start to get them to change behaviors that have nothing to do with the original purpose — when I first saw a smartphone demo, some 25 years ago, I had no idea I'd be doing my banking and shopping on a phone, or listening to podcasts on it and having it monitor my driving.

The insurance industry seems to mostly get this principle, that innovation has to fit into existing behaviors. That's why we're seeing so many dashboards that incorporate the advances in generative AI, gathering information and making evaluations in the background and presenting them to underwriters, claims professionals or agents and brokers as part of their normal workflow. I think chatbots were initially seen too much as a standalone technology but are now being integrated much better into the customer experience.  Whisker Labs' Ting device has taken off because a customer simply has to plug it into a wall socket to have it monitor for electrical issues and prevent home fires. Roost built another Predict & Prevent business by offering batteries that can be plugged into existing smoke detectors and ping a customer's cellphone when an alarm sounds, in case they aren't at home to hear it.

Still, the principle is worth keeping in mind, because the temptation — which I've witnessed across industries in my decades of writing about innovation — is to think that what you're doing is so useful that people will adapt to you, freeing you from worrying about how to adapt to them. 

If Meta and Apple can make that mistake, you can, too.

Cheers, 

Paul

P.S. While I've patted myself on the back for dumping on the Metaverse and Vision Pro right out of the gate, I need to acknowledge that I've made mistakes, too. While I can't recall a time when I savaged an idea or product and been wrong, I've certainly been too optimistic about how quickly change would happen. I try to live by the Silicon Valley dictum that "you should never confuse a clear view for a short distance," yet, well, I sometimes do.

For instance, I wrote an article in 1991 or 1992 that said paper forms no longer had a reason to live, given that we could all input information into personal computers connected to whomever or whatever needed the data. That was more than three decades ago, and, hmmmm....

But at least the article only ran on the front page of the second section of the Wall Street Journal, so only a few people read it, right? 

The insurance industry is entering 2026 with pressure mounting from every direction. Regulations are tightening, consumers are walking in more informed than ever, and agencies are still battling the same silent threat that slows them every day. Their systems can't talk to one another. Data shows up late or contradicts itself. And the very people tasked with guiding buyers through complex decisions often lack the real-time information they need to do it.

This is why 2026 is shaping up to be the year of the connected agency. Agencies that reduce fragmentation, unify their data, and adopt technology with intention will be positioned to capture the full value of the industry's next wave of digital investment. Those who cling to disconnected processes will fall behind consumers who now validate information as fast as it is shared.

Clarity Drives Connection

A connected agency is not defined by how many tools it owns. It is defined by how seamlessly those tools work together. The goal is a low-touch, high-accuracy workflow where agents have immediate access to reliable carrier data, leaders see what drives performance, and consumers receive information they can trust without hesitation.

This shift is happening fast. By 2026, digital investment in insurance is expected to climb more than 25%, driven by demand for automated workflows, integrated data systems, and more personalized consumer experiences. Technology is no longer a support function. It is shaping marketing, underwriting, distribution, and customer engagement.

Still, investment alone does not guarantee improvement. Agencies cannot unlock the value of new tools if their data foundation is fragmented.

Fragmentation Holds Agencies Back

Ask agency leaders about their greatest operational challenges and you hear the same themes. Carrier data arrives in inconsistent formats. Product details get updated in one system but not the others. Teams rely on manual workarounds to resolve discrepancies. Compliance reviews become more difficult as regulations change and competitors accelerate their digital capabilities.

This fragmentation is more than a workflow issue. It affects credibility. Consumers have changed the buying dynamic in a way the industry has never seen before. ACA marketplace enrollment reached more than 21 million people, and combined marketplace and Medicaid expansion enrollment climbed past 44 million last year. That scale reflects a buyer base that is more informed, more comfortable with independent research, and more likely to challenge any detail that feels out of sync.

It is now common for a consumer to verify a quote during a live sales call. When the information they find online contradicts what an agent is sharing, trust erodes instantly. The agency that loses trust loses the sale.

The First Step Toward Modernization

Most leaders understand they need to modernize. What they do not always know is where to start. The instinct is to focus on new platforms, upgraded CRMs, or advanced analytics dashboards. But the most effective starting point is much simpler. Agencies need a clear understanding of their historical data.

Knowing who you serve, which products perform, where gaps exist, and how your book of business has evolved gives you the context to make smarter decisions about every step that follows. It also gives you a realistic baseline to measure improvement.

This approach aligns with broader industry research. Deloitte's digital insurance analysis shows that organizations with unified, trusted data foundations see higher returns on modernization efforts because they can automate more workflows and personalize the consumer experience more effectively.

Once the data foundation is clear, the next decision is choosing the right partners. Smaller and mid-sized agencies do not need to build their own technology. They do not need to hire internal engineering teams or maintain complex systems. They need partners that can bring accurate data, integrated workflows, and automated decision support to the table. That is how they modernize without stretching budgets or shifting attention away from their core work.

The Truth About AI's Role

AI is dominating insurance headlines, and there is real potential in the technology. But agencies should approach it with clear eyes. AI can guide consumer decision-making, support call center staff, and help identify needs based on anonymized data inputs. It can also expand access to personalized insights that would be impossible to generate manually.

But AI only works when the underlying data is clean and accurate. If an agency's data is fragmented or outdated, AI amplifies those problems instead of fixing them. Advanced analytics and AI can drive meaningful value, but only with strong governance and reliable data inputs. Without those elements, AI introduces more oversight responsibility and more operational risk.

For most agencies, the path forward is not to become AI experts. It is to build the foundation that allows AI and automation to work safely and effectively when they are ready to adopt them.

Why Connected Agencies Win

Agencies that embrace a connected model see improvements across several dimensions. Efficiency rises because teams spend less time reconciling data. Institutional knowledge becomes easier to preserve as workflows are documented and digitized. Compliance risk decreases because processes are structured and repeatable. And growth accelerates because agents can spend more time selling and less time troubleshooting systems.

Connected agencies also protect themselves from external volatility. Whether regulations change or carriers update products, a unified system makes it easier to adjust without disruption.

How to Evaluate Technology Partners

As agencies evaluate their options heading into 2026, selecting the right partners matters as much as choosing the right tools. Reputation still carries weight, but agencies should look deeper. The strongest partners offer capabilities that feel like table stakes for a modern agency, not add-ons that require extensive customization or additional cost.

The right partner should reduce friction, expand visibility, and help an agency operate with confidence even when the market changes. They should support a clean data foundation, integrated workflows, and the level of accuracy required to serve a consumer base that double-checks details in real time.

What It Takes to Win in 2026

Success in 2026 is not about size or budget. It is about clarity. Agencies that know their data, streamline their workflows, and align with partners that fill technical gaps will thrive. Those who continue working in fragmented systems will find it harder to grow, harder to stay compliant, and harder to keep pace with a consumer base that expects precision.

The industry is evolving quickly, but the path forward is clear. The connected agency is not a trend or a buzzword. It is an achievable and necessary model for every organization that wants to stay competitive in the years ahead.

In an era when cyberattacks are escalating in frequency, sophistication, and financial impact, the insurance industry finds itself in a peculiar moment—not only underwriting cyber risks for clients but also struggling to protect itself from those very threats. The rising tide of digitalization has created an urgent need for cyber resilience within insurers' own operations. However, the industry is facing a growing challenge: a critical shortage of cyber talent.

As we move into 2026, insurers are increasingly asking a difficult question—can they insure themselves against cyber threats when they are struggling to hire and retain the very talent needed to defend their own systems?

The Double-Edged Sword of Cyber Risk

The insurance industry is uniquely positioned in the cybersecurity conversation. On one hand, insurers are developing and pricing cyber risk products for clients—especially businesses vulnerable to ransomware, data breaches, and phishing scams. On the other, their own systems hold vast amounts of sensitive personal and commercial data, making them lucrative targets for hackers.

Cyber insurance is one of the fastest-growing lines of business in property & casualty (P&C). According to industry projections, the global cyber insurance market is expected to exceed $30 billion in premium volume by 2027. Yet this growth is tempered by mounting internal threats and limited in-house cybersecurity capacity.

Internal Vulnerabilities on the Rise

Recent incidents have made clear that insurance firms are not immune to breaches. In fact, attackers see insurers as high-value targets due to their access to confidential policyholder data, claims histories, and financial information.

Despite increased investments in firewalls, intrusion detection systems, and endpoint protection, many insurers lack the personnel to monitor and respond to cyber incidents around the clock. The result? Gaps in defense, delayed response times, and higher exposure to reputational damage and regulatory fines.

The Cybersecurity Talent Shortage

At the heart of this vulnerability lies a growing talent crisis. Cybersecurity roles—such as threat analysts, security architects, and SOC (security operations center) analysts—are among the hardest to fill in the insurance sector. According to (ISC)², the global shortage of cybersecurity professionals stood at over 3 million in 2024, and the demand has only surged since.

Insurance firms are particularly affected because they must compete with tech giants, fintech startups, and government agencies that often offer more dynamic roles, faster career progression, and higher compensation. Many young professionals perceive the insurance industry as slow-moving or less innovative, further compounding hiring difficulties.

Legacy Systems and Innovation Drag

One of the key barriers to attracting cyber talent is the industry's continued reliance on legacy systems and outdated IT infrastructure. For cybersecurity professionals trained in modern cloud architectures, DevSecOps, and zero-trust frameworks, legacy environments are often perceived as stagnant or restrictive.

While some carriers have accelerated their digital transformation journeys, many are still in transition, which creates both technical and cultural obstacles for cybersecurity hires. This gap between modern cybersecurity demands and legacy environments makes onboarding and retention all the more difficult.

What Can Insurers Do?

To address the cyber talent crunch, insurers must rethink both their talent strategy and their organizational culture. Here are a few steps leading carriers are taking:

1. Rebrand Insurance as a Tech-Forward Industry

Firms need to reposition themselves as digital leaders. Highlighting innovation in AI-driven underwriting, blockchain-based claims processing, and cloud-native architectures can attract a new generation of tech talent who want purpose-driven and cutting-edge roles.

2. Invest in Internal Talent Pipelines

Rather than exclusively hunting externally, insurers can build internal training programs to upskill existing IT staff in cybersecurity. Partnerships with universities, bootcamps, and certification bodies like CompTIA, (ISC)², and SANS can help develop talent in-house.

3. Strengthen CISO Leadership

Chief information security officers (CISOs) must be empowered with a direct line to the board, strategic autonomy, and a clear mandate to drive transformation. Elevating the visibility and authority of cybersecurity leadership can improve team morale and signal seriousness to prospective hires.

4. Leverage Managed Services and AI Tools

Until talent gaps are fully addressed, insurers can turn to managed security services providers (MSSPs) and AI-based threat detection tools to bolster their defenses. Automation can't replace humans, but it can reduce the burden on limited teams.

5. Create Mission-Oriented Cyber Roles

Younger professionals are increasingly motivated by purpose and impact. Insurers can emphasize the role their cybersecurity staff play in protecting policyholders, critical financial infrastructure, and even disaster response systems.

A Call to Action for 2026 and Beyond

The cyber threat landscape isn't going to ease anytime soon. As quantum computing, generative AI, and decentralized finance (DeFi) introduce new vectors of attack, insurance firms must urgently fortify their digital perimeters.

But technology alone isn't the answer. The human layer—those who configure, monitor, and manage these systems—remains the most vital and vulnerable link. Closing the cyber talent gap is no longer a back-office IT issue; it is a business-critical challenge that could determine the long-term viability of an insurer.

The industry must act boldly. This includes building more inclusive pipelines, embracing flexible work models, offering competitive compensation, and nurturing a mission-driven, security-first culture.

Because in 2026, it's no longer just about insuring others. It's about ensuring the insurer can protect itself.

Workers’ compensation insurers are turning to generative AI to improve injured worker outcomes, strengthen performance, and build safer workplaces—here’s how:

Read Now

Sponsored by ITL Partner: PwC

From ROI and productivity gains to AI adoption and new market entrants, every signal influences how much trust you place in insurtech today.

Share your perspective in this anonymous survey about the insurtech your company relies on. Have a say in determining where trust is built and where it breaks.

Take the 5-minute survey

Sponsored by Benevolent Marketing