The adoption of AI in the insurance industry is no longer a distant promise.
It is happening right now, in real time, and it is reshaping how brokerages, carriers, and claims managers operate on a daily basis. We are not talking about lab experiments or pilot projects gathering dust on a shelf. We are talking about concrete changes that affect processes, teams, and actual financial results.
But there is an important detail that many people still have not stopped to think about: going beyond simple automation requires much more than picking the right tool. It requires rethinking structures, strategies, and above all, the relationship between technology and the people who use it. And that is exactly where many companies stumble, because they assume all they need to do is flip the switch and let it run.
That is precisely what a group of leaders in the U.S. insurance market gathered to discuss at the Insurance Business TV Leaders Network. During the session, hosted by Paul Lucas, experienced executives from companies like WTW (Newfront), Alliant Insurance Services, Howden, C3 Risk and Insurance Services, Baldwin Group, and Stockweller and Shepley put three central themes on the table that are dominating the most relevant conversations in the industry:
- How AI adoption is actually being implemented in underwriting, claims, and distribution
- The strategic role of data in the relationships between brokers, MGAs, and carriers
- And the structural changes this movement is demanding from the entire value chain
The conversation was honest, direct, and at times quite surprising. 👀 Spoiler: not everything is hype, but it is not as simple as it seems either.
AI adoption that is actually moving off the drawing board
One of the first things that became clear during the conversation among leaders is that AI adoption in the insurance industry is far from uniform. Some companies are already reaping practical results, while others are still trying to figure out where to begin. And that gap is not due to a lack of interest but rather a lack of clarity about what truly matters when it comes to implementing artificial intelligence in a functional and sustainable way within a complex operation like that of a carrier or brokerage.
Jennifer Wilson, cyber practice leader at WTW who came from Newfront, shared how the brokerage developed several tools focused on creating efficiency and accuracy. Among them, a platform that brings together all client data — such as applications, policies, contracts, and claims — along with a contract review tool that makes coverage recommendations and confirms compliance with contracted policies. They also built a coverage gap analysis that compares quotes and policies in a measurable way.
But Jennifer offered an honest and valuable warning: the biggest surprise was the sheer size of the investment required — and not just from engineers, but from the teams as well. To get the tools up and running, an enormous amount of data had to be entered manually, while everyone was already swamped with their existing workload. On top of that, insurance specialists had to train the tools, something nobody had anticipated as part of the job. As Paul Lucas summed it up nicely: it was a marathon to get there, but now they get to enjoy the sprint. 🏃
Receive the best innovation content in your email.
All the news, tips, trends, and resources you're looking for, delivered to your inbox.
By subscribing to the newsletter, you agree to receive communications from Método Viral. We are committed to always protecting and respecting your privacy.
What this experience shows is that the initiatives that worked best were the ones that started with a focus on real, measurable problems. Instead of trying to transform everything at once, the companies that moved forward with the most consistency chose specific pain points and built automation solutions around them. This incremental approach, which seems obvious, is exactly what many organizations overlook in the rush to show quick results.
Accessibility and smoother processes
Jackie Castellis, claims advocacy leader at Alliant, pointed out that the biggest impact of technology has been the simplification of processes, both internally and for clients, with accessibility from anywhere. She noted that the pandemic taught the lesson that a lot of work can be done remotely, and that technology empowers teams to make better decisions and support each other at the click of a button.
Still, she echoed Jennifer’s warning: the human element will always add complexity. When new tools are introduced without clear direction and without the purpose being well explained, things stall. Proper training is what truly sets teams up for success.
Data: the asset that still is not being leveraged properly
If there was one topic that dominated a large part of the debate, it was data strategy. Most companies in the industry have an absurd amount of data available, but few actually know what to do with it. And that is not empty criticism — it is a structural reality of the market, where legacy systems, information silos, and manual processes still coexist with more modern technologies in a pretty uncomfortable way.
Dale Kupravich, SVP of operations and client service at Howden, explained how any tool that gives colleagues a more complete view of the client’s risk is a positive thing. In the property intelligence space, for example, the company gathers characteristics of a property such as wildfire risk, flood exposure, hurricane threat, and loss history, delivering far more data to underwriters so they can make faster decisions. This improves the way business is submitted, identifies underwriting concerns, and allows them to advise clients on risk mitigation to make their properties more insurable.
Alan Madden, senior vice president and producer at Alliant, showed how data is changing the relationship between brokers, MGAs, and carriers. He gave concrete examples: workers comp carriers offering AI-driven solutions to reduce claims and adjust premiums from the outset, with cameras at facilities and wearables providing real-time feedback. He also mentioned that property submissions sent to London are already using AI for automated risk selection, with line slips fully driven by artificial intelligence. Alliant itself has a slip that matches capacity automatically, segmenting risks without an underwriter needing to be involved.
But Alan raised a point that many people forget: with so much information available, the broker needs to tell the client’s story and risk profile more strategically than ever. It is about bringing human intelligence to provide context for what artificial intelligence captures. As he put it, yes, that data point exists, but here is the reality of what the client actually does. 🙌
AI helping underwriters find clients
Joe Earl, cyber practice group leader at C3 Risk and Insurance Services, was straightforward: underwriters are very tech-savvy, and for more than 10 to 15 years they have been researching everything they can find on big data. Now, with AI, they plug in the company name, the application, run open-source searches, and apply tools that assign a risk score — defining pricing, qualification, and even the likelihood of winning the business.
Jessica Theer, financial institutions practice leader at Stockweller and Shepley, agreed and shared that the company partnered with a platform that is especially useful in the highly regulated financial institutions space. The tool pulls data from sources like the Securities and Exchange Commission website, quickly filters what underwriters need, and eliminates that endless back and forth of questions. The result is faster quoting, transparency about where pricing is landing in real time, and in her words, a win-win scenario for all parties. 💡
The real barriers to technology adoption
Not everything is smooth sailing, and the leaders were quite candid about the obstacles. For Jackie Castellis, the biggest challenge is change itself. As much as people say they embrace the new, they settle into their routines, their programs, and the processes they already know. That is why the most successful organizations do not just implement technology — they invest in helping employees understand why the change matters, showing the return for the individual as well.
Emily Selk, senior director and national cyber practice leader at Baldwin Group, added that there is a lack of understanding about AI and what it actually does. According to her, it is not enough to know how to apply AI in daily work — you need to understand the granularity of that application within operations. And that granularity needs to reach the C-suite, which does not always happen. Often, technology is tossed to the teams without leadership understanding every step of what is being done. The result is a messy implementation, with tools being used in ways that are sometimes appropriate and sometimes not so much.
Emily also highlighted that the process today is still very siloed. Brokers do one thing, underwriting does another, and claims does yet another, even within the same organization. Instead of creating a free flow of information, we are building everything in silos. And she stressed that the building needs to be done with each other in mind — something the industry still has not mastered.
Jennifer Wilson fully agreed and shared the practical experience of when the engineers launched the tools all fired up, expecting everyone to just start using them. It was tough. The solution was for the C-suite to step onto the field to make sure everyone understood how and when to use the tools, create guardrails, and encourage adoption. Instead of massive training sessions for the entire company, they formed small groups where people felt more comfortable asking questions and testing things out.
Jennifer also brought up two important structural points: this is an industry with centuries of history, buried in paperwork, so legacy systems are a real barrier. And there is also the generational gap — whether internally or within the client base — since certain technologies do not appeal to every profile.
The future in three to five years
When the conversation turned to looking ahead, the answers got fascinating. Joe Earl predicted three types of companies: those that use AI as an excuse to cut headcount, those that use AI to better leverage their teams by expanding what each person can accomplish, and the innovators that use AI for exponential growth — hiring people who are already AI-proficient and training their current teams. The last group will be the winners.
Joe also offered a bold warning: there could be a generational breakdown, with people who grew up outsourcing critical thinking to AI throughout school and college. In his view, companies need to keep hiring people who know how to use their own brains just as well as they use artificial intelligence. 🧠
Dale Kupravich agreed that the human aspect will always be important and that decision-makers will always be needed. He sees AI embedded into daily work within five years, handling administrative and low-value tasks so that people can focus on client relationships. The transformed broker, in his view, will not be the one who uses the most AI tools, but the one who has a model where data, technology, and human expertise work together.
Alan Madden showed that Alliant is already heading down this path in a thoughtful way. The company directed employees not to use ChatGPT or Claude until they had studied the tools, understood the responsibilities, and created guardrails. Today, about a quarter of their 16,000 employees — roughly 4,000 people — already have access to a unified Alliant AI platform, with mandatory training and certification. Every month there are seminars and webinars showing how to use AI more effectively in their workflows. The company also adopted an AI-enabled version of Salesforce, which is great for client management, prospecting, forecasting, and even identifying where future hiring needs will arise.
Jessica Theer closed with a spot-on prediction: over the next three to five years, there will be a race for quality. First, in how organizations protect and create guardrails around the information fed into AI platforms and the security surrounding them, since some companies may fall short on this front. And second, a race for quality to find who truly turns operational efficiencies and AI-driven data into sound client advice. As she summed it up, everyone will be able to place coverage easily — the differentiator will be doing what is right and in the best interest of the client.
What all of this means in practice
Connecting the dots from the discussion, one message is clear: AI adoption, data strategy, and structural change are not separate topics that can be addressed independently. They feed off each other, and the success of any initiative depends on how all three are aligned within a long-term vision.
Automation solves important operational problems and frees up team time to focus on what truly requires human judgment. But without well-structured data, automation does not have the foundation to work properly. And without structural change to support it all, any technological advancement risks becoming just another layer of complexity on top of processes that were already complicated.
The insurance industry is at a real inflection point. The conversations captured at the Insurance Business TV Leaders Network show that the most prepared leaders already understand that sustainable efficiency does not come from the most expensive tool, but from the smartest strategy combined with the human touch. And smart strategy starts with the right questions, not the easiest answers. An edition that put quality first — and just the beginning of a series of gatherings that promises plenty of great conversation ahead. 🚀
