Insurtech Insights 2026: AI Growing Pains

• June 9, 2026

By Dave Evans

Last week’s 2026 Insurtech Insights conference was dominated by all things artificial intelligence (AI). While AI has been the focus of the past two conferences, the messaging has evolved from “What is AI?” to “How is AI being deployed in your organization?”

As attendees arrived in New York City for the conference, Google’s parent Alphabet announced an $80 billion stock sale to fund more AI initiatives and development.

 Considering that Alphabet’s market capitalization is in excess of $4 trillion, the reality that it felt the need for more capital to deploy toward AI was a major topic of discussion at InsurTech Insights and one more indication of the perceived transformative power of AI, as well as Google’s desire to be an industry leader.

As AI use cases were discussed among conference panels, the distinction between generative AI and agentic AI was frequently emphasized. The evolution of task-oriented generative AI, which facilitates content creation and routine tasks, is being supplemented by agentic AI, which can execute more complex goals, such as “develop a marketing campaign.” Cindy Heismeyer, vice president of strategy and partnerships at Selective Insurance, noted that Selective’s 400 underwriters and hundreds of claims staff are using agentic AI to address their specific operating needs.

The InsurTech community concurred that agentic AI increases speed and accuracy in capturing and sharing data, while lowering the administrative burden. Sessions included numerous examples of projects that normally would take several months being accomplished in just two or three weeks.  

Previously, applying AI to business challenges was considered “innovation.” But the new normal is that AI is seen in the context of organizational transformation. This transformation also brings the challenge of keeping up with the pace of change within the organization.

The rate of adoption is so dramatic that insurance companies worry about the potential to unintentionally “cross the line” because agentic AI operates with high autonomy, particularly in the heavily regulated insurance industry.

One InsurTech panel discussed how insurance is a unique industry because there is “unlimited downsize risk.” Unlike in other industries, where the risk exposure is limited to the amount of capital that is deployed, insurance faces the possibility of a 200% or even 300% loss ratio. As a result, carrier technology leaders have to carefully weigh and balance the trade-offs that speed and efficiency bring without leading to unintended consequences.

In response, insurance companies stressed the importance of keeping the human in the loop. It also begs the question: How can the industry balance thoughtful regulation without chilling the pace of introduction?  

One panel discussion included the issue of the nonstandarized nature of reinsurance submissions and how, as a result, the underwriting process involves a lot of back and forth in gathering additional data.  However, Bill O’Reilly, head of innovation at Greenlight Re, mentioned that the company focused its use of AI on submission ingestion to accelerate the underwriting review process, reducing the time to respond with a quote or to decline to quote.

This is good news for independent insurance agents, as they benefit from the efficiencies that are created. Yet, AI can be overly confident in its output. The collective counsel of the panelists intended to maintain a human in the loop who can apply their judgment before implementing AI’s recommendations.

As generative AI increasingly allows insurance carriers and brokers to solve operational issues, “buy versus build” is a key consideration. Previously, InsurTechs were seen as a plug-and-play solution, rather than taking an individualized approach to problem-solving.

But panelists mentioned that InsurTechs still have a valuable role because they can focus on a specific operational need for the insurance company, while taking more of a consultative role as a partner rather than merely a vendor.

“InsurTech started in the role of a disrupter but now serves as a facilitator in operational areas, such as actuarial science and claims,” said Marc Frenkel, chief financial officer at Coterie Insurance. This theme of bridging the operational gap represents the evolving role that InsurTech can play in helping the insurance industry catch up with the customer experience in other industries.

Another trend panelists highlighted is that InsurTechs were initially funded by equity capital, but now more insurance capital is being used to fund the continued development of AI to deliver business solutions.

Looking ahead, AI is moving beyond improving individual underwriting intelligence and is taking on the challenge of underwriting overall portfolio risk. As existential problems like weather volatility weigh on the industry, the hope is that AI can better address how to approach underwriting risks in the lines of business that are increasingly impacted.

Independent agents can expect that both generative and agentic AI will allow carriers to address the pain points that agents still deal with on a day-to-day basis. At the same time, agencies cannot stay on the sidelines in terms of their AI strategy and tactics.

Agencies are embracing large language models (LLMs) to respond to customer inquiries. And AI is reducing the time and effort to quote new business by scraping policy PDFs to populate proposals and transfer data without reentering it. AI’s potential to free up agency staff for more frequent customer interactions will help agents develop and maintain their customer relationships. 

But the word of caution for agents is that their future competitor is not the agency down the street, but rather an AI assistant that is ready 24/7 to answer consumers’ questions and place their insurance.  

Dave Evans is a senior associate with insurance marketing firm Aartrijk, based in Fairfax, Virginia.