As insurers increasingly turn to AI for opportunities to improve insurance affordability, they must also identify and mitigate the new, complex risks it introduces. A first-of-its-kind State of the Tech Policy Brief, developed by Triple-I’s AI Policy Council, explores how one pilot from the National Association of Insurance Commissioners (NAIC) seeks to establish an AI governance framework that can keep pace with AI and machine learning systems as they evolve.
Launched in March 2026, the NAIC pilot aims to provide regulators standards for assessing how insurers leverage AI without replacing existing market conduct or financial analysis. Such assessment centers around four optional “exhibits:”
Exhibit A asks insurers to quantify their AI use across business functions;
Exhibit B evaluates the existing governance structure;
Exhibit C focuses on the mechanisms behind high-risk AI systems, or those that could impact consumers or financial stability; and
Exhibit D captures AI data details.
Participating states include California, Colorado, Connecticut, Florida, Iowa, Louisiana, Maryland, Pennsylvania, Rhode Island, Vermont, Virginia, and Wisconsin. Carriers operating in these states can expect documentation requests and should prepare accordingly, such as by maintaining accurate AI inventories, defining human oversight protocols, and verifying models are working as intended.
Involvement in the pilot does not prohibit states from pursuing their own AI regulations, so insurers should continue monitoring legislative changes at the state level as the NAIC’s program remains active through September. Once complete, the program is up for consideration for formal adoption at the NAIC Fall National Meeting in November 2026.
New AI frontier emerges
While current regulatory discussions focus heavily on predictive and generative AI, the technology has already progressed toward another major wave: agentic AI.
Unlike AI tools that depend on human review to function, agentic models can operate independently and execute their own decisions, the brief explains. As such, governance practices must adapt to manage autonomous systems capable of acting without human interaction, raising the regulatory stakes dramatically.
“As insurers evaluate governance frameworks, key considerations include transparency, accountability, regulatory compliance, and consumer trust,” the brief emphasizes. “The approaches insurers develop today will help inform how AI is deployed, monitored, and governed across the insurance value chain going forward.”
Personal auto insurance premiums represent multiple aspects of the affordability crisis U.S. consumers face today, and a panel discussion at the Brookings Center on Regulation and Markets this week helped define and clarify them.
Panel moderator Aaron Klein, Miriam K. Carliner Chair and senior fellow in Economic Studies at the Brookings Institution, began the discussion by acknowledging “the rising rates of car insurance are part of the broader set of topics that have been given the term ‘affordability.’”
Representing insurers, regulators, and consumers, the panelists included Sean Kevelighan, CEO of Triple-I; Justin Zimmerman, a former commissioner in New Jersey’s Department of Banking and Insurance; and Chuck Bell, programs director for advocacy at Consumer Reports.
All agreed that much of the blame for rising rates can be attributed to external factors such as the costs associated with safer, more technologically sophisticated vehicles, thereby raising the costs to repair and replace them. Inflation has exacerbated these impacts, with auto replacement costs up 28 percent from 2021 to 2025. Over the past 12 months, inflation increased 4.2 percent, thanks in large part to geopolitical risks, supply-chain disruptions, and rising oil prices.
Disagreement surfaced, however, around the degree of insurance-industry responsibility for insurance costs. Consumer Reports’ Chuck Bell suggested the $14 billion insurers issued in rebates to consumers during the COVID-19 pandemic was insufficient, prompting Kevelighan to point out that, “of all the refunds being given, you saw the most coming out of the insurance business and community.” Zimmerman noted that many states also froze insurers’ ability to raise rates during the pandemic, leading to some post-pandemic “rate catch-up.”
Rampant legal system abuse helps fuel the strain. While derided as a concept by some, Kevelighan cited analysis from Triple-I and the Casualty Actuarial Society that indicates excessive litigation added up to $281.2 billion in increased liability insurance losses from 2015 to 2024 – a finding that economic inflation alone cannot explain. A separate Triple-I report on civil case filings indicated roughly one-third of increasing inflation in auto liability losses stemmed from these legal trends.
Kevelighan also highlighted the $380 million spent by third-party litigation funders (TPLF) on online advertising last year, according to a study from the National Insurance Crime Bureau and 4WARN. Now “a global multi-billion-dollar asset class,” TPLF has become a target for reform in a growing number of states, notably New York.
New York affordability struggles
Wiping out billions of dollars in U.S. economic activity annually, legal system abuse costs New York residents 427,794 jobs and $7,027 per household per year, contributing to the fourth-highest auto insurance expenditures in the nation, Triple-I estimates. Moreover, the state’s average personal auto injury claim is $46,726, at more than twice the national average.
Building on legislation to tackle TPLF, New York lawmakers recently passed a package of auto insurance reform bills to disincentivize legal system abuse and fraud, one of which will introduce a $100,000 cap on noneconomic damages for drivers who were at fault, uninsured, or impaired at the time of an accident. Comparative negligence rules were also updated to ensure costs cannot be shifted away from the motorists responsible for an accident.
Kaitlin Asrow, New York State’s acting superintendent for the Department of Financial Services, told Klein in an interview before the panel, “Over the last five years, suspicious fraud reports for just no-fault auto increased 80 percent.” She added that “staged accidents were up 34 percent” in New York City alone during the same period.
While further reforms are needed to address the Empire State’s high insurance costs, Kevelighan pointed out that similar efforts in Florida have begun to drive substantial premium reductions and renewed private market competition.
Modifying behavior for risk reduction
Though many influences on insurance costs are structural, Kevelighan emphasized “a lot of this comes down to our behaviors and how we’re driving and living.” As such, insurance must shift from “a once or twice a year type of transaction” to “an open and ongoing conversation” between insurers and their customers.
Part of that conversation revolves around distracted driving, which jumped significantly after the onset of the COVID-19 pandemic and remains at elevated levels. As measured by a recent Nationwide survey, seven in ten commercial drivers have reported experiencing increased distraction as well as reckless driving from other drivers, at a 10-point increase from 2025.
Nationwide also found that commercial auto loss ratios drop by at least 30 percent when policyholders use telematics, a technology that monitors mileage, braking and acceleration, and other driving patterns to provide real-time feedback that can adjust unsafe behavior. In addition, built-in accident-avoidance systems are reducing rear-end collisions by 40 to 50 percent.
Noting telematics research is still in its early stages, Kevelighan said the “interaction and exchange” of risk information between insurers and policyholders “is where the industry is going to start shifting from just detecting and repairing after a catastrophe to predicting and preventing.”
“We’ve got to make sure we’re balancing out what it is that we’re doing to reduce our risk, because that’s the real driver,” Kevelighan explained. “When we reduce the risk, we can reduce the cost.”
At least 10 percent of property/casualty insurance claims may be fraudulent, adding up to billions of dollars in fraudulent insurance claims every year, the National Insurance Crime Bureau estimates. While legislative reforms are necessary to combat fraud and legal system abuse, many insurers are turning to artificial intelligence and machine learning models to help mitigate the risks in the near term.
Often trained on years of data, AI-powered tools can flag suspicious claims or those likely to litigate based on early risk indicators, such as attorneys or firms frequently linked to inflated claims. Some systems leverage litigation propensity scoring to predict a claim’s likelihood to escalate from the first notice of loss, providing real-time risk ratings throughout the claim cycle that better enable adjustors to prioritize high-risk claims.
By synthesizing historical data and automating the review process, such systems can give insurers the chance to intervene or settle before claims escalate. Research indicates these early-warning models can identify potentially fraudulent claims within two weeks after submission, far outpacing traditional detection methods that involve manually sifting through large, complex volumes of data.
Delivering measurable outcomes
Early intervention can facilitate fairer settlement outcomes and protect insurers and policyholders from unnecessary legal costs that keep upward pressure on premium rates for all consumers. Deloitte analysis suggests applying AI across the claims cycle could save insurers between $80 billion and $160 billion by 2032 through fraudulent claim reduction, translating to billions in savings for their insureds.
Data libraries that pool litigation pattern and claims data from insurers and companies from other industries can also improve AI model insights. Rather than leaving organizations to rely exclusively on their own internal data, these cross-industry approaches can expand base datasets and prediction accuracy, allowing insurers to keep pace with emerging risks.
To grasp insurance executive readiness for AI adoption, Deloitte conducted a separate 2025 survey that found those who reported successful AI initiatives cited “close collaboration across business, tech, data, and talent functions” as the greatest contributing factor. Among all respondents, 35 percent ranked fraud detection as one of their top five areas for implementing generative AI.
It’s no wonder why: As tools to mitigate insurance fraud have evolved, so too have the tools available to bad actors aiming to defraud the claims process. Plaintiffs’ attorneys themselves are seizing on the opportunity, with research from Suite 200 Solutions indicating “almost all litigation financing funds now use AI to identify cases likely to win,” down to “case type, venue, judge, plaintiff attorney, and other factors.”
Tactics to mislead consumers into escalating claims are also increasingly AI-driven, including automated “robocalls” and text messages that solicit receivers to file lawsuits. Another study from the National Insurance Crime Bureau and 4WARN observed that third-party litigation funders (TPLF) are using AI-generated content to scale volume and prolong settlements, as part of a larger digital marketing campaign that attracts 27.8 million clicks to TPLF-hosted websites every month.
Traditional claims review methods fail to capture these modern digital risks, necessitating AI-powered detection and mitigation to stay ahead of new threats.
Industry collaboration is key
Yet, as companies scale their AI investments, human oversight must remain at the forefront, as should maintaining a traceable actuarial record behind every model. Beyond safeguarding model accuracy, AI data understanding and preparation are crucial to ensuring carriers comply with insurance regulations and can uphold consumer trust. Attracting talent that balances actuarial knowledge with AI expertise will be pivotal to successful model deployment.
To address these challenges, Triple-I and The Institutes RiskStream Collaborative – like Triple-I, an affiliate of The Institutes – recently established two coordinating councils to develop shared AI capabilities and research and governance standards across the insurance sector.
Led by RiskStream, the AI Solutions Council brings together insurers, tech firms, and other stakeholders to prioritize multiparty AI use cases and generate AI solutions across the insurance value chain. Alongside Triple-I’s AI Policy Council, which focuses on regulatory and governance frameworks for AI use in insurance, these bodies give insurers a structured way to collaborate on AI solutions and best practices rather than leaving each carrier to build capabilities in isolation.
Despite a 34 percent decline in cyber insurance claim frequency for large U.S. businesses in 2025, average claim severity doubled to more than $4.4 million, according to Chubb’s 2026 Cyber Claims Report. Though AI-driven detection systems helped stabilize claim frequency across several global markets, advanced cyberattacks – alongside liability litigation challenges – ranked among the top cost drivers.
Drawing on historical claims data, the report explained how bad actors have begun leveraging AI for increasingly sophisticated attacks capable of “compromising multiple systems in a matter of minutes,” including large-scale incidents that involve minimal human oversight. Data-breach claims alone exceeded a historic $10.2 million in the U.S., propelled in part by the outsized impact of individual ransomware encounters.
Becoming faster and more difficult to detect, ransomware incidents can propagate across multiple businesses along a supply chain with just one attack, especially as markets become more globally interconnected. One such event in the U.K. led to roughly $568 million in losses for the targeted company but a $1.4 billion loss for the entire supply chain as manufacturing “halted for five weeks across sites in the U.K., Slovakia, Brazil, and China.” Over 5,000 U.K organizations in total were affected, Chubb said.
Consequences of cyber incidents extend beyond these losses, the report noted, as incidents increasingly escalate to legal action, often within days and “irrespective of the size of the entity or any controls perceived to be lacking.” Federal legislation enacted in 1988 to protect physical video privacy has helped lead the trend, as plaintiff attorneys continue to reinterpret the law to apply to modern streaming and social media platforms.
Similar applications of a 1967 statute in California – originally intended to prevent wiretapping – now target businesses that use website technologies such as cookies and tracking pixels, generating thousands of lawsuits in recent years. A bill that would remove these prohibitions for businesses has garnered strong bipartisan support, though faces an uncertain future after stalling in the state legislature last year.
“At a time when affordability is already one of California’s greatest challenges, these lawsuits are quietly making life more expensive for everyone,” wrote Scott Miller, president and CEO of the Fresno Chamber of Commerce, for The Fresno Bee. “[SB 690] would restore balance, reduce abusive litigation, and allow small businesses to focus on serving their customers, not defending against opportunistic lawsuits.”
A “growing body of privacy laws” are further “imposing complex, layered obligations for companies that store and/or transfer personal data,” Chubb reported, highlighting new laws in Indiana and Kentucky aimed at implementing stricter opt-in mechanisms for personal information. Companies may struggle to navigate these emerging regulations as privacy and cyber risks continue to evolve, creating compliance concerns and potentially exacerbating losses and broader supply-chain disruptions when cyberattacks occur.
Investing in threat detection, AI governance, and employee cybersecurity education are among the many ways organizations can boost their cyber resilience. A separate Chubb survey also suggests interest in cyber insurance to mitigate these risks is rising. Leaders across lower, core, and upper middle market segments identified cybersecurity and advancing technology as their chief risk concerns, with 47 percent of respondents indicating they were considering adding or increasing cyber coverage.
Given the growing ubiquity of artificial intelligence, its practical applications may seem self-evident. But for actuaries – whose work hinges on rigorous modeling and explainable risk assessment – translating AI-driven insights into analysis may pose as many challenges as solutions. A well-defined balance between technological capability and ongoing actuarial judgement is essential to navigating this shift.
“The challenge is not that there’s too much data – it’s having an awareness of what you’re looking for and then finding it,” said Dr. Michel Léonard, Triple-I chief economist and data scientist, in a recent interview for the Casualty Actuarial Society (CAS) Institute’s Almost Nowhere podcast. “If you look at all the data and it’s not focused and translated, the signal is not going to be what you need.”
Noting that many AI models train on varied language sources, Léonard stressed that data understanding and preparation are crucial to confronting the “black box,” or opacity surrounding the training and internal decision-making processes of complex algorithms. To integrate AI into risk assessment, carriers will need to demonstrate the mechanisms and actuarial record behind the models they deploy, especially for regulators and the broader public.
Though dynamic wildfire models, for instance, “very clearly show that the risk is more frequent and severe,” ongoing transparency around how these models work will be key to building “a bridge between regulators and the industry,” Léonard said.
While such models have facilitated greater access to granular, real-time data, critical information gaps continue to impede effective risk forecasting, especially following the 2025 federal government shutdown. Beyond being the longest federal closure in U.S. history, the shutdown also delayed or left permanent gaps in crucial survey data on employment, inflation, and other economic indicators, fueling more uncertainty for decision makers heading into 2026.
“Because of this uncertainty, we’re forecasting on the trend, which means that we cannot stress test or include validation for those stress tests,” Léonard said. “The lack of data on the U.S. economy is the main challenge for us right now.”
Current tariff policies – especially those targeting materials used in repairing and replacing property after insured events – add to the ambiguity. Though insurers appeared to avoid “the worst-case scenario” of COVID-19 levels of market instability last year, strategic stockpiling of imported goods to circumvent later post-tariff prices may have obscured their full impact, Léonard explained.
A pending Supreme Court ruling will determine the future of these policies, leaving global markets and consumers braced for potentially rising costs. Yet Léonard emphasized the insurance industry’s resilience in managing such “extreme, black swan-type events,” pointing out “that’s why we have a reasonable and adequate policyholder surplus” and other assets to ensure consumers remain protected.
Amid federal funding and staffing cuts to major science agencies last year, various nonprofit organizations stepped up to maintain their essential climate and weather research. Such risks may become increasingly difficult to predict and prevent, however, as key agencies, such as the National Center for Atmospheric Research (NCAR), remain targets for disinvestment or termination.
Private sector takes charge
In the spring of 2025, the federal administration attempted to rescind tens of billions of dollars in research and hazard mitigation grants, leaving many programs – like FEMA’s Building Resilient Infrastructure and Communities (BRIC) program – in legal limbo as legislators continue to debate their futures. Alongside funding delays and cancellations, mass firings led to the shuttering of several climate and weather information resources – until private associations and researchers mobilized to revive them.
Former NOAA staffers, for instance, regrouped to rescue the organization’s climate.gov website, which attracted nearly one million visitors per month – including teachers, policymakers, and media outlets – before being dismantled last June. Under a new domain, the site will both restore deleted information and resume tracking and explaining the effects of climate risk to public audiences, relying exclusively on nonprofit funding, according to project director Rebecca Lindsey in an interview with NPR.
Similarly, nonprofit Climate Central recently released its first billion-dollar weather and climate disaster report since assuming responsibility for that dataset, which former NOAA climatologist Adam Smith continues to oversee. Beyond rebuilding NOAA’s database, the organization aims to expand upon it in the coming years to track smaller catastrophes, providing insurers and other stakeholders more reliable information to understand individual disasters.
An initiative spearheaded by the American Geophysical Union (AGU) and the American Meteorological Society (AMS) is now aiming to help fill research gaps left by the elimination of the National Climate Assessment (NCA), a series of congressionally mandated reports published since 2000 to inform climate risk mitigation strategies for municipalities and businesses. Though not intended to replace NCA, the new data collection “provides a critical pathway for a wide range of researchers to come together and provide the science needed” to “ensure our communities, our neighbors, our children are all protected and prepared,” said AGU president Brandon Jones.
Grassroots efforts to archive federal climate databases and tools before they disappear have also gained traction around the globe to ensure these resources remain publicly available. The nonprofit Open Environmental Data Project, for example, saved a now-deleted tool to identify communities disproportionately impacted by climate and weather risks through its Public Environmental Data Project.
Crucial agencies under scrutiny
While the latest government spending package has largely spared science funding from further reductions, the Trump administration had proposed cuts amounting to a 21 percent drop from fiscal 2025 levels. Other agencies face potential dissolution, particularly NCAR – widely considered the largest federal climate research program in the U.S.
Managed by the University Corporation for Atmospheric Research (UCAR) in collaboration with the National Science Foundation (NSF), NCAR houses advanced computing and modeling systems to support weather forecasts, mitigation planning, flood mapping, and other datasets needed across the transportation, engineering, utility, and risk and insurance industries.
Describing NCAR’s research as critical to “protecting lives and property, supporting the economy, and strengthening national security,” UCAR president Antonio Busalacchi said in a statement that “any plans to dismantle NSF NCAR would set back our nation’s ability to predict, prepare for, and respond to severe weather and other natural disasters.”
“NCAR datasets have been vital in improving our understanding of the atmosphere and ocean,” said Phil Klotzbach, lead author of Colorado State University’s seasonal hurricane forecasts and Triple-I Non-Resident Scholar. “These tools have been critical input to CSU’s seasonal hurricane forecasts for over 25 years.”
NCAR’s pending fate coincides with a recent study from the University of Florida that suggests the budget cuts in part reflect pervasive distrust in scientific institutions, necessitating stronger efforts to communicate the value of scientific work to the public. But as more independent groups take on the responsibilities once affiliated with federal organizations, building public relationships may prove even more challenging, posing uncertain implications for the future of climate and weather data as a whole.
Deals exceeding $1 billion drove insurance industry mergers and acquisitions (M&A) in 2025, aligning with a surge of artificial intelligence-based megadeals in the broader M&A market, according to PwC’s U.S. Deals 2026 Outlook. While total M&A deal value rose to $1.6 trillion through November 30 – the second-highest value ever recorded – the insurance sector remained consistent, though ongoing economic uncertainty could challenge this stability.
Upward trends emerge
Owing 93 percent of its value to megadeals, the insurance industry’s deal volume totaled $31.8 billion during the second half of 2025, compared to $30 billion during the previous six months. Heading into 2026, PwC projects that both improved loss ratios and rising pressure on property and casualty rates will facilitate further deals, as the industry’s performance can attract investors while also leading carriers to seek growth through more acquisitions.
“In coming months, expect interest rate developments and an industry-wide search for growth to strongly influence insurance deals activity,” said Mark Friedman, PwC U.S. insurance deals leader.
Conversely, total M&A market value increased by 45 percent from 2024, with more than 20 percent of its 74 deals valued at $5 billion or more relating to A.I. Such market activity suggests “A.I. is significantly accelerating software and consumer goods development” by boosting portfolio strategies, operational efficiencies, and other gains across various industries, the report notes.
Ramzi Ramsey, a senior managing director at Blackstone Growth, emphasized the increasing role of A.I. as a core driver of M&A growth, arguing that “companies who are viewed to benefit from AI tailwinds are seeing outsized multiples and deal activities,” whereas “companies where AI is viewed to be a detractor, or if the AI impact is cloudy, may have no bid.”
Middle-market lags
Though still the third-largest in the past decade, overall M&A market volume rose by only 2 percent from 2024, with middle-market deals between $100 million to $1 billion slumping to their lowest volume since at least 2013. Tariff policy shifts largely fueled these figures, reflecting greater caution among dealmakers – particularly smaller businesses – as supply-chain risks became more unpredictable.
While sudden policy changes and recent trade disputes could persist into the new year, PwC’s report found that “interest rate cuts this year have already helped mid-tier corporates, and further Federal Reserve Board action in 2026 could go a long way in relieving pressure on them,” especially as current tariff policies continue to stabilize.
In combination with declining interest rates, “products that can withstand declining consumer performance and confidence” should help the insurance sector remain active in 2026, the report adds, pointing to A.I. investments as essential to maintaining “solid ground” in the M&A market.
The homeowners insurance market is catching up to its cost drivers while still facing challenges to affordability and availability. Rates continue to climb as natural disasters intensify and replacement costs rise, but industry analysts expect meaningful improvement over the next two years. A new Triple-I Issues Brief provides a snapshot of the market’s performance and outlook, and discusses how some trends are shaping its future.
The latest results for the product line have helped narrow the anticipated 2025 gap between the performance of the personal and commercial lines. Despite a volatile start to 2025 driven largely by January’s destructive Los Angeles wildfires, homeowners insurance is still headed for double-digit net written premium growth this year.
With nearly half of all homes in the United States at risk of “severe or extreme” damage from weather related events, climate risk looms large. In January 2025, the U.S. Department of the Treasury released “Analyses of U.S. Homeowners Insurance Markets, 2018-2022: Climate-Related Risks and Other Factors.“ a report based on the most comprehensive and granular snapshot of the homeowners insurance market to date. The agency found that climate risk is making it more costly for insurers to operate, as insurers’ costs in 2018-2022 were higher in areas with the highest expected losses from climate-related perils. The paid loss ratio, which reflects how much insurers paid for claims relative to the premiums they collected, was highest in the highest-risk ZIP Codes.
In 2025, the U.S. experienced its first hurricane season without a single landfall in a decade. However, the Triple-I issue brief explains, while 2025 economic losses from natural catastrophes are running below recent averages, other perils — such as severe convective storms, wildfires, and flash flooding — are becoming formidable sources of insurer loss. These increasingly frequent moderate disasters are challenging traditional catastrophe models built around infrequent peak perils, such as major hurricanes.
At the same time, soaring replacement costs have become the new normal for the homeowners market. Repair and rebuilding expenses have jumped nearly 30 percent over the past five years, fueled by inflation, supply-chain disruptions, rising construction material prices, labor shortages, and, more recently, new federal tariffs. Although the full impact of these tariffs has been milder than expected so far, the worst effects may simply be deferred until 2026 as inventories decline. Rising replacement costs translate directly into higher claim payouts, placing additional pressure on insurers and, ultimately, policyholders.
Beyond tariffs, other political and regulatory shifts are adding a new uncertainty as federal disinvestment in climate monitoring and mitigation may impede the insurance industry’s ability to accurately price risk, predict future losses, and, ultimately, provide affordable coverage. Meanwhile, several states grapple with balancing affordability with the stability and solvency of their insurance markets.
Insurance pricing must reflect these increased risks to maintain policyholder surplus, the funds regulators require insurers to keep on hand to pay claims. If premium rates fail to reflect increased costs, insurers may rapidly drain their policyholder surplus. This issue brief discusses how emerging technologies, such as advanced predictive analytics, aerial imagery, and smart-home sensors, could pave the way for more accurate pricing, faster claims processing, and improved risk prevention.
An Insurance Research Council (IRC) study indicates that homeowners familiar with some AI-driven insurance solutions view pricing using those technologies as fairer and express fewer concerns overall. These tools may play a critical role in bolstering affordability, rebuilding trust, and strengthening the resilience of the homeowners’ insurance sector amid escalating climate and economic pressures.
The issue brief’s list of factors and trends impacting the homeowners’ market isn’t intended to be exhaustive. Accordingly, future briefs on homeowners (or property lines in general) may highlight other pertinent topics, such as the link between insurance premiums and property prices. While home values in high-risk areas can often be diminished by rising premiums, higher home values can generally mean higher replacement costs, and consequently, lead to higher premiums. As of early 2025,home prices are up 60 percent nationwide since 2019 and still rising by 3.9 percent YoY, according to the Joint Center for Housing Studies at Harvard University. The Harvard report cites Freddie Mac data indicating home insurance premiums jumped 57 percent from 2019 to 2024.
Global insurtech funding hit $1.01 billion in the third quarter of 2025, marking three consecutive years of average quarterly funding near $1.1 billion, according to a recent Gallagher Re report. Raising $751.72 million, A.I.-based insurtechs accounted for almost 75 percent of all funding across 49 deals, raising $751.72 million, largely attributed to the commercial insurance industry’s evolving risk profile.
Fewer, bigger deals
Though down 7.3 percent from the previous quarter, third-quarter results reflected less quarterly volatility compared to the preceding three-year period, which fluctuated with greater uncertainty around a higher average of $2.7 billion. Deal count also dropped to 76 – the lowest total count since the second quarter of 2020 – as average deal size rose from $12.83 million in the second quarter of 2025 to $15.70 million.
Property/casualty insurers financed eight of the quarter’s 10 largest deals, propelling the industry’s total third-quarter funding to $690.28 million – a 90.5 percent increase from its seven-year low in the prior quarter.
Reinsurers led another dramatic market shift by backing a record sector high of 51 tech investments, suggesting “the appetite for pure venture risk is alive and well” even as investors place fewer “massive, high-risk bets” in favor of “a more balanced approach,” said Andrew Johnston, global head of InsurTech at Gallagher Re.
A.I. takes center stage
With over 25 percent of commercial lines now sold through digital channels, the report outlines how insurers can meet the demands of changing customer expectations and business practices – particularly the digital collection and storage of customer data – by leveraging A.I.
By improving data extraction, pattern identification, and fraud detection, A.I. tools streamline routine decision-making while freeing up underwriters’ capacity to build client relationships and assess complicated, high-value risks, the study found. On a concrete level, such efficiency gains include high-resolution aerial imagery to quickly verify property damages, dashcams to monitor real-time driving behaviors, and wearable IoT solutions to prevent workplace injuries, demonstrating the utility of A.I.-powered insurtechs across commercial lines.
Effective integration, however, transcends simple adoption. Freddie Scarratt, Gallagher Re’s global deputy head of InsurTech, emphasized the enduring challenges of applying A.I. to legacy administration systems and of an emerging talent gap to bridge data and A.I. literacy expertise with traditional underwriting.
Business leaders expressed similar concerns in a Gallagher Re survey released earlier this year, highlighting a skills shortage and pervasive ethical implications as barriers to A.I. adoption.
Underscoring the industry’s goal to “supercharge” underwriting judgement rather than replace it, Scarratt concluded that “the most successful (re)insurers of the future will be those that combine their expertise in relationship management, complex deal structuring, and cycle management with the speed, scale, and analytical power of A.I.”
I wrapped up my first-ever Climate Week NYC last week at ClimateTech Connect. After their two-day April event in Washington, D.C., I could hardly miss this special half-day update when it was so close to home.
Fifty-plus attendees crammed a room near Grand Central Station, and I immediately spotted familiar faces and had the opportunity to meet with a mix of industry veterans and relative newbies spanning all insurance disciplines, from underwriting and claims to the cutting edge of modeling and artificial intelligence. Top insurance thought leaders and influencers were there to speak on climate-related issues of pressing interest to my industry and everyone it serves. The panel themes and the panelist themselves made it clear from the start that a blog post was not going to do the event justice.
The first panel – Pioneers Shaping the Future of Climate Resilience – was moderated by Francis Bouchard, managing director for climate at Marsh McLennan, whose bona fides include senior positions with Zurich Insurance and the Reinsurance Association of America. Francis moderated a no-holds-barred panel of young insurance leaders: Angela Grant at Palomar, Michael Gulla of Adaptive Insurance, and Valkyrie Holmes of Faura. The energy and expertise of these panelists left me feeling that the industry – in the face of myriad challenges – is being put into good hands.
The next discussion was moderated by Jerry Theodorou, a director at the R Street Institute whose professional background includes roles at Conning, AIG, and Chubb. It featured Dan Kaniewski, managing director and U.S. public sector lead for Marsh McLennan and a former FEMA deputy administrator, and Raghuveer Vinukollu, head of climate insights and advisory for Munich Re. The depth and timeliness of these three experts’ insights made for an engaging and thought-provoking session.
The third panel was both engaging and accessible – a bit surprising to me, given that it consisted entirely of PhDs. Steve Weinstein, CEO of Mangrove Property Insurance led a discussion among Joanna Syroka of Fermat Capital Management, Catherine Ansell of JPMorgan Chase, and M. Cameron Rencurrel at Mercury Insurance on not only “Why Science Needs to Be in the Boardroom,” but HOW young scientists can find their way there and decide IF that’s where they want to be.
Between these panels were presentations from representatives of several insurtechs who shared their data-driven solutions focused on understanding and addressing climate-related panels. All this in a period of about three hours (not including the networking reception afterward). Despite all the information shared, the event did not feel at all rushed.
If you weren’t able to make it and are feeling a bit left out, don’t fret! ClimateTech Connect 2026 will be held in Washington, D.C., on April 8 and 9, 2026.
