Companies in the insurance industry are finding that the best way to benefit from AI — and to keep the AI tools they run accurate and honest — is to keep a tight grip on the data that they feed into them.
MSI, for example, is concerned about
Paul Templar, VIPR Solutions co-founder and CEO, echoes this sentiment: "Anyone can build technology — and many people have. What you cannot build quickly is
Meanwhile, AGI is looking at whether vendor relationships deprive it of long-term control over its own data. To this end,
Read Digital Insurance's recent coverage of this trend and others below:
MSI cuts product dev time using Claude, eyes AI drift risks
Millennial Specialty Insurance overhauled its product management processes in response to rising property data complexity and AI proliferation — compressing what once took one to two months of product development into a single day using Anthropic's Claude. Chief Product Officer Andrew Dalton warns that as AI modeling becomes commoditized, MGAs that fail to keep pace face accelerating adverse selection risk. Continuous monitoring for AI drift is non-negotiable: data patterns must be tested regularly, not at launch alone. Proprietary data relationships with distribution partners and vendors — not modeling tools, which will soon be universally accessible — will determine competitive advantage.
Read more:
AGI's in-house AI flags risk gaps, contract flaws for brokers
American Growth Insurance's proprietary AI brokerage platform — live for two months with Maryland-based Heller Kowitz as its first client — automates identification of coverage deficiencies, contract risk transfer failures and cross-sell opportunities. Built with backing from Rockbridge Growth Equity and Atomic, the platform updates every six months, with full impact projected within 24 to 36 months. AGI built the system in-house to maintain direct control over adjustments and avoid third-party dependencies — a model worth examining for carriers weighing build-vs.-buy decisions. The platform's longer-term value hinges on data accumulation; as the client portfolio grows, trend detection and proactive servicing capabilities are expected to sharpen.
Read more:
Specialty insurers face 12-18 month AI governance window
Specialty insurers, MGAs and brokers operating in North American program business markets face a critical 12-18 month window to establish AI governance frameworks and audit-ready oversight as regulatory and rating agency scrutiny intensifies across U.S. states, according to VIPR Solutions CEO Paul Templar. Firms still managing fragmented bordereaux data and legacy workflows are poorly positioned to capitalize on AI-driven efficiency gains. Templar's guidance: domain-specific platforms with deep regulatory logic and proprietary program data will grow more valuable as AI adoption accelerates — not less. The competitive moat in specialty insurance lies not in AI models themselves, but in the trusted, validated data those models require.
Read more:
Allstate opens data-driven claims training campus in Dallas
Allstate has opened a 33,000-square-foot Claims University campus in Dallas, featuring two-story model homes with staged fire, flood and structural damage — built on insights from more than 100 million property claims — alongside 22 damaged vehicles, including EVs and hybrids. The facility updates Allstate's original Tech-Cor training site from the 1970s and is designed to pair virtual instruction with hands-on practice. As homes grow more complex and severe weather drives claims volume, carriers investing in workforce development may gain a measurable edge in adjuster accuracy and customer outcomes. Every classroom supports simultaneous remote participation.
Read more:
Insurers report AI gains in underwriting, claims processing
Second-quarter 2026 earnings calls reveal AI is moving from pilot programs to core operations at major carriers. The Hartford reports underwriting activities completed "in a fraction of the time," with net income up 31% to $1.3 billion for the quarter. Travelers attributed a 0.5-point underlying loss ratio improvement partly to AI-driven submissions processing and straight-through claims handling. Liberty Mutual, posting $2.6 billion net income in the second quarter, is scaling AI enterprise-wide while emphasizing model validation and human accountability. UnitedHealth Group is targeting 80% real-time prior authorization processing by the end of 2027. Executives across all four carriers stressed that AI augments — rather than replaces — human decision-making.
Read more:
AI insurtech funding hits $613M in 2026, but ROI bar rises
AI insurtech startup funding has surged from $69.95 million across seven startups in 2023 to $613.1 million across 14 startups already in 2026 — surpassing all of 2025's $609.4 million total. But the capital comes with tighter expectations. Investors are compressing timelines to profitability and demanding stronger unit economics, a shift experts say distinguishes this cycle from the more permissive insurtech boom of the 2010s. For carriers evaluating vendor partnerships, the pressure to "show the money sooner" means prioritizing startups with clear revenue models over those still building toward scale. Questions about AI ROI are intensifying on the insurer side as well.
Read more:
AI insurtech funding flows to underwriting, claims workflows
Venture funding for AI insurtechs is gravitating toward targeted workflow tools rather than broad platforms, according to Accenture insurance practice CEO advisory lead Kenneth Saldanha. The use cases gaining the most traction — commercial lines submission intake, first notice of loss processing, contact center agent assist and back-office automation — share a common thread: ingesting and summarizing high volumes of documents faster than manual processes allow. Carriers evaluating vendor partnerships should prioritize solutions that integrate directly into existing friction points rather than horizontal AI platforms. Insurance remains attractive to AI investors because significant value remains locked in legacy manual workflows across underwriting, claims and distribution.
Read more:
AI underwriting cuts quote times, builds broker loyalty
Carriers investing in AI-powered underwriting stand to gain a measurable competitive edge in broker retention — but only if modernization extends beyond front-end submission tools. According to Capgemini's 2026 research, underwriters spend the majority of their time on routine data gathering rather than risk evaluation; 61% struggle to improve quote-to-bind conversion rates and 49% struggle to maintain agent satisfaction. Brokers currently navigate seven to 10 separate applications to assemble a single report. Eliminating that internal fragmentation — not just accelerating submissions — is what converts weeks-long turnaround times into minutes and builds durable broker relationships.
Read more:
AI drives policy changes for 37% of insurance researchers
Nearly a third of auto and home insurance customers — 29% — now use AI to research coverage, manage accounts or shop for quotes, according to a J.D. Power analysis of 8,352 responses. The stakes are significant: 37% of those using AI for research changed their policy based on AI-provided information, and 42% who used AI to shop for quotes purchased a new policy. With customers split between insurer-provided tools (34%) and third-party platforms (33%), carriers must optimize both their own AI offerings and monitor how large language models ingest and interpret their policy data and content.
Read more:
Keep humans in the loop as AI reshapes underwriting workflows
As AI prompt engines and agentic AI reshape insurance workflows, human oversight remains non-negotiable — particularly in underwriting. Accenture's global commercial insurance lead warns against ceding decision-making to AI, recommending instead a hybrid model combining conversational AI, system-driven recommendations and underwriter review. Underwriting workbenches are already shifting from task managers toward AI agents that autonomously select systems and respond to unexpected events. The stakes are highest for mid-tier agents: analysts warn that voice-enabled AI interfaces could displace distributors who rely on routine renewals rather than complex problem-solving. As AI adoption accelerates industrywide, firms that delay integration risk falling behind competitors who are using it to move faster on risk decisions.
Read more:
90% of agents reduced business with carriers over workflow issues
Carriers risk losing agent business without investing in automation, according to an Ivans survey of more than 700 independent agents conducted in April–May 2026. Nine in 10 agents have reduced business with a carrier due to workflow friction. Agents' top automation priorities: commercial submission automation (79%), real-time AMS data upload (61%) and claims loss-run delivery (44%). AI appetite varies sharply by line — 94% of commercial lines agents are open to AI tools, while 22% of personal lines agents reject them outright. Automated data prefill leads AI feature requests at 51%, signaling where carriers can gain the most immediate competitive ground with distribution partners.
Read more:
This roundup was created with AI assistance. A Digital Insurance editor reviewed each item before publication.
