From Our Desk Original analysis · AI
All AI insights →How Actuaries Price AI Liability Coverage When the Loss Triangle Has No Rows
AI liability insurers are pricing coverage with almost no loss history: full credibility needs 1,082 claims, GenAI litigation is up 978% since 2021, and actuaries lean on analogical transfer and scenario loading.
By Shruti · 13 min read · Jul 10
AI Replaces the Weekly PML Run: Cat Accumulation Goes Real-Time
A cohort of primary carriers is replacing weekly batch PML runs with AI-driven accumulation monitoring updated at each bind, not after a one- to three-week lag, a shift with real consequences for RBC cat charges.
By Sarah · 11 min · Jul 10
Chubb's 150-Basis-Point AI Savings Promise Faces Its Q2 2026 Test
Chubb pledged 150 basis points of combined ratio savings from nine AI projects over three to four years; its July 22 Q2 call is the first checkpoint, arriving amid cycle tailwinds that could mask the AI signal.
By Shruti · 12 min · Jul 10
Agentic AI Patents Enter Their System-Architecture Phase
Insurance AI patents are shifting from single-function classifiers to multi-agent orchestration claims, and the harder drafting bar likely widens the 77% patent moat held by State Farm, USAA, and Allstate.
By Shruti · 13 min · Jul 9
NAIC's AI Evaluation Tool Turns Bulletins Into a Scored Exam
The NAIC's AI Systems Evaluation Tool pilot spans 12 states through September 2026, turning Model Bulletin disclosure into a scored exam. We map what Exhibit D demands of a rating engine's data lineage.
By Shruti · 13 min · Jul 9
Duck Creek Buys Send: The Build-vs-Buy Shift in Agentic Underwriting
Duck Creek's July 2026 acquisition of Send folds a $26 billion-GWP underwriting orchestration engine into its Agentic AI Platform, reframing insurance AI's build-vs-buy question around who owns the triage rules.
By Shruti · 12 min · Jul 8
Lemonade Prices Tesla FSD Miles at Half Rate: A Live Test of Exposure-Base Ratemaking
Lemonade prices Tesla FSD miles at half the human-driven rate within one policy, an exposure base shifting mid-trip that forces classification and credibility questions actuaries have mostly argued only in theory.
By Shruti · 11 min · Jul 8
SR 26-2's Banking Model Risk Framework: A Translation Guide for Actuaries
SR 26-2 is banking guidance, not insurance regulation. Its materiality-based model risk framework still gives actuaries a useful template for pricing, reserving, vendor, and AI model governance.
By Shruti · 12 min · Jul 8
LexisNexis Scores Home Risk Across Six Perils: The Validation Problem
LexisNexis says its new AI home risk score shows a 20x claim-frequency lift across six perils, but actuaries need peril-level, cohort-level proof before trusting that number in a rate filing.
By Shruti · 13 min · Jul 7
- AI infrastructure investment is reshaping commercial risk pools: Swiss ReReinsurance News · 10h ago
- Flood risk, rewritten: how AI is filling coverage gapsDigital Insurance · Yesterday
- NN Group Uses In-House AI to Settle Claims in Around Six MinutesInsurtech Insights · 2d ago
- Optalitix Adds Agentic AI to Quote PlatformInsurance Innovation Reporter · Yesterday
- Vertafore Adds AI Submission Agent for MGAsInsurance Innovation Reporter · Yesterday
Carrier Deploys What's actually in production this quarter
Full tracker →Disclosures from earnings calls, 10-Q filings, and vendor press releases. Live indicates production deployment, not pilot.