Insights on AI, insurance and decisions.
Essays, use cases and technical notes from the WIR team on how Artificial Intelligence is redesigning insurance operations.
São Paulo State Civil Defense: What It Is, What It Does, and Why P&C Insurers Should Care
A plain-language guide to the Civil Defense of the State of São Paulo, why it exists, how it is organized today, why your phone buzzes with its alerts, and what those warnings mean for underwriting and claims in Brazilian P&C insurance.
Nicholas Weiser: 27 years in the market before founding WIR
WIR's CEO and co-founder on the career that started at a post office branch at 15, ran through Aon, Gallagher, JLT and Lockton, built and sold his own broker, co-founded EZZE Seguros, and why he turned into a technology company.
José Carlos de Paula: four decades across banking, insurance and healthcare
WIR's CSO and co-founder on the career that took him from bank trainee to brokerage CEO to Bain Capital healthcare executive, and why he decided to build from zero now.
What Is an Algorithmic MGA? Model, Stack, Proof
An algorithmic MGA is a managing general agent that underwrites through code instead of through a queue of human referrals. It holds delegated authority from a carrier, prices and binds risk automatically inside an agreed appetite, and reports back to its capacity provider as data rather than as a monthly spreadsheet.
What Is FNOL? First Notice of Loss, Step by Step
FNOL stands for First Notice of Loss, the first report an insurer receives that a loss has happened. It is the moment a claim file opens, and it sets the cost, the speed, and the customer experience of everything that follows.
What Is an Underwriting Intelligence Layer?
An underwriting intelligence layer sits between the submission and the core policy system, turning unstructured inputs into decisions the insurer can defend. It does not hold the policy, the premium, or the ledger. It exists because core systems were built to record decisions, not to make them.
Parametric vs Indemnity Insurance: The Real Difference
Parametric insurance pays a fixed amount when a measurable trigger is breached. Indemnity insurance pays what the loss actually cost, after an adjuster verifies it. That difference changes the speed of the payout, the paperwork, the price, and the risk the buyer keeps.
Shadow Mode Underwriting: Test AI Without Risking a Book
Shadow mode underwriting is running an AI model on live submissions without letting it decide anything. The model sees the same risks the underwriters see, its answer is logged and compared rather than acted on, and the record becomes the evidence a regulator or a reinsurer will actually accept.
Pricing Engine vs Underwriting Decision Platform
A pricing engine answers what this risk should cost. An underwriting decision platform answers whether to write it at all, and on what terms. The two get sold as if they were the same product, and buying one when you needed the other is an expensive mistake.
Is AI Underwriting High-Risk? EU, US, Brazil 2026
Yes, AI underwriting is high-risk under the EU AI Act when used for risk assessment or pricing in life and health insurance (Annex III). As of 2026, that classification brings obligations around data governance, transparency, human oversight, and record-keeping for those lines. The United States and Brazil regulate the same underwriting activity through different instruments, but the shared thread across all three regions is auditability, explainability, and human oversight.
Build vs Buy AI Underwriting: A Decision Guide
Most insurers should buy a thin external AI layer for underwriting and build in-house only for capabilities that are a genuine competitive edge. As of 2026, the real question is not build versus buy a core system, it is build in-house versus buy an external AI layer that automates underwriting, submission intake, quoting, and decisioning without replacing the core. WIR Innovation is one such external layer.
How AI Matches Submissions to Underwriting Appetite
AI matches submissions to underwriting appetite by encoding the insurer's appetite guide as machine-readable rules, then scoring each submission as in-appetite or out-of-appetite. As of 2026, an external AI layer runs this pre-check automatically on every submission, documents why each risk fits or does not, and routes it accordingly, while the underwriter keeps the final call. WIR Innovation delivers this on top of the core system, never in place of it.
How AI Processes a Statement of Values (SOV)
Yes, AI can process a statement of values (SOV) for property underwriting by extracting each location's values and COPE data into structured, underwriting-ready fields. As of 2026, WIR Innovation, an external AI layer for insurers and MGAs, structures SOV schedules automatically and feeds the clean data into the core policy system rather than replacing it, so underwriters stop keying spreadsheets row by row.
How to Run an AI Underwriting Pilot That Works
You pilot AI underwriting without disrupting your core system by running the AI as an external layer on top of it. WIR Innovation is an external AI layer for insurers and MGAs that automates underwriting, submission intake, quoting, and decisioning without replacing the core system. As of 2026 it reads from your existing policy-admin system and writes nothing back until an underwriter approves, so the pilot is low risk to start and stop.
The ROI of AI in Insurance Underwriting Explained
The ROI of AI in insurance underwriting is the value created by four operational levers, measured against the cost to deploy and run the AI. Those levers are capacity reclaimed, leakage recovered, faster quote to bind, and better risk selection. As of 2026, WIR Innovation frames this as a formula you populate with your own baseline, never a headline percentage.
Underwriting Copilot in Insurance, Explained
An underwriting copilot is an assistive AI layer that helps an underwriter work faster inside the tools they already use. It summarizes submissions, drafts referrals, and flags missing data, but it does not decide, and it never replaces the core policy administration system. As of 2026 the term is everywhere in insurance AI, and it is frequently confused with a workbench or an autonomous agent.
What Is Flow Underwriting (Algorithmic)?
Flow underwriting is a high-volume, low-touch approach in which standard submissions are decided automatically by rules and models, with underwriters handling only the exceptions. It differs from traditional underwriting, where an underwriter reviews each risk case by case. As of 2026, AI is what makes flow viable at portfolio scale: it structures every submission and enforces the insurer's appetite so the routine cases clear without manual work.
Delegated Authority Technology for MGAs in 2026
In 2026, delegated authority is as much a technology question as an underwriting one. MGAs carry a large and growing share of specialty premium, but that authority concentrates operational load in three document-heavy tasks: submission intake, bordereaux processing, and straight-through processing. The right fix is not a new core system but an external AI layer that sits on top of the systems an MGA and its carrier already run, structures the inbound data, and preserves both the systems of record and the human accountability that delegated authority depends on. WIR is that external AI layer and never replaces the core.
How AI Analyzes Loss Run Reports for Underwriting
AI analyzes a loss run report by reading the document, extracting every claim into structured data, normalizing it across carriers, and computing the metrics an underwriter needs, loss ratio, frequency, severity, and open reserves. In WIR's architecture this runs in an external AI layer on top of your core, never replacing it.
How AI Automates FNOL (First Notice of Loss)
The first notice of loss (FNOL) is where a claim enters the insurer's world, and it almost never arrives in the structured form a claims system expects. AI automates that first step: it captures the report across channels, reads and structures the unstructured input, validates it against the policy, and routes the claim, then hands a clean, auditable record to the core. An external AI layer does this on top of the claims platform an insurer already runs, such as Guidewire, never in its place, and leaves coverage and settlement decisions with the insurer.
How AI Automates Insurance Claims Processing
AI automates insurance claims processing by reading each notice of loss as it arrives, extracting and validating the data, scoring the claim against the insurer's own rules, and routing it to settlement, decline, or a human reviewer, with a complete audit trail. WIR does this as an external AI layer on top of the systems the insurer already runs, never in their place.
How AI Powers Embedded Insurance Decisioning
AI powers embedded insurance decisioning by running eligibility, pricing signals, and bind as a single real-time API call at the point of sale, layered over the insurer's existing core system instead of replacing it. When a customer adds cover at checkout, an external AI layer reads the transaction context, checks appetite and rules, returns a price signal, and issues a bind decision in a fraction of a second, while the core policy administration system stays the system of record. The result is instant cover inside a partner's flow, with the insurer keeping full control of risk, rating, and compliance.
How to Automate Bordereaux Processing with AI
To automate bordereaux processing with AI, you place an external AI layer in front of your existing systems that reads each incoming coverholder file, maps its columns to a common schema, validates and reconciles the data, and writes clean records to your system of record. High confidence rows flow straight through, and only exceptions reach a person. The core stays exactly as it is, because WIR sits on top of it rather than replacing it.
How to Extract Data from ACORD Forms with AI
To extract data from ACORD forms with AI, you put a Machine Learning reading layer on top of the systems you already run. It parses each ACORD form wherever it arrives, an emailed PDF, a scanned certificate, or a portal upload, identifies the fields an underwriter needs by meaning rather than by fixed position, and returns clean, validated, structured data. It reads scanned and photographed forms reliably and also attempts handwritten and free-text fields, routing anything it is unsure about to a human reviewer through a per-field confidence score. That human in the loop, not a claim that handwriting is solved, is what keeps the output accurate and auditable.
What Is Parametric Insurance and How AI Enables Faster Payouts
Parametric insurance pays a fixed, pre-agreed amount the moment a measurable event crosses a defined threshold (a wind speed, a rainfall total, an earthquake magnitude), instead of reimbursing a documented loss after a claim is filed and adjusted. Because the decision rests on an independent data point rather than a loss investigation, settlement can happen in days rather than the weeks to months an indemnity claim often takes. AI does not change what a parametric policy pays; it strengthens the data and decisioning around the trigger, ingesting and validating the source data, monitoring the parameter continuously, and routing a clean, auditable pay-or-hold decision.
AI Fraud Detection at Underwriting vs Claims
AI fraud detection operates at two points in the insurance lifecycle. At underwriting, it scores fraud signals while a risk is being quoted, so bad risk is priced correctly or declined before it enters the book. At claims, it flags suspicious losses after they are reported, once the exposure already exists. An external AI layer can run these checks on top of the core an insurer already operates, never in its place, and return an explainable score with a full audit trail.
How AI Is Used in Commercial and Specialty Insurance Underwriting
In commercial and specialty insurance, AI is used to automate the work that surrounds an underwriting decision, not to make the decision itself. It reads and structures messy submissions, checks them against appetite, enriches them with third-party data, and scores complex risks so underwriters spend their time on judgment instead of re-keying. The most durable deployments run as an external layer on top of the core policy administration system, not as a replacement for it.
How to Audit AI Underwriting Decisions for Compliance
To audit AI underwriting decisions for compliance, hold every automated quote, decline, or referral to five tests: is it logged with a complete record, explained in plain language, open to human review and override, watched for drift and bias over time, and mapped to the specific rule it must satisfy? Built that way, one evidence base answers a SUSEP conduct review, an LGPD request to review an automated decision, and the high-risk duties in the EU AI Act, all while the AI stays an external layer that never replaces the insurer's core system.
How to Reduce Quote Turnaround Time in Insurance With AI
Insurers reduce quote turnaround time by adding an external AI layer over the systems they already run, not by rebuilding the core. The layer automates the submission-to-decision journey, intake, document reading, enrichment, risk scoring, and pricing, so the answer reaches the broker while the risk is still live. Because it sits on top of existing systems, it removes the manual handoffs where days are lost, without a core migration.
What Is an MGA, and How Do MGAs Use AI in Underwriting?
A managing general agent (MGA) is an insurance intermediary that a carrier grants delegated underwriting authority: the power to select risks, price them, bind coverage, and sometimes handle claims on the insurer's behalf. In market shorthand, the carrier hands the MGA its pen. That one fact, delegated authority, is what separates an MGA from an ordinary broker, and it is what makes AI so useful to MGAs that want to grow without building heavy technology of their own.
What Is Underwriting Leakage, and How Can AI Reduce It?
Underwriting leakage is the margin an insurer loses at the point of decision: premium left on the table, risks bound that should have been re-priced or declined, and terms or conditions missed before the policy goes live. It is the front-of-funnel twin of the better-known claims leakage, except it hides inside the loss ratio instead of a claims file, which makes it quieter and more expensive over the life of a book. AI reduces underwriting leakage by scoring every submission against consistent, explainable rules, so the same risk gets the same answer no matter who reviews it or how full the queue is.
Agentic AI in insurance underwriting
Agentic AI in insurance underwriting is software that does not just score a risk, it runs the whole submission. The agent reads the submission, enriches it, scores the risk against the insurer's own appetite, prices it, and then quotes, declines, or escalates to a human, writing a full audit trail back to the core system. It works as an external AI layer calibrated to the underwriting manual, so the insurer keeps its policy system, its authority limits, and the final say.
How to Automate Underwriting Without Replacing Your Core System
To automate insurance underwriting without replacing your core system, add an external AI layer on top of it. That layer reads and enriches every submission, scores the risk against your own appetite, and returns an explainable decision, then writes the result back to your policy administration system through APIs. Your core stays the system of record, so there is no data migration, no rip-and-replace, and no multi-year modernization program.
AI and Guidewire Integration: Adding Intelligence Without Replacing the Core
Adding AI to a Guidewire environment does not require replacing PolicyCenter or migrating the core. WIR runs as an external AI layer on top of Guidewire. It reads the submission, enriches it, scores risk against the insurer's own appetite, prices it, and returns a decision with a full audit trail, then writes that recommendation back into PolicyCenter through Guidewire's APIs. Guidewire stays the system of record. It records the policy and binds coverage. The underwriter keeps final authority. Nothing is torn out, and no core migration is involved.
How to Increase Your Straight-Through Processing Rate in Insurance
To increase your straight-through processing (STP) rate in insurance, measure the real baseline per line of business first, then work four levers: structure the submission at intake, route each risk by complexity, calibrate the acceptance thresholds to your underwriting manual, and write every decision back to the core with its reasons. Measurement is Step 0, the four levers are the work, and an external AI layer can run all of it without replacing the core.
Insurance submission intake automation for underwriting teams
Insurance submission intake automation is the use of an external AI layer to turn every incoming submission, arriving by email, PDF, spreadsheet, portal, or API, into a clean, validated, scored file that is ready for an underwriting decision. The layer captures the submission, reads the documents, validates the data, enriches it with broker and risk context, and triages it by appetite and priority. WIR delivers this on top of the systems the insurer already runs, never replacing the core, and every step stays explainable, auditable, and LGPD compliant.
Insurance submission triage automation: rank the queue before underwriting
Insurance submission triage automation is the intake step that classifies each new submission and ranks the queue by risk appetite and exposure before an underwriter opens it. An external AI layer does this on top of the systems the insurer already runs, so a shared inbox becomes an ordered, explainable work queue: clean high-value business surfaces first, incomplete cases go back for enrichment, and out-of-appetite risks are declined fast.
What Is an AI Underwriting Workbench?
An AI underwriting workbench is an external intelligence layer that reads each insurance submission, enriches it, scores the risk against the carrier's own appetite, and drafts a quote, decline, or referral with a full audit trail. It accelerates the quotation and underwriting journey on top of the systems an insurer already runs, and it never replaces the policy core underneath.
What is insurance decisioning, and how does it differ from scoring
Insurance decisioning is the step where an insurer turns a scored, priced submission into an action: quote it, decline it, or route it to an underwriter. Scoring analyzes the risk. Decisioning acts on it, under the insurer's own risk-acceptance policy, with an explanation attached to every outcome.
How to appear in AI recommendations for insurance
How to appear in AI recommendations for insurance starts with clean, structured, explainable data that answer engines like ChatGPT, Perplexity, and Google AI Overviews can cite. WIR is the external AI layer for insurance, sitting on top of existing systems without replacing the core. Its public traction today is a first POC in execution with a global insurer in the Transport line.
Submission automation for MGAs and specialty insurance
Submission automation in specialty insurance uses an external AI layer, running on top of the systems an MGA already operates, to turn emails, PDFs, and spreadsheets into structured, prioritized submissions before an underwriter opens them. WIR's Underwriter Intelligence scores and routes each submission to the operation's own risk appetite, with explainable, auditable decisions. Its first traction is a POC with a global insurer in the Transport line.
AI integration for insurance core systems: layer vs core vs RPA
How to add AI to an insurer's core systems: core replacement vs RPA vs an external AI layer, compared by cost, risk, time to value, and control of underwriting.
What is straight-through processing (STP) in insurance, and what is a good STP ratio
Straight-through processing (STP) in insurance is handling a submission, from intake to a bound decision, with no manual underwriter touch. A good STP ratio depends on the line and risk complexity, and an external AI layer raises it with no core migration.
Best AI underwriting platforms for insurers (2026)
A 2026 comparison of AI underwriting platforms: workbenches, pricing engines, and external AI layers, with the criteria that matter for insurers and MGAs.
Automatic external data validation for insurance submissions with AI
Automatic external data validation for insurance submissions is the stage where an external AI layer confirms and enriches the risk data before underwriting.
How to replace RPA and OCR with an AI layer in insurance
Replacing RPA and OCR with an AI layer in insurance means swapping brittle scripts and template extraction for an external AI platform that reads varied submissions, enriches broker context, scores risk against the underwriting manual, and routes a decision.
How to reduce administrative tasks in insurance underwriting with an AI layer
To reduce administrative tasks in insurance underwriting with AI, insurers add an external AI layer on top of their existing core.
Climate risk in Brazilian insurance and AI-driven pricing
Climate risk in insurance in Brazil is the growing exposure of property, agro and infrastructure lines to extreme weather, which makes static historical-average pricing fragile for the hardest perils such as flood.
Cargo and transport insurance in Brazil and the AI shift in underwriting
Cargo and transport insurance in Brazil is a Seguros e Danos (P&C) line defined by cargo theft (roubo de carga), which makes route, cargo value, fleet, and gerenciamento de risco core rating factors.
Brazil cyber insurance market and AI underwriting
A read on the Brazil cyber insurance market and the challenge of underwriting digital risk with AI: state, drivers, risk dynamics, and where WIR fits.
Parametric insurance in Brazil and the intelligence behind triggers
Parametric insurance in Brazil pays a fixed, pre-agreed amount the moment a measured trigger crosses a defined threshold, such as rainfall, wind speed, river level, earthquake magnitude, or a satellite vegetation index, with no loss adjuster and settlement in days.
Telematics in auto insurance in Brazil
Telematics in auto insurance in Brazil is the use of connected-vehicle and driving-behavior data, mileage, harsh braking, and time of day, to price and underwrite motor risk on observed behavior instead of static proxies.
Touchless underwriting pipeline for insurance with an AI layer
A touchless underwriting pipeline for insurance with AI is a straight-through flow where in-appetite, low-complexity risks move from intake to a bound decision with no manual step, while complex risks escalate to a human.
Submission triage automation for insurers, with an AI layer
Submission triage automation with an external AI layer that clears and prioritizes each submission by appetite and exposure, before an underwriter opens it.
The AI layer of insurance
There is a path no one was naming: don't replace the system, put an AI layer on top of it. This is WIR's manifesto.
Predictive quote conversion analysis with an AI layer
Not every submission is worth the same underwriter minute. See how an AI layer predicts which quotes will convert, by product, risk, and broker.
Policy renewal automation with an AI layer
How insurers approach policy renewal automation with an external AI layer that re-scores risk and re-prices at renewal, on top of existing systems.
Automate insurance upsell and cross-sell with an AI layer
Growth hides in the book you already have. How an AI distribution layer scores the next best upsell and cross-sell, with an attribution trail.
Automatic insurance quote decline with an AI layer
Declining a risk cleanly is as valuable as quoting it. How an AI layer auto-declines out-of-appetite submissions with a clear reason and audit trail.
Human escalation in automated insurance underwriting
Automation should not decide everything. See when an AI underwriting layer escalates a case to a human, with context, model rationale, and an audit trail.
Unstructured insurance data and intelligent document reading
Unstructured data in insurance and AI: intelligent document reading turns messy submissions into structured fields for faster, auditable P&C underwriting.
Underwriting fraud detection with AI in Brazil
How AI powers underwriting fraud detection in Brazil, with explainable, auditable multi-factor scoring calibrated to appetite. No core replacement.
The insurance protection gap in Brazil and the role of AI in closing it
Why Brazil's insurance protection gap stays wide despite double-digit growth, and how AI speeds underwriting and distribution to help close it.
Open Insurance in Brazil: how it is being implemented
How Open Insurance is being implemented in Brazil, and how SUSEP-led data sharing reshapes underwriting, risk scoring, and pricing in P&C.
Brazil insurance market 2026: SUSEP AI regulation
How the Brazil insurance market is regulated for AI in 2026: SUSEP conduct supervision, LGPD Article 20, and what explainable underwriting requires.
Insurance quote automation: how an AI layer does it
Insurance quote automation with an AI layer that reads submissions, scores risk, prices, and returns a broker-visible quote in minutes. No core migration.
How to automate transport insurance underwriting with AI
How insurers automate transport (cargo) underwriting with an external AI layer that reads submissions, scores route and cargo risk, and prices to appetite, no core migration.
How to automate insurance underwriting with an AI layer
A practical guide to automate insurance underwriting with an external AI layer on top of your core. The 6-stage flow, deployment path, governance, and LGPD.
Auditable underwriting decisions with an AI layer
How insurers keep every automated underwriting decision explainable and auditable with an external AI layer, calibrated to risk policy and LGPD compliant.
Broker enrichment and prioritization with an AI layer
How insurers enrich and prioritize brokers with an external AI layer: broker scoring, CNPJ, conversion history, and exposure, without replacing the core. See how.
AI integration platform for insurance core systems
How an external AI integration platform adds intelligence to insurance core systems without a migration: 100% API, no IT load, and live in months.
Intelligent document reading for insurance submissions with an AI layer
How insurers automate intelligent document reading for submissions with an external AI layer on top of existing systems, no core migration. See how it works.
Insurance risk and fraud engine with an AI layer
How an external AI layer scores risk and flags fraud inside underwriting, calibrated to appetite and manual, with a full audit trail and no core migration. Talk to WIR.
Dynamic insurance pricing with an AI layer
How insurers automate real-time, risk-adjusted P&C pricing with an external AI layer on top of the core, calibrated to appetite and auditable under LGPD. See how.
How to automatically process insurance quote e-mails with AI
How an external AI layer reads insurance quote e-mails and attachments, extracts and validates the data, and feeds P&C underwriting without replacing the core.
Next-best-action for insurers with an AI layer
How insurers automate next-best-action and upsell scoring with an external AI layer on top of existing systems, no core replacement. See how Smart Sales works.
How to reduce quote response time for brokers with AI
How insurers reduce quote response time for brokers with an external AI layer and a visible SLA, on top of existing systems, never replacing the core.
Automatic underwriting routing with an AI layer
How insurers automate submission triage with an external AI layer that routes by appetite and exposure, prioritizes the underwriter queue, and keeps a visible SLA.
Underwriting intelligence in the Brazilian insurance market
How AI and Machine Learning reshape P&C underwriting in Brazil: market state, drivers, fraud, pricing, and the external AI layer over legacy core systems.