Skip to content
AI Agent · Underwriting

Meet your AI Agent for ACORD Form Extraction.

Every commercial submission starts with an ACORD form — and every quote stalls there. Bevaya reads every form and variant, extracts every field with 99%+ accuracy, and lands the data clean in your underwriting workbench.

Trusted by leading P&C carriers, brokers, and TPAs
How your AI Workforce works

From broker inbox to underwriting workbench.

 
Commercial Insurance Application
Applicant Information Section
Date (mm/dd/yyyy)
04/22/2026
AI CLASSIFICATION: ACORD 125 · COMMERCIAL APPLICATION · ED. 2016/03 (0.99)
Agency
 
Carrier
 
Contact Name
 
Phone
 
Policy Number
 
Status of Transaction
Quote
Issue Policy
Bound
Renew
Applicant Information
Named Insured
Stoltz Trucking & Logistics LLC
Effective Date
 
Lines of Business
Commercial GL
Commercial Prop
Workers Comp
Business Auto
RECOGNIZED — EVERY FORM, EVERY VARIANT
ACORD type, edition, and broker layout variant detected per page
Handwritten edits, faxed scans, and non-standard layouts all read
No broker re-keying — Bevaya accepts what comes in
Recognizing every form
 
 
 
 
 
 
 
InsurGPT™ · every field with X-Ray
CROSS-PACKAGE CHECK
Named insured matches across all 4 ACORDs
0.99
NAICS CODE
484121 · Long-distance trucking · validated
0.98
ADDRESS NORMALIZATION
HQ + 6 locations · USPS-validated
0.97
PRIOR CARRIERS
3 prior · listed on 125 and 130
0.96
LOSS HISTORY RECONCILE
$240K on 125 · $312K in loss runs · variance
FLAG
CLASS CODES
WC 7228 · GL 99317 · validated
0.95
Errors caught upstream
 
 
Guidewire · Duck Creek
Verified data populated · high-confidence STP
Salesforce · Majesco
Synced via API in your target schema
UW Workbench
Submission ready for triage
Review Queue · 2 minutes
1 low-confidence field · loss variance flagged
Lands clean in your workbench
Recognizes every form
Picks up submissions from Outlook, broker portals, MGA platforms, SFTP, and API
Recognizes each ACORD form by type, edition, and variant — 125, 126, 130, 140, and more
Handles handwritten edits, faxed scans, and broker-custom layouts without re-keying
Pulls every field
Applicant info, locations, loss history, class codes, prior carriers — every field captured
InsurGPT™ scores every extraction so underwriters know exactly where to look
X-Ray Mode source-links every value back to the exact spot on the form
Catches errors upstream
Cross-checks values across the entire submission package — named insured, dates, addresses
Validates NAICS, normalizes addresses, and verifies class codes against current schedules
Reconciles loss history on the application against the actual loss runs
Lands it clean in your workbench
Verified data pushed into Guidewire, Duck Creek, Salesforce, Majesco, or your workbench
High-confidence fields flow straight through — true straight-through processing
Low-confidence fields route to a 2-minute Review Queue with full X-Ray context
Production performance, not pilot numbers.

Built and proven at scale

Extraction accuracy — on every ACORD field, every form, every edition.
99%+
Turnaround per submission — down from 10–15 minutes of manual review.
<2 min
Processed straight-through — Top 20 P&C insurer, production deployment.
80%
Bevaya — Case Studies Module

Bevaya’s AI agents reduced our claims indexing from 5 days to under an hour. The accuracy was better than our most experienced staff on day one.

Harry Talbert

Harry Talbert SVP of Information Systems · Eastern Alliance

Read the full story

99% straight-through processing and 246% ROI in just 6 months.

F5

Fortune 500 Carrier Property & Casualty

Read the full story

We evaluated six AI vendors. Bevaya was the only one that understood our underwriting workflows from day one — no six-month education period.

CU

Chief Underwriting Officer Specialty Carrier · $2B+ GDP

Read the full story

The confidence scoring changed everything. Our reviewers know exactly which items need attention and which can go straight through. We trust it.

DO

Director of Operations Workers’ Compensation Insurer · 74 NSP

Read the full story

Cut COI turnaround from 24 hours to minutes and saved millions.

T5

Top-5 Broker National

Read the full story
Bevaya · Engine Inside (4-tab) preview
The engine inside

Most AI sees a PDF.
InsurGPT™ sees an ACORD form.

Trained on 300M+ insurance documents — not the open web. That's why the field-level accuracy lands at 99%+ instead of the 67% generic GPT benchmarks deliver on insurance docs.

Trained on every ACORD variant | Coverage

Every state edition, every revision year, every broker customization.

Reconciles across the package | Reconciliation

Catches mismatches between the 125, the loss runs, and the SOV before they reach the underwriter.

Built for messy reality | Real-world

Faxes. Photos. Handwritten margin notes. Forms pasted into email bodies. Not the demo version — the real one.

Traceable to the source | Cited

Every value carries a confidence score and a source link back to the page where it was found. A second model double-checks the first.

 
What It Does

ACORD Form Extraction AI Agent

Every ACORD form, every variant, every edge case — including the supplements, SOVs, and loss runs brokers send alongside.

ACORD 125 — Commercial Application: applicant info, premises, prior carriers, loss history, coverage requested — every field, every edition.
ACORD 126 — General Liability: class codes, exposure data, operations descriptions, sub-contractor and additional insured detail.
ACORD 130 — Workers Comp: class codes by state, payroll by class, owner/officer detail, experience modification, prior carrier history.
ACORD 140 — Property Section: location schedules, construction details, protection class, occupancy, COPE data, building values.
State variants & year revisions — every state edition, every line-of-business supplement, every annual revision, automatically.
Handwritten edits & broker customizations — margin notes, crossed-out values, broker-modified templates, handwritten supplements.
Supplemental applications — carrier-specific and class-specific supplements across every format and line of business.
Faxed, scanned, or photographed forms — whatever the broker sent, the AI Agent reads. GutenOCR handles the worst real-world image quality.
Field-level confidence + X-Ray Mode — every value carries a 0–100 score and a source link back to the page where it was found.
How it works

An AI layer that sits on top of your existing stack.

Bevaya orchestrates the tools your team already uses. Your stack stays in place. The intelligence layer over it.

01 Sources
Microsoft Outlook
Mailroom / SFTP
Custom API / IPA
02 Bevaya Platform
Bevaya

Bevaya

Intelligence + orchestration layer for insurance workflows

03 Destinations
Systems of record
Guidewire ClaimCenter
Salesforce Industries
Duck Creek Claims
Document management
Alerts
Slack
Microsoft Teams
MORE AI AGENTS FOR UNDERWRITING

Build the full underwriting AI program.

Underwriting Automation Overview

End-to-end automation across the underwriting lifecycle — from submission intake through quote, bind, and renewal.

Underwriting

Submission Intake

Monitors broker inboxes, classifies submissions, extracts data from ACORDs and supplementals, and populates underwriting systems automatically.

Underwriting

Loss Run Processing

Extracts and normalizes loss data across carriers, identifies trends, and produces structured loss histories ready for underwriting review.

Underwriting

Schedules & SOV Extraction

Reads Statement of Values spreadsheets and property schedules in any format, normalizes the data, and delivers clean structured exposure data.

Underwriting

ACORD Form Extraction

Extracts data from all ACORD forms, validates fields against carrier rules, and pushes clean structured data into underwriting systems.

Underwriting

Policy Renewal Handling

Pulls expiring policy data, refreshes exposures and loss runs, and assembles a complete renewal package for the underwriter.

Underwriting

Policy to Policy Comparison

Compares expiring and renewal policy terms, flags coverage and limit changes, and surfaces material differences for underwriter review.

Underwriting

Claims Automation Overview

End-to-end automation across the claims lifecycle — from FNOL through investigation, settlement, and payment.

Claims

FNOL / FROI Setup

Captures first notice of loss and first report of injury details from email, phone, or portal, structures the data, creates the claim file, and assigns it to the right adjuster.

Claims

Claim to Policy Comparison

Compares loss details against policy terms to surface coverage, exclusions, and obligations before the adjuster makes a call.

Claims

Claim Indexing

Classifies, indexes, and routes claims documents to the right adjuster or team. Handles hundreds of document categories with up to 99% STP.

Claims

Legal Demands Identification & Extraction

Reads demand letters and legal correspondence, extracts allegations, damages, and deadlines, and routes them with full context to the claims team.

Claims

Medical Bill Identification & Extraction

Reads medical bills and provider records, extracts CPT/ICD codes, charges, and dates of service, and links them to the claim for adjuster review.

Claims

Invoice Payment Processing

Processes vendor and provider invoices against claim files, validates against authorized services, and routes for payment approval automatically.

Claims

Claim File Summarization

Reads complex claim files and produces role-specific summaries. Key details, timelines, and action items surfaced for the adjuster.

Claims

Policy Servicing Automation Overview

End-to-end automation across policy servicing — from endorsements and certificates through billing inquiries and audits.

Policy Servicing

Endorsement Processing

Reads endorsement requests, validates against the policy, drafts the change, and routes it for approval and issuance.

Policy Servicing

COI Creation

Generates Certificates of Insurance from policy data, validates additional insured language, and delivers certificates to brokers and insureds automatically.

Policy Servicing

Premium Audit Processing

Classifies audit documents, extracts payroll and classification data, populates systems, and requests missing information from policyholders.

Policy Servicing
Custom AI Agents

Don't see the AI Agent you need?

Bevaya's platform extends to fit your operation. We can build a custom AI Agent for your specific underwriting workflow — trained on Bevaya's insurance-native AI and integrated with your systems.

Resources & insights

From the field and from the lab

Case Study - claims
Research

Page stream segmentation with LLMs

How Bevaya Labs approaches a foundational problem in insurance document AI.

Case Study - claims
Case Study

Workers' comp carrier processes claims 100x faster

How indexing automation delivered 432% ROI in 12 months.

2026.06.02-library-webinar-registration-how-to-establish-clear-ai-ownership-in-your-insurance-organization
Architecture

Inside the Bevaya platform architecture

How specialized models, HITL controls, and integrations come together in production.

FAQ

Frequently asked questions.

Bevaya extracts data from every commercial ACORD form, including ACORD 125 (Commercial Application), 126 (General Liability), 130 (Workers Compensation), and 140 (Property Section), plus all state-specific variants, year editions, and supplemental applications. The model is also trained on broker-customized layouts and handwritten edits.

Bevaya's ACORD Form Extraction achieves 99%+ field-level accuracy in production. Every extracted value carries a 0–100 confidence score and an X-Ray Mode source link back to the original form. Generic GPT-class models benchmark around 67% on insurance documents; InsurGPT™ benchmarks at 93%+ in head-to-head testing on the same documents.

Yes. Bevaya's GutenOCR engine handles low-quality scans, faxed pages, and photographed forms. InsurGPT™ recognizes handwritten edits, margin notes, crossed-out values, and broker-modified templates — the messy real-world submission package, not just the clean demo version.

Bevaya connects to all major policy admin and underwriting systems, including Guidewire PolicyCenter, Duck Creek Policy, Salesforce Financial Services Cloud, and Majesco. Inbound channels include Outlook, Gmail, broker portals, MGA platforms, mailroom scanners, SFTP, and any custom source via API. External data sources include SIC/NAICS verification, ISO, and LexisNexis. No rip-and-replace required.

Yes. Every step in every run is logged — actor, action, resource, timestamp — with immutable entries. Every extracted field carries a confidence score and X-Ray Mode showing exactly where the value came from in the source form.

Submissions Intake is the workflow AI Agent that monitors channels, classifies broker emails, and orchestrates the full submission package end to end. ACORD Form Extraction is the field-level specialist for the ACORD forms themselves. Most carriers deploy both: Submissions Intake orchestrates the package; ACORD Form Extraction handles the forms inside it.

Deployment typically takes 8–12 weeks from kickoff. ACORD Form Extraction ships as a pre-configured workflow on the Bevaya Canvas, powered by InsurGPT™ models already trained on the full ACORD form family. Implementation focuses on connecting your intake channels, mapping to your workbench, and tuning to broker-specific variants — not building from scratch. Bevaya manages deployment end to end.

No. Bevaya replaces the keying, not the underwriting. The AI Agent reads, extracts, and validates the data; your underwriters evaluate risk, price, negotiate, and bind. Customers consistently report that AI handles the intake volume so the same team carries more submissions to bind — not fewer underwriters.

GET STARTED

Ready to design, deploy, and goven your AI workforce?

Bevaya AI Agents can help you triage, analyze, and recommend across underwriting, claims, and policy servicing.
Let's connect and show you how it works.