Wirewalk Instituteof AI
fin-230 Intermediate AvailableFlagship

Building and Deploying AI in Insurance

For pricing actuaries, underwriting and claims leaders, SIU managers, model validators and AI governance leads who build, buy or defend specific AI systems in a US insurer, one job task at a time.

Level
Intermediate
Study time
16 hours
Lessons
21 in 6 modules
Exam
20 questions

About this course

This is not an "AI in insurance" overview. Every lesson is anchored on one task a named insurance role performs: a pricing actuary answering a rate analyst's objection letter about a GBM-derived interaction, a compliance analyst computing adverse impact ratios on a preferred-tier placement model, an SIU manager setting a fraud-referral threshold that fits investigator capacity, an auto physical-damage manager deciding which photo estimates may close without an appraiser, a commercial underwriter checking an extracted schedule of values, a property underwriting manager writing the renewal rule for aerial roof scores, an adjuster auditing every policy provision a copilot quoted, and a governance lead assembling a regulator data-call response. For each task you see how it is done today, where AI helps and where it must not decide, the data and core systems actually involved, the product categories and named vendors in use (described from public materials, with vendor performance figures labeled as vendor claims and no endorsement), the controls that apply to that exact task, and how to measure success.

Six Harborview decision records connect the existing worked examples. US state applicability, model-text versus enacted-law status, input and evidence boundaries are taught at the task. All examples and case records are synthetic. The planned 16-hour core uses supplied outputs; code extensions and full source reading add time. See the dated source ledger for authenticated sources and explicit access limits.

What you will be able to do

  • Compare GLM/GBM pricing evidence, selected rating terms and impact while identifying independent confirmation and state/source-use limits
  • Compute and interpret aggregate inferred-group disparity with proxy error, uncertainty and no insurance legal pass threshold
  • Compare less discriminatory alternatives and accelerated life cutoffs with loss, slippage, workload and specific notice evidence
  • Design staffed SIU, photo-estimate and FNOL routing with claim-specific evidence, applicable running clocks and escalation
  • Reconcile field/location extraction and loss conventions, then assess imagery using parcel, date and dispute evidence
  • Repair generated coverage answers against issued forms and test category-complete evaluation and bounded UM routing without clinical approval claims
  • Build an entity/use inventory, vendor evidence controls and a candid versioned regulator-response index
  • Distinguish source-unit drift and immature outcomes from model failure and design impact review, monitoring and fallback

Syllabus

6 modules · 21 lessons · about 16 hours

Module 1: Pricing Actuary: A GBM-Informed Auto Rating Plan You Can Defend in a Filing

Use gradient-boosted models to find risk signal, translate it into filed rating factors, and answer what state rate reviewers ask under Model 1780 rating law, the NAIC predictive-model white paper, Colorado Regulation 10-1-1 and NY Circular Letter 7.
3 lessons
  1. Fitting a GLM and a GBM on Synthetic Auto Loss Data and Using SHAP to Find the Missed Interaction
  2. Turning a GBM Finding Into a Filed Rating Term and Answering a Rate Analyst's Objection Letter
  3. Telematics and Other External Data in the Rating Plan: Consent, Provenance and Colorado's ECDIS Rules
  • Decision Lab · 12 minThe factor that lost its provenanceA Harborview evidence conflict changes a specific insurance AI recommendation.

Module 2: Compliance Analyst and Life Underwriting Owner: Testing Models for Unfair Discrimination

Compare matched disparity and life-triage evidence with uncertainty, less discriminatory alternatives and specific applicant notices; verify current state requirements separately.
4 lessons
  1. Estimating Race and Ethnicity With BIFSG When Insurers Do Not Collect It, and Its Error Rates
  2. Computing Adverse Impact Ratio and Standardized Mean Difference on a Synthetic Tier-Placement Model
  3. Writing the Circular Letter 7 Three-Step Memo and Searching for a Less Discriminatory Alternative
  4. Accelerated Life Underwriting Triage: Setting the Cutoff, Measuring Mortality Slippage and Sending the 15-Day Notice
  • Decision Lab · 12 minA fairness pass with the wrong denominatorA Harborview evidence conflict changes a specific insurance AI recommendation.

Module 3: SIU, Auto Physical-Damage and Claims Operations Leads: Automating Claims Without Automated Wrongful Denials

Set fraud-referral, photo-estimate and FNOL routing rules that speed legitimate claims while keeping every adverse action with a human under unfair claims settlement rules.
4 lessons
  1. SIU Manager: Setting a Fraud-Referral Threshold That Fits Fixed Investigator Capacity
  2. Finding a Planted Claims Ring With a Network View and Documenting the Human Decision Point
  3. Auto Physical-Damage Manager: Auto-Approval Rules for Photo-Based Estimates Against a Supplement-Rate Target
  4. Claims Operations Lead: FNOL Triage for Injury Recall, Human Handoff and State Acknowledgment Clocks
  • Decision Lab · 12 minEighty investigations and a running payment clockA Harborview evidence conflict changes a specific insurance AI recommendation.

Module 4: Commercial and Property Underwriters: Submission Intake and Imagery-Based Renewal Decisions

Extract and verify broker submissions and loss runs with traceable citations, and write renewal rules for aerial roof-condition scores that meet state imagery bulletins.
3 lessons
  1. Commercial Underwriter: Extraction Schemas for Applications and Schedules of Values, Scored Field by Field
  2. Summarizing a Five-Year Loss Run With Page Citations and Catching Paid-Versus-Incurred Errors
  3. Property Underwriting Manager: Renewal Rules for Aerial-Imagery Roof Scores Under State Bulletins
  • Decision Lab · 12 minThe roof invoice and the missing zeroA Harborview evidence conflict changes a specific insurance AI recommendation.

Module 5: Adjusters, Agents and Utilization Managers: Generative Assistants With Audit Trails

Ground and test generative drafting and question-answering against the issued policy, the carrier's own guides and physician-review laws so no misrepresentation or AI-only denial leaves the building.
3 lessons
  1. Adjuster Copilot: Drafting a Reservation-of-Rights Letter and Auditing Every Cited Provision
  2. Agent Coverage Assistant: Building and Grading an Evaluation Set With Endorsement-Dependent Questions
  3. Health-Plan UM Leader: Prior-Authorization Rules Where No Path Ends in an AI-Only Denial
  • Decision Lab · 12 minA real quote from the wrong controlling formA Harborview evidence conflict changes a specific insurance AI recommendation.

Module 6: AI Governance Lead, Vendor-Risk Lead and Model Validator: Running the AIS Program

Build the inventory, vendor controls, drift monitoring and data-call response that the NAIC Model Bulletin, Colorado and New York expect insurers to produce on request.
4 lessons
  1. Building the AI Systems Inventory and Tiering It by the NAIC Bulletin's Degree-of-Potential-Harm Factors
  2. Vendor-Risk Lead: Due-Diligence Questions and Contract Clauses for a Third-Party Score
  3. Model Validator: Drift Monitoring for a Pricing GBM and a Claim-Severity Model
  4. Assembling the Response to a Regulator Data Call Under Section 4 of the NAIC Bulletin
  • Decision Lab · 12 minGreen book, broken sourceA Harborview evidence conflict changes a specific insurance AI recommendation.
  • Capstone simulation · 30 minThe six-system evidence hearingDefend a cumulative portfolio after a cross-affiliate scope conflict and a ten-day data call.

Exam and certificate

The exam unlocks once you have completed every lesson and played each Decision Lab to an ending. It draws 20 questions from a larger bank, spread across the modules, with the answer options shuffled for each attempt. You have 40 minutes and need 80% to pass. After a failed attempt you can try again after 24 hours, up to 3 attempts in total. Feedback is given per module, never as the answer key.

Passing earns a certificate of completion and assessment from the Wirewalk Institute of AI with a unique ID that anyone can check at institute.wirewalk.com/verify. It is not an accredited qualification or a vendor certification. How certification works.

The Wirewalk Institute of AI is independent. This course teaches publicly documented products and is not affiliated with or endorsed by Akur8, Shift Technology, CCC Intelligent Solutions, Tractable, Hi Marley, LexisNexis Risk Solutions, CAPE Analytics (Moody's), ZestyAI, Cambridge Mobile Telematics, Arity, Cytora, Federato, Sixfold, Guidewire, or Duck Creek. Product and company names are trademarks of their respective owners.

More in AI in Financial Services

All 15 courses in this track
fin-001 Free intro Available

AI in Financial Services: The One-Hour Briefing

A free one-hour briefing built on eight real financial-services jobs, not an overview: spreading financials for a credit memo, writing adverse-action reasons from an ML score, drafting a SAR narrative, setting fraud-score thresholds, checking an earnings-call summary, reviewing an advisor's AI meeting note, summarizing a commercial loss run and reviewing a photo-based auto estimate. Each lesson shows where AI helps, where it must not decide, and the rule that governs that task.

1 h · 9 lessons Free intro
fin-110 Basic Available

AI for Banking Professionals

Practice banking work on a self-contained fictional bank packet: repair service drafts, reconcile credit facts and reason records, summarize investigation leads, compare review burdens and prepare a scoped recommendation. Supplied outputs let you complete the work without a model account, code installation or real customer records.

12 h · 18 lessons See price at checkout
fin-210 Intermediate Available

Building and Deploying AI in Banking and Lending

Follow one composite $46 billion bank through a year-long AI program with public promises and OCC examinations on the calendar. Build and test the systems bank AI teams own, from an ML credit model to a servicing assistant, prove what they are worth in savings the CFO can book, and make the strategic trade-offs in six Decision Labs and a capstone simulation.

16 h · 18 lessons See price at checkout