# Liability Forecasting Engine

*/Opportunities/Liability_Forecasting_Engine*

## Opportunity Overview

**Wedge**: Target commercial auto and workers compensation books within regional carriers. These lines suffer from high claim frequency and volatile severity due to social inflation, offering fast proof of concept when the engine identifies escalating claims early. Expansion moves horizontally into general liability and professional lines, then vertically into pricing engines once the claims data establishes a reliable feedback loop.
**Timing**: Large language models can now reliably extract unstructured text from adjuster notes, medical records, and legal filings, translating them into structured risk vectors. This capability allows for continuous recalculation of reserve needs without waiting for manual, quarterly actuarial reviews.
**Why This I C P**: Mid-market commercial carriers carry significant long-tail risk but lack the internal quants and proprietary data lakes of top-tier giants. They face acute financial pressure from reserve surprises and readily adopt predictive software to stabilize their balance sheets.
**Size Of Prize**: Approximately 2,500 mid-to-large US commercial property and casualty insurers and large self-insured entities spend an average of $150,000 annually on actuarial consulting and legacy modeling software, yielding a $375 million addressable market.
**Gap Narrative**: Actuaries and risk managers currently rely on static, historically weighted models to predict long-tail claims, failing to account for emerging litigation trends or granular claim-level changes. They need a system that continuously re-evaluates open claims against real-time medical costs, legal precedents, and unstructured adjuster notes to project ultimate liabilities dynamically.
**Defensibility**: The system builds a proprietary cross-carrier mapping of claim attributes to ultimate settlement costs, creating a localized data network effect. As the engine ingests more unstructured adjuster notes and pairs them with final payout data, its predictive accuracy outpaces any single carrier's internal models. Switching costs lock in as the actuarial team embeds the engine into their regulatory filings and quarterly reserve workflows.
**Why This Thesis**: A deterministic software model fed by AI extraction bridges the gap between the actuarial team and claims adjusters. This structure fits the problem precisely because reserving requires strict, auditable mathematical outputs, while the leading indicators of liability reside in messy, unstructured text that AI parses continuously.

## Opportunity Linked Thesis

**Thesis**: [Software](/Theses/Software)

## Opportunity Linked I C P

**Icp**: [Insurance Carrier](/CompanyTypes/Insurance_Carrier)

## Opportunity Market Sizing

_Illustrative — target and order-of-magnitude estimate figures, not an achieved track record (this Thing is concept-stage)._

**S A M**: ~$400-600M US and UK mid-to-large casualty and specialty carriers
**S O M**: ~$15-40M
**T A M**: ~8,000-12,000 global P&C carriers and reinsurers × ~$150k-250k/yr ≈ ~$1.2-3.0B
**Growth Rate**: ~12-18%/yr, driven by rising casualty claim severity, social inflation, and the failure of historical models to price emerging risks
**Paid Comparable Spend**: ~$300k-800k/yr on external actuarial consulting, legacy risk modeling software licenses, and manual claims analysis labor

## Opportunity Incumbents

- [FIS Prophet](/Products/FIS_Prophet) — Tool
- [Moody's AXIS](/Products/Moody's_AXIS) — Tool
- [WTW ResQ](/Products/WTW_ResQ) — Tool
- [Excel Actuarial Templates](/Products/Excel_Actuarial_Templates) — Spreadsheet
- [Milliman Consulting](/Products/Milliman_Consulting) — Service
- [Oliver Wyman Actuarial](/Products/Oliver_Wyman_Actuarial) — Service
- [In-House Python Models](/Products/In-House_Python_Models) — DIY

## Opportunity Win Conditions

**Kill Thresholds**:
- Pilot-to-paid conversion rate < 25% after 90 days of live testing
- Sales cycle exceeds 6 months for a $150,000 ACV contract
- More than 50% of active users export engine data to Excel to complete their primary reserving analysis
- System fails to match manual reserving accuracy within a 2% margin of error during backtesting
**Leading Metrics**:
- Data ingestion and normalization time per 10,000 claims
- Weekly active actuarial users per deployed carrier
- Number of mid-cycle pricing adjustments triggered by engine outputs
- Model backtest variance against verified historical loss runs
**What Proves Right**: Actuarial teams load live claims data into the engine weekly instead of relying on quarterly spreadsheet rollups. Pilot customers sign $150,000 annual contracts after validating that the engine isolates emerging casualty severity trends faster than legacy tools. Carrier pricing desks directly consume the engine outputs to adjust premium rates mid-cycle.
**What Proves Wrong**: Chief Actuaries refuse to trust the engine outputs without manual Excel recreation, rendering the workflow redundant. Procurement rejects the target price point because they view the tool as a simple visualization layer on top of their existing Moody's AXIS deployment. The engine fails to backtest accurately against complex historical casualty claims, causing pilot participants to churn.

## Opportunity Build Profile

**Hardest Part**: Extracting structured fact patterns from unstructured legal demands and financial documentation to train a predictive model that achieves reliable confidence intervals for CFO reporting. Handling the severe long-tail distribution of settlement amounts without skewing baseline forecasts is computationally demanding.
**Min Viable Scope**: Forecast exposure strictly for US employment practice liability claims using historical company HR data and public settlement benchmarks. Deliberately exclude general commercial litigation, insurance policy routing, and settlement workflow automation.
**Cold Start Problem**: Predictive models require thousands of historical settlements to identify patterns, but corporate liability data is highly siloed and strictly confidential. Break this by bootstrapping with scraped public court dockets and securing a foundational data partnership with a specialized litigation finance firm to access private settlement figures.
**Time To First Value**: 2-4 weeks to ingest internal historical claims data, map it to the industry ontology, and run the first localized backtest
**Data Moat Available**: true
**Technical Difficulty**: High

## Neighborhood

### Where the gap lives

- [State and local government agencies](/Employers/State_and_local_government_agencies) — latent gap · Employers
- [Post-Notice Inquiry Rate](/Metrics/Post-Notice_Inquiry_Rate) — latent gap · Metrics

### Incumbent in

- [WTW ResQ](/Products/WTW_ResQ) — incumbent in · Products
- [Moody's AXIS](/Products/Moody's_AXIS) — incumbent in · Products
- [Oliver Wyman Actuarial](/Products/Oliver_Wyman_Actuarial) — incumbent in · Products
- [Excel Actuarial Templates](/Products/Excel_Actuarial_Templates) — incumbent in · Products
- [FIS Prophet](/Products/FIS_Prophet) — incumbent in · Products
- [In-House Python Models](/Products/In-House_Python_Models) — incumbent in · Products
- [Milliman Consulting](/Products/Milliman_Consulting) — incumbent in · Products

### Applies thesis

- [Insurance Carrier](/CompanyTypes/Insurance_Carrier) — applies thesis · CompanyTypes

### Embodies

- [Software](/Theses/Software) — embodies · Theses

### Similar Opportunities

- [Synthetic Actuarial Analyst](/Industries/Finance_and_Insurance/Opportunities/Synthetic_Actuarial_Analyst) — similar · Opportunities
- [Actuarial Reserve API](/Knowledge/Mathematics/Opportunities/Actuarial_Reserve_API) — similar · Opportunities
- [Algorithmic Claims Administrator](/Industries/Finance_and_Insurance/Opportunities/Algorithmic_Claims_Administrator) — similar · Opportunities
- [Valence Claims](/Opportunities/Valence_Claims) — similar · Opportunities
- [AI Claims Adjudication for Insurance Carriers](/Opportunities/AI_Claims_Adjudication_for_Insurance_Carriers) — similar · Opportunities
- [Litigation Underwriting Engine](/Occupations/Legal_Occupations/Opportunities/Litigation_Underwriting_Engine) — similar · Opportunities
- [Autonomous Claims Desk](/Opportunities/Autonomous_Claims_Desk) — similar · Opportunities
- [Automated Claim Appraisal](/Opportunities/Automated_Claim_Appraisal) — similar · Opportunities
- [Incident Loss Forecaster](/Opportunities/Incident_Loss_Forecaster) — similar · Opportunities
- [AI Medical Claims Adjudication for Carriers](/Opportunities/AI_Medical_Claims_Adjudication_for_Carriers) — similar · Opportunities
- [Liability Audit Engine](/Opportunities/Liability_Audit_Engine) — similar · Opportunities
- [AI Claims Adjudication for Health Insurers](/Opportunities/AI_Claims_Adjudication_for_Health_Insurers) — similar · Opportunities
- [AI Subrogation Investigator](/Opportunities/AI_Subrogation_Investigator) — similar · Opportunities
- [Workers Comp Engine](/Opportunities/Workers_Comp_Engine) — similar · Opportunities
- [Warranty Settlement Engine](/Opportunities/Warranty_Settlement_Engine) — similar · Opportunities
- [Runoff Compliance Automation](/Opportunities/Runoff_Compliance_Automation) — similar · Opportunities
- [Freight Liability Oracle](/Opportunities/Freight_Liability_Oracle) — similar · Opportunities
- [AI Rule Validator](/Skills/Systems_Analysis/Opportunities/AI_Rule_Validator) — similar · Opportunities
- [Property Damage Appraiser](/Opportunities/Property_Damage_Appraiser) — similar · Opportunities
- [Dynamic Freight Risk Engine](/Occupations/Transportation_and_Material_Moving_Occupations/Opportunities/Dynamic_Freight_Risk_Engine) — similar · Opportunities