Independent regulatory scrutiny auditing 11 market-leading claims automation vendors against NAIC Model Bulletins, state adjuster licensing statutes, ERISA due process, and unfair claims settlement practices acts.
11Vendors Audited
5Core Categories
4Statutory Risk Vectors
100%Zero Vendor Sponsorship
The Non-Delegation Doctrine
Under Section 1 of the NAIC Model Bulletin on Artificial Intelligence Systems and New York DFS Circular Letter No. 7, an insurance carrier remains directly and non-delegably liable for any statutory violations, unfair settlement practices, or unlicensed adjusting conduct caused by third-party vendor software.
Filter Audited Platforms
Search by vendor name, platform features, or filter by claim operation category.
Showing 11 of 11 Audited Platforms
No matching vendor audits found
No commercial software systems matched your current search query or category filter. Try clearing your search.
Primary Operational Function: Generates personal injury demand packages, medical chronologies, and claimed damages for plaintiff law firms.
Regulatory & Compliance Dimension
Operational Finding & Statutory Scrutiny
Evidentiary Accuracy & Hallucinations
Automated demand generators frequently introduce phantom injuries, mischaracterize chronic degenerative findings as acute trauma, and double-count medical billing. Defense examiners must implement page-by-page Bates verification.
Algorithmic Negotiation Anchor
Settlement databases generate artificial anchors. Insurers facing EvenUp demands risk bad faith if they rely on black-box counter-algorithms instead of verifiable document reconciliations.
Primary Operational Function: Workers' compensation and casualty document summarization, medical chronology generation, and workflow triage.
Regulatory & Compliance Dimension
Operational Finding & Statutory Scrutiny
Licensing & Discretionary Adjusting
To remain compliant with state adjuster licensing statutes (e.g., Cal. Ins. Code § 14021), platforms summarizing workers' comp files must operate strictly as clerical assistance. Any automated recommendation regarding disability ratings or settlement reserves triggers statutory adjusting oversight.
Explainability Standards
When adjusters use AI chronologies to justify indemnity reserve changes, the chronology must provide direct visual links to the underlying doctor's narrative to survive state workers' compensation board audits.
Primary Operational Function: Machine learning platform predicting disability and workers' compensation claim durations and nudging adjusters toward early return-to-work closures.
Regulatory & Compliance Dimension
Operational Finding & Statutory Scrutiny
ERISA § 503 Full & Fair Review
Terminating or modifying employee disability benefits based on statistical cohort duration scores rather than individualized clinical proof creates substantial arbitrary-and-capricious liability under 29 U.S.C. § 1133.
Reserve Steering Bad Faith
Adjuster notes reflecting automated suggestions to cut reserves or initiate IME surveillance within hours of an AI alert provide powerful evidence in bad-faith litigation.
Primary Operational Function: Predictive casualty AI scoring claim litigation risk, profiling medical providers, and ranking plaintiff attorneys.
Regulatory & Compliance Dimension
Operational Finding & Statutory Scrutiny
Bad-Faith Attorney Profiling
Adjusting settlement offers based on plaintiff counsel's perceived trial willingness violates the carrier's affirmative duty to evaluate claims solely on their individual factual merits under state UCSPA statutes.
Discoverable AI Prompts
Examiner workflow screen logs and algorithmic prompts are 100% discoverable in civil litigation, creating severe exposure during adjuster depositions.
Initial photo estimates regularly underestimate physical repairs by 35% to 60%, shifting the administrative burden of filing supplements onto consumers and collision repair shops.
Prompt Payment Penalties
Supplement delays on closed virtual claims risk mandatory statutory interest penalties under Texas Ins. Code § 542 and Florida Stat. § 627.70131.
Primary Operational Function: Dominant collision estimating platform automating line items, repair-vs-replace scoring, and pre-teardown total loss thresholds.
Regulatory & Compliance Dimension
Operational Finding & Statutory Scrutiny
Anti-Steering (Cal. § 758.5 / Tex. § 1952.301)
Automated substitution of OEM components with Like Kind & Quality (LKQ) parts without required written policyholder disclosure violates state anti-steering mandates.
Visual Estimating Discrepancy
2D smartphone photo AI models under-estimate hidden structural and ADAS calibration damage by an average of 34.6% compared to physical teardowns.
Primary Operational Function: High-volume medical bill review and fee-schedule repricing engine in workers' compensation and auto PIP.
Regulatory & Compliance Dimension
Operational Finding & Statutory Scrutiny
Silent PPO Discount Stacking
Routing provider bills through secondary discount networks without direct contractual privity violates California Labor Code § 4609 and Texas Labor Code § 413.011.
Automated CPT Downcoding
Algorithmic truncation of Evaluation & Management (E&M) codes without medical director chart review violates administrative due process and medical fee dispute rules.
Touchless claim rejections without an individualized investigation by a licensed human adjuster breach non-delegable insurer duties and state adjusting statutes.
Biometric & Video Profiling Risks
Scoring claimant video statements for deception risks severe civil liability under Illinois BIPA and Colorado SB 21-169 unfair discrimination rules.
Primary Operational Function: Generates automated personal injury liability assessments, settlement demand letters, and litigation valuation metrics.
Regulatory & Compliance Dimension
Operational Finding & Statutory Scrutiny
Unauthorized Practice of Law (UPL)
Deploying generative AI algorithms to draft substantive legal arguments and evaluate claim settlement ranges for non-lawyer claims staff risks state bar unauthorized practice of law sanctions under ABA Model Rule 5.5.
Rule 11 Evidentiary Foundation
Settlement demand packages containing unverified algorithmic multipliers face immediate sanctions under Fed. R. Civ. P. 11 and state bad-faith discovery rules.
Primary Operational Function: Machine learning fraud detection platform analyzing claim networks, social graphs, and behavioral markers to generate automated fraud risk scores and SIU referral flags.
Regulatory & Compliance Dimension
Operational Finding & Statutory Scrutiny
Disparate Impact & Proxy Discrimination
Utilizing non-traditional data markers and machine learning graph networks without verifiable bias audits violates NY DFS Circular Letter No. 7 and Colorado SB 21-169 governance mandates.
Unfair Claim Settlement Delay
Automatically freezing, delaying, or suppressing claim payments based on automated fraud likelihood scores without independent human SIU corroboration constitutes bad-faith claims delay under state UCSPA statutes.
REGULATORY DILIGENCE BENCHMARK
Comparative Compliance Scorecard for Insurers
The following mandatory diligence framework outlines the baseline statutory questions carrier technology and compliance committees must require every claims AI software vendor to verify before enterprise procurement:
Vendor Diligence Requirement
Statutory Authority
Compliant Architectural Standard
Adjuster Discretion
State Licensing Codes (Cal. § 14021, Tex. § 4101)
Zero automated valuations or denial mandates. Software assists with evidence extraction; licensed human adjusters retain 100% discretionary evaluation authority.
Explanatory Provenance
NAIC Model #900 & State UCSPA
Every single date, dollar figure, and diagnostic code links directly to an underlying Bates-numbered document page in the claim file.
Algorithmic Fairness
Colorado SB 21-169 & NY DFS Circular 7
No black-box regional multipliers or demographic discounting models. Transparent, statistically reproducible factual reconciliation.
Data Custody & Privacy
NAIC Model #668 & HIPAA Privacy Rule
Zero data egress to multi-tenant public APIs. Processing remains inside single-tenant, isolated infrastructure with zero commercial LLM retention.
Research & Editorial Methodology
This report was authored by the Claims Governance Institute Research Group. CGI is an independent, non-partisan research monitor examining artificial intelligence, algorithmic accountability, and regulatory compliance across the insurance and legal-tech sectors.
Our analysis relies strictly on publicly available documentation, statutory frameworks (including state insurance codes, NAIC model acts, and HIPAA guidelines), and technical audits. Read our Editorial Standards & Conflict Disclosures.