AI claims litigation
The two cases that make the AI-claims risk concrete — they are the loss events the Glacis Tech-E&O product would respond to, and the disputes the TPA / claims- attribution stream is built to answer. Both grounded in the Law360 article. GROUNDED · Law360
Why these matter
Section titled “Why these matter”The NAIC AI Model Bulletin’s enforcement leans on unfair trade practice and unfair claim settlement practice statutes (see 00 NAIC AI Model Bulletin). These two lawsuits are real-world tests of exactly those theories applied to AI — “the same issues that concern regulators.” GROUNDED · Law360
Case 1 — Kisting-Leung v. Cigna Corp.
Section titled “Case 1 — Kisting-Leung v. Cigna Corp.”| Field | Detail (per the article) |
|---|---|
| Court | U.S. District Court, Eastern District of California |
| Status | The court allowed the suit to proceed (March 2025) |
| Allegation | Cigna used AI to deny medical claims without proper review |
| Theory | Improper AI-driven claim denial (unfair claims practice) |
This is a health-claims dispute — directly relevant to the healthcare-AI vertical of the Glacis product. It’s the canonical example of “wrongful AI claim denial” as an insurable exposure. See 07 Tech E&O for healthcare AI. GROUNDED · Law360
Case 2 — Kelly v. State Farm Fire & Casualty Co.
Section titled “Case 2 — Kelly v. State Farm Fire & Casualty Co.”| Field | Detail (per the article) |
|---|---|
| Court | U.S. District Court, Middle District of Alabama |
| Filed | October 2025 |
| Allegation | Homeowners allege State Farm used “cheat and defeat AI algorithms” as discriminatory tools in claims processing that “disproportionately impact[ed] Black and non-white policyholders” |
| Theory | Algorithmic discrimination in claims handling |
This is the canonical algorithmic-discrimination / bias case — the same fairness concern that drives the NAIC’s bias-testing expectations and the risk-classification guardrail. GROUNDED · Law360
The two theories, side by side
Section titled “The two theories, side by side”| Case | Theory |
|---|---|
| Kisting-Leung v. Cigna | AI DENIED claims WITHOUT proper review (improper denial / unfair claims practice) |
| Kelly v. State Farm | AI claims algorithm was DISCRIMINATORY (algorithmic bias / unfair discrimination) |
Together they cover the two big AI-claims failure modes: wrongful denial and biased outcomes — exactly the exposures a healthcare-AI Tech-E&O policy underwrites.
How attribution answers them (the TPA tie-in)
Section titled “How attribution answers them (the TPA tie-in)”Both cases hinge on what the AI actually did and whether it was reviewed/ controlled. That’s precisely what cryptographic attribution of AI decisions makes answerable — turning “we believe the model did X” into “here is the attested record of what the model did, and the human review applied.” GROUNDED · Glacis
This is the foundation of Glacis revenue stream #3 (TPA / claims error-source attribution) — see 07 TPA & claims attribution — and it dovetails with the NAIC push toward documentation of model origins and explainability in 01 AI evaluation tool & third-party WG. GROUNDED · Law360
Honest-sourcing note (VERIFY case status)
Section titled “Honest-sourcing note (VERIFY case status)”The article is a December 2025 snapshot, read here in mid-2026. Litigation moves — do not treat the procedural posture as current.
→ 15 Glacis strategic arc: The three revenue streams
Drill this page →5 bank questions stand behind what you just read. Check it while it’s warm.