AI AGENT INTELLIGENCE · INSURANCE
The report that tells your board whether your insurer's AI agent is safe to sell and settle claims
Lexic Compass audits insurance conversational agents on product recommendation, claims handling, and risk pricing — and delivers a Trust Score from 0 to 100 with an independent verdict.
An insurance conversational agent audited by Lexic Compass receives a Trust Score from 0 to 100 across 4 pillars — Integrity & Safety, Regulatory Trust, Operational Reliability, and Experience Trust — with a compliance pack specific to insurance that verifies the IDD demands-and-needs test before recommending a policy, consistent and non-discretionary handling of claims decisions, and the risk that life or health insurance scoring and pricing falls under High-Risk classification under Annex III of the EU AI Act — the only vertical where the Regulation names this use case explicitly. The result is an independent report/verdict — Cleared, Cleared with Conditions, or Not Cleared — not a certification: Lexic is not a notified body under the EU AI Act.
What the Insurance pack covers
| Regulation | What the Compass report verifies |
|---|---|
| IDD (Insurance Distribution Directive) | Demands-and-needs test documented before recommending, comparing, or selling a policy |
| Annex III — explicit case | Life/health insurance risk scoring and individual pricing is High-Risk, named directly by the Regulation itself |
| Claims handling — market conduct | Consistent, traceable criteria; no full agent autonomy in denying a claim |
| Fairness in pricing/scoring | No use of proxies for protected characteristics in risk assessment |
| DORA | Incident traceability classifiable for the customer's ICT register |
| Vulnerable customer handoff | Real escalation to a human for health or bereavement-related claims |
“In insurance, the agent that recommends your policy and the agent that denies your claim are held to completely different legal standards — and most companies are only auditing one of them.”