Loss History Report

There is no formal loss history report. The underwriter asks the agent 'has this account had any claims?' and trusts the verbal answer. Prior loss information is anecdotal — 'I think they had a fire a few years back' — with no verification against any database.

None — AI cannot perform predictive loss modeling because no prior loss records exist in any system.

Loss history reports are ordered from CLUE or A-PLUS and arrive as PDF documents. The underwriting assistant attaches the PDF to the file. The underwriter reads the report manually, scanning for large claims or suspicious patterns. Each line of business pulls loss history from a different source — personal lines from CLUE, commercial from the carrier's own claims system, specialty from broker-provided loss runs. Comparing loss patterns across sources requires side-by-side manual review.

AI could OCR the PDF loss reports, but the varying formats across sources (CLUE, A-PLUS, broker loss runs) make reliable structured extraction inconsistent.

Loss history records are stored as structured fields — claim date, claim type, incurred amount, and claim status — parsed from CLUE/A-PLUS reports and internal claims records. The underwriting workbench displays a summary table of prior losses. Sorting by claim type, size, or date range is possible. But external loss reports and internal claims records are stored separately — the full picture requires checking both sources manually.

AI can flag high-frequency or high-severity loss patterns from the structured loss history table. Cannot provide a complete loss picture because external loss reports and internal claims records are not unified.

Loss history is a unified, consolidated record per insured combining external reports (CLUE, A-PLUS) and internal claims records. Duplicate entries are reconciled. Each loss event links to the original source. An underwriter can query 'show me all property losses for this insured exceeding $50K in the last five years with their current reserve status' and get a single, reliable answer without checking multiple systems.

AI can perform predictive loss modeling using the consolidated loss history — identifying trends, predicting future loss frequency, and flagging accounts that deviate from expected loss patterns for their risk class.

Loss history records are schema-driven with formal entity relationships. Each loss event links to the policy that responded, the specific location or unit involved, the cause-of-loss code, and the reserve development timeline. An AI agent can query 'across all accounts in this agency's book, what is the correlation between roof age and water damage claim frequency for frame construction properties?' and get a computed, evidence-backed answer.

AI can perform fully autonomous loss analysis — portfolio-level trend identification, account-level loss prediction, and cross-variable correlation analysis. Autonomous underwriting decisions for loss-history-dependent risks.

Loss history is a living record that updates in real-time. New claims events post immediately. Reserve adjustments propagate within minutes. Subrogation recoveries and salvage update the loss profile as they occur. The loss history report is never stale — it reflects the current state of every loss event as it evolves through its lifecycle.

Fully autonomous loss history management. AI operates on continuously current loss profiles that reflect every claims event in real-time.

Ceiling of the CMC framework for this dimension.