Module Boundaries: Verdict-Heavy vs Settlement-Anchored Data

TL;DR — Verdict and judgment data belong in a tail-risk module. Settlement data belongs in a clearing-price module. Verdicts are useful for trial severity; they are the wrong training target for settlement value. Mixing the two distributions corrupts the forecast.

Jury verdicts are a useful record of what happens when a claim is tried. They show trial severity, venue behavior, and the tail of exposure. They are the wrong target for a settlement value. Most litigation data providers sell scraped public dockets: verdicts, trial judgments, and motion outcomes. That corpus is valuable for modeling trial risk. It is not a settlement sample. When engineers ingest it without module boundaries, the model treats a public fight as a private compromise. Two distributions go into one training set. The forecast for routine claims drifts up. The public trial record is selected for disagreement, not for typical resolution. The vast majority of claims still resolve privately. Mixing those realities in a single store produces a warped expected value.

The cases that reach a jury are the ones where plaintiff and defense fundamentally disagree on the facts or the liability probability. That is Priest-Klein selection bias. Routine cases settle early because the clearing price is obvious to both sides. A model trained on a blended set of settlements and verdicts learns the wrong object. It pulls expected settlement value up. Verdict data is not empty of signal. The signal is about trial risk, not the settlement number. You cannot average a private compromise with a public fight and treat the result as a clearing price.

The Architecture of Separation

We enforce strict module boundaries between verdict-heavy data and settlement-anchored data. At Canotera, our generative AI layer has one highly specific job: read the raw case file. It ingests the pleadings, the medical records, and the correspondence. It processes thousands of pages per claim to structure the messy, unstructured facts. It identifies the injuries, extracts the chronological timeline, and isolates the legal theories. It does not predict the financial outcome. We isolate the reading comprehension task from the mathematical forecasting task. This strict separation of concerns allows us to process highly sensitive medical records securely, stripping out personally identifiable information before the structured data ever reaches the predictive layer.

The prediction happens in separate, purpose-built mathematical models trained exclusively on resolved cases with known outcomes. The settlement module operates strictly on private, resolved claims data. This data is inherently sparse because settlements are highly confidential. Triangulating a clearing price requires mapping the specific drivers of a new claim against the geometric space of these known, resolved settlements. We represent claims as high-dimensional vectors where each dimension corresponds to a material fact extracted by the generative AI. The model outputs a calibrated settlement range, not a single point guess. This range reflects honest uncertainty bands based on the density of comparable settled cases in that specific region and injury profile. If the local space is sparse, the uncertainty band widens organically.

The verdict data lives in its own isolated module. We use it to model tail risk, trial severity, and what juries actually do in a venue. When a claim shows signs of escalation, we query that module for the ceiling of exposure. A jury award is rarely the final amount paid. Hyman and other researchers have demonstrated that post-verdict settlements, policy limits, collectability constraints, and appeals asymmetry drastically reduce the actual cash transfer. Building a system that treats a fifty million dollar verdict as a fifty million dollar payout is a fundamental engineering failure. We model the verdict as a distinct event probability, entirely separate from the expected baseline settlement payout.

Triangulation and Honest Uncertainty

Maintaining these boundaries requires rigorous data hygiene during the initial ingestion phase. Every resolved case we use for training must be cleanly tagged by its resolution mechanism. Did this specific claim resolve via early settlement, post-discovery mediation, or trial? Our API exposes these distinctions directly to the claims workflow. When an adjuster reviews a reserve delta generated by Canotera, they see exactly which comparable resolved cases drove the calculation. The specific drivers behind each number are fully traceable to the source documents. If the escalation probability is high, the system surfaces the verdict-heavy comps to justify the warning, rather than silently inflating the baseline settlement range. This traceability eliminates the black-box reserve volatility that plagues legacy predictive models.

Running distinct models in parallel introduces strict latency and compute constraints. Reading thousands of pages of unstructured medicals and pleadings requires heavy compute resources. We decouple the document ingestion pipeline from the predictive queries to solve this. The generative AI processes the file asynchronously, updating the structured claim record continuously as new correspondence arrives. The geometric machine-learning models run low-latency inferences against that structured record. This asynchronous architecture ensures that claims teams get immediate updates to their settlement ranges without waiting for a monolithic system to re-read the entire file. It also preserves strict tenant isolation, keeping sensitive client data completely walled off from the aggregated mathematical models.

Industry forces like social inflation and third-party litigation funding break static models. Verdicts are climbing significantly faster than historical economic inflation. If settlement and verdict data bleed together, it becomes mathematically impossible to track the velocity of this change. Isolating the verdict module is how we keep that signal usable. We can measure how nuclear verdicts are pulling up the settlement floor in specific jurisdictions over time. We track the drift independently. The models adapt to shifting liability probabilities and rising defense costs without contaminating the baseline settlement data. Claims teams can allocate defense spend accurately and negotiate from hard data, knowing whether a plaintiff demand is priced off comparable settlements or off trial risk that belongs in a separate module.

Building a reliable forecasting platform for insurance claims means respecting the underlying structure of the legal system. You have to model the reality of the dispute. This requires acknowledging that trials and settlements are completely different financial mechanisms driven by completely different incentives. Verdict data is how you see the fight. Settlement data is how you price the compromise. The engineering architecture must keep those modules apart.

Bad architecture averages the extremes; good architecture maps the boundaries.