Compare · Predict vs Carrier-side AI

The carriers built theirs in 2014. Now we have ours.

Carriers have run claims-side ML on their internal settlement data for over a decade. By 2019, every major US property-casualty insurer was setting reserves with a model. Carriers price your cases with AI. Now you can too.

Predict

Carrier-side claims AI

Who sees it

Plaintiff attorneys only — contractually

Internal to the carrier; adjusters and reserve teams

Training data

Over 20,000 precedent cases, 200,000 data points — plaintiff outcomes

The carrier's own claims history — paid losses, reserve setting

What it optimizes

Estimated fair value to the plaintiff, with confidence bands

Reserve adequacy and combined-ratio targets — the carrier's economics

Transparency

You can read how it works and the cohort behind every number

"Proprietary model" — opaque to the claimant and to plaintiff counsel

Confidence band

Always shown — ± dollar range, 90% CI, sample size

Internal to the model; rarely surfaced in settlement memos

Recalibration

Refreshed on a regular cadence as new verdicts land

Internal cadence; not disclosed to the other side

Side of the table

Plaintiff-side only by design — we will never sell to carriers

Carrier-only by design — the asymmetry is the product

What you do with it

Defensible counter-anchor; exports a full demand letter in your firm's style

Sets the reserve and the opening offer — the floor of the negotiation

Over 20,000 precedent cases and 200,000 data points across MVA and premises liability. On a held-out test set, Predict reaches 90–92% median accuracy (MdAPE) against the realized settlement on higher-value cases, each shown as a gross value (before attorney fees, case costs, and liens) with a 90% confidence band. See the accuracy results →

Same kind of instrument. Opposite side of the table.

In plain terms: the carrier shows up to negotiation with a number from a model. Until now you showed up with comps and instinct. Predict gives you a number from a model built on plaintiff outcomes, with the same statistical rigor, so the table is even.

The deliberate differences are upstream of the model. The training data is plaintiff outcomes, not carrier paid losses. The side of the table is fixed by the brand commitment, not by the buyer — we will never sell to a carrier. And every number arrives with the comparable cohort it was built on, so you can read the work instead of taking it on faith.

For 12 years the plaintiff bar priced cases with Verdict Search and institutional memory while the carrier opened with a modeled number. Predict gives you a model of your own: a sourced, confidence-banded number to counter-anchor at demand and at the negotiation table, carried the same way from intake through settlement.

See how it works →

Illustrative example

$185,000

± $28,000 · 90% CI

$157K Median $213K

Jurisdiction-tuned to Harris County, moderate-severity rear-end MVA, banded to a 90% CI on its comparable-verdict cohort. This is the plaintiff-side number to bring to the table.

TX · HARRIS · MVA

Carriers have priced cases like this against an internal model since the 2010s. The negotiation used to start with their number setting the anchor. Now both sides bring an instrument to the table.

Bring a model to the next negotiation.

Predict a current case and see what a defensible counter-anchor looks like: confidence-banded, jurisdiction-tuned, sourced to the comparable-verdict cohort.

Enter the jurisdiction, injury type, and basic case facts. In about 60 seconds you get a jurisdiction-tuned gross case value — before fees, costs, and liens — with its confidence band and the comparable-verdict cohort it is sourced to.

A 14-day free trial gives you a defensible, confidence-banded number on every case in your pipeline — from intake through demand prep to negotiation. $499 per seat / mo, cancel anytime in one click. See pricing.

Predict your case Why we built this →