JustineAI™ — Mass Tort Edition.
MDL specialists. One Justine supervisor coordinates thousands of plaintiff sub-agents in a single context — bellwether scoring, settlement-matrix modeling at scale, lien negotiation, plaintiff fact-sheet auto-population. The supervisor + sub-agent pattern is the moat.
MT runs on the full Eve-Legal Operating System — case management, deadlines, conflict check, trust accounting, time tracking, native e-signature (ESIGN / UETA), the firm-branded public intake portal, and Microsoft 365 · Dropbox integrations (plus a bring-your-own DocuSign connection). The OS layer is shared across every edition; the legal-reasoning calibration is what changes.
This edition is not yet generally available. The page describes the workload it will serve and lets mass tort attorneys register interest. We don’t publish dates — we publish the architecture and reach out when the edition is ready.
Nine stages. One Justine. The full mass tort lifecycle.
This is the lifecycle JustineAI™ MT is designed to run end to end — the same supervisor pattern that runs personal injury today, recalibrated for mass tort practice. Every stage will be auditable, every output attorney-attested work product.
- Stage 01Inventory intake & vetting
When the MT edition opens, Justine will triage the full plaintiff inventory at acquisition — product exposure, injury type, use dates, prior counsel, lien posture, statute and tolling-agreement status per jurisdiction. Marginal and time-barred claims get flagged before they enter the MDL workup.
- Stage 02Plaintiff fact sheets
One supervisor will auto-populate each plaintiff fact sheet from intake records, medical chronologies, and prior productions — every field with per-source provenance. The validator council confirms exposure and injury entries, surfacing cross-source conflicts for attorney resolution before the PFS is attested and served.
- Stage 03Common-issue discovery
Justine is designed to read the common-benefit document depository — internal corporate memos, regulatory submissions, adverse-event reports — in one context, surfacing the general-causation and notice evidence that recurs across the pool rather than re-deriving it plaintiff by plaintiff.
- Stage 04Bellwether selection & scoring
The bellwether sub-agent will compose PFS density, injury severity, jurisdiction risk, and settlement leverage into a defensible ranking. When CMO 1 sets the trial-selection protocol, the firm walks in with a representativeness argument grounded in the actual inventory, not intuition.
- Stage 05Science & Daubert
Justine is designed to assemble the general-causation record — epidemiology, dose-response, mechanism — against each defense expert’s prior testimony and methodological gaps, producing a Daubert pre-read. Every cited authority is grounded against a verifiable source, never invented.
- Stage 06Bellwether trial workup
For each selected trial pick, the supervisor will spin a dedicated case workup — complaint refinement, plaintiff-specific causation, deposition designations, motions in limine, exhibit assembly — while keeping every other plaintiff sub-agent in the same reasoning context for cross-matter consistency.
- Stage 07Global settlement matrix
Justine will model the inventory settlement as a tiered matrix — injury grids, qualifying criteria, base-and-enhancement point schedules — with per-tier values anchored to public verdicts where available. The matrix exports as a procurement-ready spreadsheet the negotiating committee can defend.
- Stage 08Allocation & point scoring
When a settlement is reached, the supervisor is designed to score every plaintiff against the agreed matrix — documenting tier placement, enhancements, and offsets per claimant with provenance — so the special master and claims administrator receive a fully reasoned allocation the firm can stand behind.
- Stage 09Lien resolution & distribution
The lien sub-agent will orchestrate Medicare, Medicaid, ERISA, provider, and subrogation liens across the entire pool — computing each plaintiff’s net before distribution, not after. Disbursement statements and an auditable disbursement trail are prepared for every claimant.
One Justine. Sub-agents built for mass tort attorneys.
The MT edition will run the same architecture as PI: one Justine coordinating stage-specialized sub-agents across the matter, reasoning in a single context. The sub-agents aren’t separately branded — they’re facets of the Digital Employee. These are the ones this edition is designed around.
- Inventory sub-agent
Reasons over the entire plaintiff pool at once: product-exposure proof, injury cohorting, use-date windows, jurisdiction and venue, statute and tolling-agreement status, prior-counsel and referral provenance, duplicate and lien-conflict detection across thousands of claimants.
For example — An inventory of 4,200 acquired claims in a hernia-mesh MDL. The inventory sub-agent will cohort them by implant model and revision-surgery status, flag 137 with no implant record on file, surface 19 whose two-year statute ran before any tolling agreement attached, and identify 6 already represented in a parallel state consolidation — each finding shown with the source page that triggered it, for the intake attorney to clear before workup.
- Bellwether sub-agent
Reasons over selection-protocol fitness: plaintiff fact-sheet completeness, injury severity, treatment and causation density, jurisdiction risk, defense exposure, demographic and geographic representativeness against the pool, and the CMO’s randomization or strike rules.
For example — From a 4,200-plaintiff inventory, CMO 1 calls for each side to designate eight trial picks from a representative pool. The bellwether sub-agent will score the inventory, propose a 40-plaintiff candidate set that mirrors the pool on implant model, injury grade, and venue, and explain why its slate is more representative than a naive severity sort — handing the steering committee a Rule-defensible selection argument the special master can scrutinize.
- Settlement-matrix sub-agent
Reasons over inventory-wide valuation: injury-tier definitions, qualifying medical criteria, base point schedules, enhancement and offset factors, public-verdict comparables, holdback and common-benefit assessments, and the global-deal participation threshold.
For example — A proposed inventory settlement with a $1.2 billion common fund. The settlement-matrix sub-agent will model a five-tier injury grid, assign base points by surgery status with enhancements for documented complications and offsets for pre-existing conditions, anchor each tier to public bellwether verdicts where available, and project the per-tier average recovery and the firm’s blended fee — exporting it as a spreadsheet the negotiating committee attests before presenting it to the court.
- Lien sub-agent
Reasons over recovery-side obligations across the pool: Medicare conditional payments and MSA exposure, Medicaid state recovery, ERISA-plan subrogation and Ahlborn reductions, provider balances, and per-plaintiff net-to-client computation before distribution.
For example — A 4,200-plaintiff settlement with mixed payer profiles. The lien sub-agent will batch-query CMS for conditional-payment ledgers, separate injury-related charges from unrelated treatment, assert procurement-cost reductions on each ERISA plan, and compute the net distribution for every claimant before a single check is cut — flagging 84 plaintiffs whose Medicare exposure exceeds their tier recovery so the attorney can intervene before, not after, disbursement.
The MT workload — described before it ships.
What follows is the architectural commitment for this edition — the workload Eve-Legal F5/reasoner will carry once we open it. The reasoning core is shared with PI. The calibration, taxonomy, and document pipeline are MT-specific.
- Supervisor over thousands of plaintiff sub-agents
One Justine coordinates plaintiff fact-sheet drafting, intake triage, and discovery prep across the MDL’s entire plaintiff population — in a single reasoning context.
- Bellwether scoring
Plaintiff fact-sheet density, injury severity, jurisdiction risk, settlement leverage — composed into a bellwether ranking the firm can defend in CMO 1.
- Settlement-matrix modeling
Per-tier point-value modeling for inventory settlement. Comparables anchored to public verdicts where available. The matrix exports as a procurement-ready spreadsheet.
- Lien negotiation at scale
Medicare, Medicaid, ERISA, provider, and subrogation liens orchestrated across the plaintiff pool. The lien total per plaintiff is computed before settlement, not after.
- Plaintiff fact-sheet auto-population
PFSs drafted from intake records, medical chronologies, and prior productions. Attorney-attested before service.
The same depth of reasoning — on mass tort matters.
We proved the reasoning on personal injury. Here is what the same compositional fabric is designed to surface once the MT edition opens — the findings a mass tort practice would want caught, each presented with its evidence for the attorney to weigh.
- Pool-wide triage
Justine will surface — On a typical inventory acquisition, Justine will read every intake file in one context and surface the claims that don’t belong — the plaintiffs with no proof of product exposure, the statutes that ran before a tolling agreement attached, the duplicates already pending in a parallel consolidation — each tied to the source page, so the firm vets the inventory before it funds the workup.
- Common causation
Justine will surface — Across a common-benefit depository of internal memos, regulatory filings, and adverse-event reports, Justine is designed to surface the handful of documents that establish corporate notice and general causation for the whole pool — the date the defendant first knew, the buried safety signal — reasoned once and reused across thousands of plaintiff sub-agents rather than re-discovered file by file.
- Bellwether representativeness
Justine will surface — When CMO 1 sets the trial-selection protocol, Justine will show why a proposed slate of picks does or doesn’t mirror the inventory — flagging that the defense’s eight nominees skew toward the mildest injury grade and the most defense-friendly venues — and propose a representative counter-slate, with the distribution math the steering committee can defend to the special master.
- Settlement-matrix modeling
Justine will surface — On an inventory deal, Justine is designed to model the injury-tier grid end to end — projecting the per-tier average recovery, the share of the fund consumed by the most severe cohort, and the blended attorney fee after common-benefit holdback — anchored to public bellwether verdicts where they exist, so the negotiating committee sees the consequences of each tier definition before agreeing to it.
- Lien exposure at scale
Justine will surface — Before distribution, Justine will compute each plaintiff’s net across the entire pool and surface the claimants whose Medicare, ERISA, or provider liens exceed their matrix recovery — the cases that net to zero or below — so the firm renegotiates those liens or revisits tier placement before checks issue, not in a post-distribution scramble.
- Single-context consistency
Justine will surface — Because the supervisor holds the whole matter in one reasoning context, Justine is designed to catch what fragmented review misses — a causation theory argued for the bellwethers that contradicts a PFS allegation in the broader pool, an expert opinion that helps one tier and undercuts another — keeping thousands of plaintiff sub-agents internally consistent for the attorney to ratify.
The AI reasons; the attorney decides.
Tell us about your MT practice.
We onboard new editions in waves. When MT opens, founding-firm slots and beta access go to attorneys who told us about their practice early. The form tags your inquiry as MT automatically — you don’t need to repeat the practice area.
Solo and small MT practices: we will reach out the moment self-serve opens. Mid-size and enterprise: we will schedule a discovery call in the wave-prep window before general availability.