ClauseCast
Private litigation screening intelligence

Screen litigation before committing to full diligence.

ClauseCast helps litigation funders and ATE insurers assess English breach-of-contract disputes at intake. Provide the information available today, resolve decision-critical gaps, and receive a structured statistical assessment to help prioritise deeper professional review.

Private pilots · English breach-of-contract scope · Professional decision-support

The screening problem

Too much diligence can begin before the case has earned it.

Early referrals arrive with different levels of preparation, evidence and factual clarity. ClauseCast is designed for the decision before full diligence: does this matter warrant deeper investigation?

01 · Intake

Incoming matters vary enormously.

Initial referrals can be incomplete, inconsistently structured and supported by very different levels of documentation.

02 · Resource

Professional review is valuable.

Counsel, underwriting and investment-team time should be concentrated where the available record justifies deeper work.

03 · Consistency

Material gaps can surface late.

Different reviewers can focus on different information, while a missing fact may not become decision-critical until significant work has already started.

ClauseCast provides a structured first-pass signal so professional teams can prioritise deeper diligence without pretending the first screen is the final decision.

How ClauseCast works

From intake to a structured first-pass assessment.

The workflow is designed around information available at initial screening rather than a fully prepared litigation bundle.

01

Submit what you have.

Describe the dispute in plain language, provide appropriately authorised supporting documents, or use both routes.

02

Resolve material gaps.

ClauseCast identifies decision-critical missing information and asks focused questions rather than silently treating unknown facts as neutral.

03

Review the assessment.

Receive a statistical estimate together with information sufficiency, decision sensitivity, supporting and adverse considerations, key uncertainty and the issue most likely to change the result.

Built for: litigation funders, ATE insurers and professional case-screening teams.Role: supplements legal, underwriting and investment judgement.
The output

More useful than a headline probability.

A screening number is only useful if reviewers can understand its scope, uncertainty and what information is driving it. ClauseCast keeps those elements visible.

ClauseCast evaluation workspace showing a fictional breach-of-contract scenario and a targeted clarification question
Fictionalised product example. Interface and model performance remain subject to testing and validation.
Estimated success rangeA statistical screening range against the current prediction target.
Information sufficiencyWhether the supplied record is adequate for the current assessment.
Decision sensitivityWhether additional information could materially alter the result.
Supporting and adverse factorsA structured summary of information currently weighing for and against the claim.
Decision-critical issueThe factual or evidential issue most likely to change the assessment.
Evidence refinementMaterial new facts or documents can trigger a complete reassessment.
Professional workflows

Built for the first screening decision.

ClauseCast is deliberately narrower than a general legal assistant. It is designed to support professional triage and prioritisation in defined commercial-litigation workflows.

Litigation funders

Prioritise matters for deeper diligence.

  • Initial referral screening
  • Identification of missing decision-critical evidence
  • More structured early-stage review
  • Reassessment when material new information arrives
  • Support for investment-team prioritisation
ATE insurers

Structure early merits screening.

  • Initial case triage
  • Identification of evidential gaps
  • Consistent pre-underwriting review
  • Reconsideration as the information record develops
  • Decision-support before more intensive professional work
Professional decision-support only. ClauseCast does not make funding, underwriting or litigation decisions and does not provide legal advice. Outputs must not be used as the sole basis for a professional decision.
Evidence

Historical testing shows meaningful ranking separation.

In an internal retrospective benchmark of 960 historical English contract cases, higher-ranked and lower-ranked matters showed substantially different observed claimant outcomes.

960 historical English contract cases

Higher-ranked and lower-ranked cases separated materially in retrospective testing.

The figures below are observed claimant outcomes in the highest- and lowest-ranked deciles of the internal out-of-fold benchmark.

Internal · retrospective · group-level
Highest-ranked 10%88.5%

Observed claimant success in the highest-ranked decile.

Lowest-ranked 10%10.4%

Observed claimant success in the lowest-ranked decile.

These results demonstrate historical ranking separation, not overall accuracy, prospective performance, improved customer outcomes or the result of any individual dispute. Independent prospective validation remains necessary.
Explore the evidence →
Security & data handling

Designed for sensitive professional workflows.

The current service uses controlled access, data-minimised application storage and explicit provider disclosures. The limitations matter too, so they are documented rather than hidden behind generic security claims.

Access
Controlled accounts

Passwordless authentication, authenticated backend requests and server-side access controls support invitation-only use.

Application storage
Data minimisation

The application is designed not to deliberately persist case bodies, uploaded files or extracted document text in Firestore.

Processing
Transparent providers

Relevant infrastructure, authentication and AI-processing providers and their current arrangements are publicly identified.

Limitations
No zero-retention claim

Gemini project logging is currently enabled for seven days; ClauseCast does not claim external certification or independent penetration-test assurance.

Current pilot data policy: submit anonymised, pseudonymised or otherwise appropriately authorised material and avoid unnecessary personal, privileged or highly sensitive information. Review the current privacy, security and subprocessor documentation before using live matter data.
Defined scope

Purpose-built, not general-purpose.

A narrow scope makes it easier for professional users to understand what the current product is intended to assess — and what it is not.

Current dispute scopeEnglish breach-of-contract claims
Current usersFunders, ATE insurers and professional screening teams
RoleEarly-stage statistical decision-support
Not intended forLegal advice or autonomous funding, underwriting or litigation decisions
Private demonstrations & pilots

See ClauseCast on a real screening workflow.

We run private demonstrations and pilot discussions with litigation funders, ATE insurers and professional teams evaluating early-stage commercial litigation screening.

Invitation-only accessEnglish breach-of-contract scopeProfessional use