Legal confidence thresholds with Jev
Intake forms and emails mix contract, employment, privacy, and “please just send a letter.” Counsel still owns advice. Jev routes the text you already collected after your privilege/PII filter. It does not draft opinions or file anything.
This unofficial page is the confidence thresholds slice of the legal intake routing pack. Intent: apply the Jev (TypeSafe System One) decision model to legal intake routing confidence thresholds. Primary search language: Legal Jev confidence thresholds. Confirm patterns on docs.typesafe.ai. This site does not sell, issue, or proxy TypeSafe keys. Use a credential you already have from the console or a documented gateway.
Independent angle (cover ≠ clone): Matter-type + privilege-risk Noul on redacted text — not a clone of a legal-intake SaaS page or “AI lawyer” voice.
Legal use-case context
Thresholds turn legal intake routing answers into act / review / abstain. They are product policy, not a hyperparameter TypeSafe ships. Official 0.5 / 0.9 sketches are illustrations. This slice also carries the false-reject discussion: over-gating legal intake routing hides calibration.
Hub: Use cases. Compare, when the other tool is the real job: legal intake.
Confidence Thresholds inputs
You need (1) pinned answers on a frozen contract and (2) labels for matter-type gold from intake counsel, privilege-risk gold, and whether the queue was acceptable. State shape:
{
"intake": { "id": "m-302", "text": "Vendor wants to use our logs to train models; need redlines this week." },
"playbook": { "privacy": "Training-on-logs or model-training clauses → privacy_queue." },
"meta": { "requestor_role": "procurement", "deadline_band": "this_week" }
}Decision signals and actions
| Axis | Where it lives | Legal use |
|---|---|---|
choice / score / noul |
answer payload | What to do with the intake narrative (redacted) |
confidence |
Choice & Score only | Whether to trust the argmax |
| Distance from 0.5 | Noul | Whether privilege_risk is decided |
FLOORS = {
"tag_type_only": 0.55, # illustrations — replace
"auto_send_external": 0.95,
}
NOUL_TAU = 0.70 # for privilege_risk
def allow(ans, action):
return ans.confidence >= FLOORS[action]Do not treat a Noul of 0.5 as a “medium” legal intake routing score — it means yes and no are equally likely. Conjunctions stay in your code.
Guardrails and escalation
TypeSafe’s confidence-gated examples use a lower bar for recoverable reads than for irreversible actions. Those numbers are illustrations. For legal intake routing, treat auto_send_external as the high bar (sending work product outside the firm or closing a matter). Tune on labels — see offline evaluation.
Band around 0.5 on privilege_risk always reviews. Do not copy 0.70 onto Choice confidence.
Evaluation and rollout notes
- Missed restricted_queue on planted privilege-seeking text
- Wrong matter_type at your floor
- False restricted (throughput vs safety)
Fit loop: pin jev-1.13.0 → replay → plot error vs confidence → pick floors where auto-act error ≤ your SLA. Pin jev-1.13.0 (the versioned id) after you fit thresholds. jev-latest and the marketing line jev-1.13 can move. Log the response model. TypeSafe’s published list price for jev-1.13 is $0.042 per million input tokens (vendor claim — confirm on the models page); output tokens are free on that same page. Unused distractors still bill as input.
Official Python and JavaScript SDKs read TYPESAFE_API_KEY and retry documented 429/529. This site does not sell, issue, or proxy TypeSafe keys. Use a credential you already have from the console or a documented gateway.
Pack map
| Slice | Page |
|---|---|
| Graph and primitives | decision workflow |
What may enter state |
input contracts |
| What to gather first | evidence collection |
| Atomic rules | policy checks |
| Act / review / abstain | you are here |
| Reviewer payload | human handoff |
| What to persist | audit trail |
| How it breaks | failure modes |
| Labeled replay | evaluation |
| Shadow → canary | production rollout |
FAQ
Should auto_send_external use 0.9 everywhere? No. Over-gating hides calibration and dumps the queue on humans. Fit per action.
Can I reuse a Noul τ as Choice confidence? No. Jaggedness: they are not interchangeable. See confidence.
Where is the rest of the Legal pack? Start with Legal decision workflow and Legal human handoff. Cluster hub: Use cases.
Is this legal advice? No. Independent routing notes only. Counsel owns advice. Confirm facts on docs.typesafe.ai.
Can Jev write the contract redline? Not in this pack. Generation is a different job; keep this URL as typed routing.
What this page does not claim
- Not legal advice, not a law firm, not a privilege ruling.
- No accuracy claims.
- Not official TypeSafe.
- Official TypeSafe status, or that jev.pro issues API keys.
- That a schema-constrained answer is automatically factually correct.
Disclaimer
This is an independent unofficial site and is not affiliated with TypeSafe AI; official documentation is available at https://docs.typesafe.ai.
Primary documentation: https://docs.typesafe.ai. Hub: Use cases.
Sources
Public TypeSafe or adjacent documentation only. No private claims.