Transparency Report
What actually ran
Every figure on this page is computed from a real record — the published evaluation scorecard, the metadata-only request ledger (counts, never dollar figures; no document content; no personal data), the append-only adjudication trail (aggregate shares only), and the permission matrix the code actually enforces. Nothing here is authored copy. Companions: DraftBench · Engine status · Coverage · Security & Governance.
Published evaluation results
The committed DraftBench scorecard (run 7/17/2026) — skips are honest gaps awaiting credentials, never counted as passes.
202/207 passed· 0 failed · 5 skipped
- clauses (deterministic assembly fidelity) — 44✓
- cite-check (format + existence honesty) — 12✓ 2○
- grounding (cited-or-flagged validator) — 10✓
- red-team (adversarial guardrail checks) — 13✓ 3○
- gateway — 13✓
- drafting-gates — 10✓
- matter-os (deterministic extraction) — 11✓
- court-packet (readiness guidance honesty) — 6✓
- attorney network & paralegal mode (finalization-never) — 14✓
- small firm console (deterministic firm playbook) — 11✓
- argument war room (framing invariant + grounding) — 13✓
- source-locked assistant (boundary routing invariant) — 11✓
- upl capability engine (fail-closed matrix + ledger) — 34✓
Model-request ledger (counts only)
2
Requests (all time)
2
Last 30 days
0.0%
Error rate (all time)
0
Errors (all time)
Attorney adjudication of AI findings
2
Decisions recorded
2
Approved as proposed
0
Revised
0
Rejected
Attorney reversal share (revised + rejected): 0.0% — we publish this number because a supervised tool should be able to show how often the supervisor overrides it. Aggregates only; no reviewer identities, no matters.
Complaints, refunds & corrections
Every case opened on the support pageis counted here — including the ones we haven't resolved. Counts only; never content or identities.
1
Cases opened
1 (0 resolved)
correction
Resolution share: 0.0%.
The permission matrix, as enforced
This table is generated by calling the same capability()function the routes enforce — the page cannot claim a posture the code doesn't implement. L5 is disabled for everyone: the platform never files, serves, negotiates, or appears.
| Output class | Consumer (LegalDraft) | Attorney desk (DraftCounsel) |
|---|---|---|
| L0 · Published information | Enabled | Enabled — attorney adjudicates |
| L1 · Clerical processing | Enabled | Enabled — attorney adjudicates |
| L2 · User-directed drafting | Enabled — only from your explicit selections | Enabled — attorney adjudicates |
| L3 · Personalized analysis | Enabled — only for a state you select | Enabled — attorney adjudicates |
| L4 · Recommendation / strategy | Unranked alternatives only — you choose; software-selected recommendations only under an express state safe harbor (e.g. TX), labeled as automated suggestions | Enabled — attorney adjudicates |
| L5 · Representation | Never — for anyone | Never — for anyone |
13 surfaces are classified in the registry; unregistered surfaces and unknown states fail closed. Hard gates currently enforced: 14 of 14 registered (jurisdiction, coverage, claim-state-exact, attestation, access, attorney-finalization, citation-label, execution, upl, required-facts, conflict, currentness, court-form, upl-capability).
Why this page exists: regulators and courts are actively deciding how AI legal tools should be supervised, and the reform frameworks converge on disclosure — publish what the tool does, how often it errs, and who exercises judgment. DraftCounsel publishes those numbers from its own records rather than asking to be taken on faith. This report is factual disclosure, not a certification, and DraftCounsel is not a law firm.