Review Level (LAM): sufficient supervision, made auditable
How much human review is enough for a task? LexCore turns that judgment into an explicit, auditable dial that drives how many findings escalate to you.
"Review Level" is LexCore's implementation of the Lawyer Analysis Metric (LAM) — the question of how much human review counts as *sufficient* supervision of AI for a given task. Supervision runs on a spectrum from review-everything (safe, no time saved) to review-nothing (sanctionable); the right point is task- and risk-dependent.
| Review Level | Typical task | Effect |
|---|---|---|
| Low | Routine extraction, defined-term pulls | High auto-pass; few escalations |
| Moderate | Standard contract review | Balanced; the default |
| High | Unusual clauses, liability, novel risk | Low auto-pass; more findings escalate |
| Highest | Deal-critical, strategy | Almost everything reaches a lawyer |
The level sets the confidence bar a finding must clear to auto-pass. The higher the level, the more findings escalate to a Human Gate. Every review records the level applied — turning a hand-waved judgment into a defensible, auditable record.
What LAM is not
Review Level is a supervision policy and record, not a correctness guarantee. A finding that auto-passes is still only as trustworthy as its CitationGate-verified grounding. The dials are policy settings, and they can be calibrated against measured accuracy over time.
Related technical documentation
Human Gates & the triage worklist
Keeping a lawyer in the loop without drowning them: what escalates, and how the survivors are ranked by risk and materiality.
CitationGate: grounding every finding
The trust core. A finding with no verifiable quote is rejected; a quote that doesn't match the source is dropped. This is why LexCore's output is traceable.
Data sovereignty & model routing
How the model router keeps a sovereign tenant's documents away from third-party model providers — and why where your ANALYTICS goes is a separate question with its own control.