Understanding finding verdicts

The three controls on every AI review finding — your call, your note, and what changes in the document. What they mean, how they differ from the AI's check result, and how to use them well.

Updated 27 Jul 2026

Every AI review produces a set of findings — one per clause your rulebook told Clment to look at. Each finding card reads the AI’s analysis to you, then walks you through three controls of your own. They’re separate on purpose: what you think of the recommendation, why, and what actually changes in the document are different questions.

The AI’s check result is not your verdict

Before your controls, one distinction: every finding carries the AI’s check result — did this clause pass the rulebook’s standard?

Failed — the check found the requirement is not met; the finding says what and why.
Warning — partially met or risky; read the finding before deciding.
Passed — the requirement is met, no change needed.
Out of scope — this rulebook item doesn’t apply to this document.

You’ll see it as the icon on each finding card (hover it for the meaning), the Check column in the findings grid view, and the Passed / Warnings / Failed / Out of scope tiles on the Overview tab. That’s what the AI concluded. The controls below are what you conclude — you can disagree with a Failed finding or change the wording on something the AI merely warned about.

Your controls

1. Your call — what you think of the AI’s recommendation

Three buttons:

Agree — you agree with the AI’s recommendation for this clause.
Disagree — you don’t agree.
Partially — you agree with part of it; say which part in your note.

The framing trips people up: “Agree” means agree with the recommendation, NOT “agree with the clause as drafted”. If the AI recommends rewording a liability cap and you say Agree, you’re saying yes, reword it.

We chose this framing deliberately because it aligns the verdict with the action. “Do I agree?” maps directly to “is this change going in the redline?”. The alternative — “do you agree with the clause?” — is ambiguous because the clause might already be fine and the AI flagged it for a different reason.

2. Your note — the reasoning

A free-text note explaining your thinking. Always optional, but the card prompts you — “What would you change?” — when you pick Partially or Disagree, because that’s exactly the signal that makes everything downstream better.

Notes stay attached to the finding and show up in:

The exported review report.
The activity log for the contract.
Anyone else’s view of the finding (reviews can be shared with colleagues).
Rulebook refinement — your pushback, in your words, is what Clment uses to draft rulebook improvements.

Useful notes capture why you decided what you decided — not just what you decided. Future you (or your replacement) will thank present you for “Partially — agreed indemnity scope needs widening but their proposed language is too broad; counter with our standard cap” instead of just “see redline”.

3. What changes in your updated contract?

One control settles both questions — whether this finding changes the document, and what the change should say. Three choices:

Use this fix — take the AI’s suggested replacement language as drafted. (Only offered when the AI drafted one.)
Custom instructions — tell the drafter what you actually want instead (“cap at 24 months of fees, carve out confidentiality and IP”). The generator follows your instructions rather than the AI’s suggestion. This is the tool when the AI’s wording is too aggressive (or too soft) for the relationship, when you have house-style language, or when you’re steering to a specific position — “the AI recommended uncapped indemnity, but we’d accept a $5M cap”.
No change — leave the clause alone. The finding still appears in the report; it just doesn’t touch the document.

Clment sets each finding’s starting position in a clear order of precedence — the first rule that applies wins:

Your explicit choice. Whatever you picked here always wins.
Your wording. Saving custom instructions selects the finding for change.
Your call. Once you’ve given one — Agree → in, Disagree → out, Partially → you decide.
The review strategy’s severity floor. For findings you haven’t touched yet, the review strategy you picked decides: Understand & Negotiate includes medium-severity and up, High-priority only includes high-severity and up, and Strict includes everything.

So a finding you’ve engaged with follows your decision; a finding you’ve ignored follows the strategy. This is what lets you run a noisy Strict pass and still produce a focused redline — decide the ones that matter and the rest fall away.

Why is this separate from your call?

You might Agree with the analysis but choose No change for now — you’re saving it for a later round, or the finding is informational (“the indemnity scope is broad — flagging for awareness, not changing”).
You might Disagree with the AI’s recommendation but still want an edit — your edit, not theirs. Custom instructions are how.

Needs direction: a finding set to change the document with nothing to go on — no AI-drafted fix and no custom instructions — is flagged as Needs direction (there’s a filter chip for it on the Findings tab). When you create the updated contract, Clment lists any such findings first and jumps you to them, so nothing lands in the document without a steer from you.

A worked example

Finding: “Limitation of liability cap is set at fees paid in the prior 12 months. Recommended: increase to fees paid in prior 24 months.”

Scenario	Your call	Note	What changes
Accept the AI’s position as-is	Agree	(empty or short note)	Use this fix
Agree with the principle but use our standard cap	Agree	”We want 24mo cap as standard”	Custom instructions: “Cap = fees paid in prior 24 months; carve-outs for breach of confidentiality, IP infringement, gross negligence.”
Cap is fine, no change needed	Disagree	”Standard relationship — 12mo cap is acceptable”	No change
Push for 36mo, not 24mo	Partially	”Agree with principle but pushing for 36mo on this risk profile”	Custom instructions: “Cap = fees paid in prior 36 months …”

Closing the review

When you’re done working through the findings, mark the review Closed from the Finish up tab (or the last stage on the Overview tab). Closing is a one-off, review-level state: the review records who closed it, and anyone with access can reopen it if there’s more to do. A closed review stays intact as the record of what was decided — if the contract moves again, run a fresh review, which gets its own REV-N identifier.

You can assign a review to a colleague before closing (e.g. for a partner review). They get an in-app notification + an email with the ?review=REV-12 deep-link.

Still have questions?

Instant article search