Precedent & Knowledge Finder
A fee-earner asks a question in plain language and gets the answer from your own precedents, advice, policies and matter files — with the exact document, version and clause quoted, and an honest "not on file" when it isn't there.
It indexes your document library behind your existing access controls and answers each question with the matching passage, naming the source — or refusing and routing to a person when the answer isn't in the corpus.
Where this works well
The slow, invisible problem this makes visible is that a firm's most valuable knowledge — the precedent that fits, the advice already given to a similar client, the clause that survived the last negotiation — lives in inboxes, drives, and the heads of a few senior people. Fee-earners spend roughly a fifth of the day looking for it, and the search interrupts the people whose time is most expensive. The information exists; finding it on demand is the drudgery.
This pattern earns its keep where a firm has a real corpus worth searching — a precedent library, an engagement-terms and risk policy, a conflicts and independence manual, a staff handbook with delegations — and a steady stream of "where's the rule on this?" questions. The biggest beneficiaries are the senior solicitors and partners who currently field those interruptions, and the junior fee-earners who lose hours not knowing the firm already answered the question two years ago. A grounded answer that names "Conflicts & Independence Manual v4, §7.4" is worth far more than a confident paragraph with no source.
Where it works badly
It works badly when the library is thin, contradictory, or out of date, because it answers honestly from what it is given. If three versions of an engagement-terms policy are all indexed and none is marked current, it can surface a superseded clause and present it with full citations — correct that the words exist, wrong that they still govern. It is only as reliable as the corpus is curated.
It is also the wrong tool for questions that are judgments rather than lookups. "Can we waive the liability cap to win this tender?" is a risk-and-delegations decision, not a retrieval — and the honest behaviour there is refusal and escalation, not a generated opinion. If most of what your people need is judgment under uncertainty rather than "find me the clause we already wrote," this will underwhelm.
The test to apply to yourself: pick last week's five most-interrupting questions. If three or more could be answered by quoting an existing document, this fits. If most needed a partner to weigh competing considerations, it does not — yet.
What it doesn't do — and shouldn't
It surfaces and cites; it does not decide. It will tell you what the Conflicts & Independence Manual says about a retention period and quote the clause. It will not decide whether a particular conflict is live, whether a precedent applies to the matter in front of you, or whether a costs-disclosure obligation under the Legal Profession Uniform Law has been met. Those are calls a person makes, and the citation exists precisely so that person can verify the source before relying on it.
When the answer is not in the approved documents, it says so and routes the question to someone who can authorise — the Risk Partner or the document owner — rather than inventing a plausible answer. That refusal is the deliberate boundary, not a failure. In a firm carrying professional-indemnity exposure and confidentiality duties, a tool that guesses is a liability; a tool that says "not on file, here's who decides" is an asset.
What your data has to look like for this to work
The corpus has to be the right documents, identifiable by version, with the live one distinguishable from the superseded. Concretely: a precedent library where each precedent is a discrete, findable item rather than buried in a matter folder; policies and manuals carried with a version (v4, v6, 2026 edition) so the index can prefer the current one; and document-level permissions in your DMS or SharePoint that the index can honour, so a user only retrieves what they are already entitled to read. The demo's citations — "Conflicts & Independence Manual v4 · §7.4" — only exist because the source documents carry that structure.
Most firms have some of this in good shape and a lot of it scattered: precedents emailed rather than filed, the controlling policy living on a partner's desktop, three near-identical templates with no owner. Getting the source documents to a state where the current version is identifiable and access rules are clean is usually the real first job — and it is typically a matter of how knowledge is captured and governed, not of buying a new tool. That work is what RMAI helps with, and it is usually larger and more valuable than the AI layer sitting on top of it.
Could it give someone a confidently wrong answer and have them act on it?
The safeguard is that every answer is tied to a quoted passage with its document, version and section — there is no answer without a source the reader can open and check. When nothing in the corpus matches the question, it refuses rather than guessing and routes the question to the Risk Partner or the document owner. It retrieves and cites; the fee-earner still decides whether the precedent actually fits the matter in front of them.
What if our precedents and policies are a mess — duplicates, old versions, things filed in three places?
Then it will faithfully surface the mess, including superseded clauses, because it answers from what is indexed. Version metadata helps it prefer the current document, but it cannot know a precedent is stale if nothing in your library says so. Getting the corpus to a state where the live version is identifiable is usually the real first job, and it is the work we help with before the AI layer earns its keep.
Does this replace our knowledge lawyer or the senior staff people currently interrupt?
No. It takes the routine retrieval load off senior staff — the fifth of the day spent hunting for the clause that already exists — so their judgment goes to the questions only they can answer. Deciding whether a precedent applies, whether a conflict is live, or whether to waive a liability cap stays with the person. We accelerate the thinking; we do not replace the thinker.
How current does the indexed content have to be?
As current as the decision it informs. For a retention rule under the Australian Solicitors' Conduct Rules or a costs-disclosure obligation under the Legal Profession Uniform Law, an answer from a superseded policy can be wrong in a way that matters. Re-indexing is triggered when a controlling document is updated, and each answer shows the version it drew from so the reader can see how fresh it is.
Where does our data go — does this send privileged matter content to a public AI?
The index sits inside your environment behind your existing DMS or SharePoint permissions, so a user only ever retrieves from documents they are already entitled to see. We scope at the start exactly which sources are indexed and which model arrangement is used, with privileged and client-confidential material kept inside your tenancy rather than sent to a public service.
Estimated build: 3–4 weeks. Most of it is template work we've already done.
Fixed scope, fixed price, fixed dates.
The cost band reflects the engagement shape, not a per-feature line item. We work on fixed scope, fixed price, fixed dates — see the services catalogue for what falls inside each band.
Considering this for your org?
The honest place to start is a bite-sized first piece — one contained change, low risk. Tell us where it hurts; we’ll play it back, scope it, and show you what’s possible.