Born or made?
The new AI law firms say transformation has to be born, not made. They are wrong, and the price of believing them is the firm itself: the offer is that you join the platform, take its brand, run your clients through its system and end up a licensed component of somebody else's practice. What stops an ordinary firm operating like them is not the technology. It is two things. The AI-native workflow has to be mapped out, matter type by matter type, so the machine does the first pass and your method survives the redesign. And there has to be a defensible answer to one question: who supervised this machine's work, and can you prove it? NativeLaw is both, built as infrastructure, and it leaves your name on the door.
The SRA has authorised firms that deliver legal services through AI, and sits inside the government's AI Growth Lab. The permission exists. What is missing is the mechanism by which an existing firm crosses over.
The Upper Tribunal has held that using public AI tools on client matters can waive legal professional privilege. Supervision has stopped being best practice. It is now the difference between a protected file and an exposed one.
Survey data suggests a majority of UK lawyers use AI without their firm's authorisation. The question is no longer whether to govern AI use, but whether to do so before or after something is lost.
The gap
The gap is not between firms that use AI and firms that do not. It is between firms that redesigned their practice and firms that did not. Your fee earners have used AI: a letter here, a research query there. It helped with some things. It did not change the firm. That is not your fault, and it is not the tool's fault. AI was added to the old way of working rather than built into a new one.
Used as a faster typewriter, you get faster typing. Used as a better search engine, you get better searches. The firms and legal teams that are moving ahead are not doing the same things faster. They are doing things differently: the machine does the first pass of the work, and lawyers supervise it. Nobody gave the rest of the profession the infrastructure to make that inversion safely, inside the regulator's perimeter. That is what NativeLaw is.
The difference
There are two ways to become AI-native. One is to hand your practice to somebody who already is: join the platform, adopt its brand, route your clients through its system, and accept that the technology, the client relationship and the economics all belong to someone else. It works, and for a lawyer who wants to stop running a business it may be the right answer. It is also how a profession stops owning itself.
It rarely arrives looking like a sale. It arrives as a favour, and usually under the word co-counsel. The platform has the clients, the marketing budget and the technology; you have the practising certificate it cannot operate without; so you come on as co-counsel, take the matters it sends you, and put your name to what its system produces. The work arrives without you having to win it, which is the hardest part of running a practice solved at a stroke.
Look at what has changed hands. The client found the platform and will go back to the platform, because the relationship was never yours. The workflow, the drafting, the file and the record sit on somebody else's rails, so the means of delivering the output is not yours either. What you contributed was the signature, and the signature is the only thing in that exchange that could not simply be bought. Access to clients is a poor price for the one asset the profession does not let anyone else hold. Sold in instalments, a matter at a time, it still amounts to selling the business, and most of the lawyers doing it have not noticed that is what the arrangement is.
The other way is to buy the rails and keep the railway. That is what this is. Your letterhead, your retainers, your client relationships, your professional indemnity arrangements and your judgement on every matter stay exactly where they are. What changes is underneath: how the first draft is produced, and how the supervision behind it is proved. A three-partner practice gets the production economics that were supposed to belong only to the funded ones; a two-hundred-lawyer firm gets a governance record it can put in front of a regulator. Neither has to become somebody else to have it.
AI-native is not AI-professed. The machine may do everything up to the line; the line itself belongs to the professed, and the professed is you.
Work is produced on our rails and delivered under your brand, to your clients, on your retainer and your terms of business. NativeLaw never appears between you and the person who instructed you, never holds the relationship, and will never ask to be on the record beside you. We do not send you work in exchange for your signature.
We are the supplier, not the firmBrief encodes your method, not ours. Nothing is standardised across subscribing firms, because the thing a client is buying is the particular way you practise. The machine is taught how you work; it does not teach you how to.
Your know-how, written down and keptStart with supervision over the AI tools you already run. Add matter types when you decide they are ready. Stop where your own judgement says stop, and keep the mark that describes where you actually are.
Hybrid by choice is a finished positionThe platform
A suite built to work together, or independently if you want to start with one. Each solves a different part of the same problem.
Most firms carry their knowledge in their people's heads. Brief is where you write it down. Build a workflow for each matter type once: questionnaire, document request, research check, drafting step, review gate. From then on every matter of that type runs the same way, for every fee earner, every client, every time.
Institutional knowledge that does not leave with anyoneNot a drafting assistant that hands your lawyer a starting point to rewrite. Opinion takes the workflow Brief defined and runs it on a live matter: it reads what was submitted, identifies the applicable law, runs the specified checks and produces the outputs. Every step logged, every uncertainty flagged before it reaches anyone who relies on it.
From intake to a file ready for reviewA qualified lawyer must take professional responsibility for every piece of work that leaves the practice, including work produced with AI. Control is built around that requirement: a five-layer supervision process ending in an immutable audit record and a named lawyer's sign-off. Run it as the final stage of the suite, or standalone over the AI tools you already use.
Nothing reaches a client without a signature on recordStart anywhere. Control runs as a standalone supervision layer from day one, powered by Countersigned. Brief and Opinion come in matter type by matter type, one workflow at a time.
The other platforms want your firm. We would rather you kept it. In a regulated profession the signature is the product, and a signature cannot be licensed from anybody.
Chris Dias, founder
Who it is for
Convert, without ceasing to be yourself
For managing partners and COLPs who can see the direction of travel: AI-native competitors on price, clients on assurance, the courts on privilege. NativeLaw is the conversion mechanism, and conversion here means what it says. Your firm keeps its clients, its brand, its professional indemnity arrangements and the judgement of its own lawyers. What changes is how the work is produced underneath.
Production cost per matter falls, because expensive human hours move from drafting to reviewing. Capacity rises without headcount. And every output carries a named signature and an immutable record, which is the answer your regulator and your insurer will ask for. Size is not the qualifier here. Small, medium or large, a firm runs the same rails and keeps the same independence; what differs is how many matter types you convert and how quickly.
Supervised AI for the legal department
For general counsel under pressure to adopt AI and answerable for what it produces. Run the same infrastructure inside your department: your playbooks encoded in Brief, execution by Opinion, and every output signed off through Control with an audit record your board, your auditors and your regulators can inspect.
Or use the full vertical: instruct Lawyery for supervised, signed legal work delivered on NativeLaw rails at machine economics. Either way, you get the thing a due diligence questionnaire cannot argue with: proof of who supervised the work.
The standard
The standard has two marks, and which one fits is the firm's decision, not a ranking. A firm holds Certified AI-Hybrid when it runs AI on some of its workflows, at the discretion of the solicitor conducting the matter, with a named lawyer signing that work. It holds Certified AI-Native when the whole practice operates that way: AI on every matter type, the machine taking the first pass, every output signed and recorded, and the standard independently monitored. Any firm can reach AI-Native if it chooses to work that way, and any firm can hold at Hybrid for as long as that is the right answer. The supervision is the same under both, and neither mark permits unsupervised work.
Hybrid is not a consolation prize, and it is not a waiting room. It is the honest mark for a firm that has worked out which of its work belongs to the machine and which does not, and some firms will decide to stop there for good reasons of their own. The standard exists to describe where a practice actually is, not to push it somewhere its own judgement says it should not go.
Our ambition is to align the marks' criteria with the operational controls of ISO/IEC 42001, the international standard for AI management systems, so that the badge and the live record beneath it speak the language your clients' due diligence already asks in. The marks are ours, not ISO's; the discipline behind them is built to match.
Self-declared
For a firm using AI services on some, but not all, of its workflows, with a named lawyer supervising and signing that work. Where a firm draws that line is its own decision, not ours.
Independently verified
For a firm that has chosen to run this way: every matter type redesigned for AI, every output supervised, signed and recorded. The standard set out in the professed.ai manifesto.
The person behind it
Chris Dias has been an immigration solicitor for twenty-five years. In 2022 he and his wife Crystal, both solicitors, founded Lawyery, a regulated virtual-first firm in Holborn. He set out the argument as Professed, built the supervision layer as Countersigned, shipped free public tools through Legalaid, and NativeLaw is the conclusion: the infrastructure that lets any firm do what his does.
NativeLaw is coming soon. Tell us about your practice and we will show you what conversion looks like, matter type by matter type.
Coming soon