pro·fess, from the Latin profiteri: to declare publicly

The machine can draft.
A lawyer can profess.

professed.ai is the home of a manifesto and a thesis about the future of legal practice: supporting the case for embracing AI where it saves time and money, yet upholding the ancient, public, binding declaration that marks the line AI should never cross.

The professed seal, pressed in green wax
The machine drafts. The professed verify. The record proves it.
The word

To profess is to declare publicly.

From the Latin profiteri, to avow before others; pro, forth, and fateri, to acknowledge, the same root as confess. When the word entered English it meant taking the vows of a religious order: out loud, before witnesses, binding for life.

The professions

A professional is someone who has professed.

Divinity, law and medicine became the professions because their members did something structurally similar to the vow: they publicly declared mastery of a body of learning and bound themselves to duties above their own interest and above the market.

The lawyer

The vow is still live.

Lawyers are admitted in open ceremony, entered on a public roll, bound by duties to the court that override commercial interest, and can be struck off: the unmaking of the vow. Everything a client trusts about a lawyer flows from that one public act.

The idea

AI-native, professed.

The race to make law firms AI-native is real, funded and already authorised by regulators. It should be embraced; the machine doing the first pass of legal work is what frees lawyers to spend their time on the parts of the job that required a vow in the first place.

But there is broad consensus that AI needs a human in the loop, and in law that consensus undersells the problem. In other domains the human is a checkpoint, and a review and a button click will do. In law the human is the product. The machine can draft, research, assemble and check; it cannot profess. It has taken no vow, sits on no roll, owes no duty to the court and stakes nothing when it is wrong. Advice becomes legal advice at the moment someone professed takes responsibility for it, and no capability curve changes that.

The thesis ends by admitting that none of this is new. The law has always known it is flawed: every right of appeal, every dissenting judgment, says so out loud, and the system survives because of that imperfection rather than despite it. AI is a product of humanity and of nature; it inherits our imperfection at scale and at speed, so it can never be the still point the system checks itself against. Only someone who knows they are flawed, and has publicly promised to answer for it, can be that.

The line

AI-native where it saves time and money: intake, drafting, checking, research triage, the work that never deserved the hourly rates it was billed at.

Professed where judgement, experience and responsibility are sought; where regulated advice requires a human to exercise oversight and answer for the outcome.

The machine may do everything up to the line. The line itself belongs to the professed.

A solicitor who clicks approve on machine output they have not genuinely engaged with has not supervised it. They have counterfeited the signature.

From the thesis

The documents

A manifesto and a thesis.

The manifesto

Twelve declarations

The whole position in twelve short, public declarations, written the way a profession makes commitments: out loud, in public, as a line in the sand. Two minutes to read; a career to keep.

Read the manifesto →

The thesis

The full argument

Why the AI-native race is right, where it goes wrong, the history of the word profess, why the public declaration at the heart of the legal profession is the foundation of everything a lawyer does and is, and an honest admission that none of it is new; that is rather the point.

Read the thesis →

First the idea, then the practice

The idea spreads by being read.

Once the manifesto and the thesis have had their say, see where they are already at work, and where the author declares his own interest.

See it in practice