Every legal AI product built for people without a lawyer

We catalogued 264 named products from the 95 companies in our legal AI funding survey. Fourteen are sold to a person handling a legal problem without a lawyer.

Fourteen is a small enough number to print, so here they are — and then a fifteenth that our own method had quietly filtered out, which turns out to be the most interesting entry of the lot.

The whole list

Divorce. Hello Divorce sells five of the fourteen on its own: a DIY forms plan, a plan where a non-lawyer specialist prepares and files the paperwork, a flat-fee attorney plan in 16 states, a chatbot called Hallie that its own site says is not a lawyer, and Homesplit, which pairs divorcing clients with brokers to divide the house.

Immigration. Boundless sells attorney-reviewed packages — spousal green card, K-1, adjustment of status, naturalisation — through an Arizona-licensed firm. Casium files work visas, though it sells mostly to employers rather than individuals.

Buying a house. In England, Keith does fixed-fee freehold conveyancing at £1,200 to buy and £1,125 to sell. Conveyd does the same work at a fixed fee, though Conveyd is not itself a regulated firm — it is a front end over one.

General consumer matters. Lawhive sells fixed-price legal help matched to solicitors in its network, and now owns a law firm outright. DoNotPay handles subscription cancellations, bill negotiation and ticket appeals.

Mass tort. Justpoint mines medical records and FDA data to find people harmed by drugs and products, then routes them to partner firms or to its own Arizona-licensed firm.

That is all of it — from the funded companies.

The one our own method could not see

Our catalogue starts from a funding survey. To be in it, a company had to have raised enough for the round to be reported. That criterion has a blind spot, and we walked straight into it.

Courtroom5 has raised roughly $420,000 since 2017, most of it grants — NC IDEA, Google for Startups' Black Founders Exchange, Techstars Kansas City — rather than venture rounds. No priced round has ever been announced. By the rule we set, it did not qualify. By any reading of the market, it belongs.

It is self-operated software for self-represented civil litigants, rebuilt in 2026 around AI as the LAW Accelerator. Proof maps your claims to their legal elements and scores how well your facts support each one. Strategy summarises the other side's filings and tells you what procedurally comes next. Documents runs a seven-step drafting workflow ending in an oral argument simulator. Pricing is published: $1 for the first hundred credits, or $150 a month, with an estimate of about $1,500 to take a case to trial.

Founded by Sonja Ebron and Debra Slone in Durham, North Carolina, after Ebron — an electrical engineering PhD — was sued in a debt collection case and could not afford a lawyer.

In May 2026 it became a launch Justice Partner in Anthropic's Claude for the Legal Industry. On 25 August 2026 it appeared in Google's Gemini Enterprise for Legal connector list, alongside Harvey, Thomson Reuters, iManage, Relativity and NetDocuments — the only tool for people without lawyers on a list otherwise made of enterprise incumbents.

A company with $420,000 is shipping into the same connector list as companies with a billion.

Three things the list tells you

First: almost none of the funded ones are software. Thirteen of the fourteen are legal services — a firm, a forms specialist, an attorney on a flat fee. Genuinely useful, frequently cheaper than the alternative, but somebody else doing the work. Among funded companies, what a person actually operates themselves comes down to DoNotPay, the Hallie chatbot and a DIY forms plan. Courtroom5 is the substantial exception, and it is the one that took almost no money.

Second: one of them is not what it appears to be. Justpoint is tagged for unrepresented people because its public site recruits claimants directly. But the product is claim sourcing. You are not the customer; you are the inventory. That is a legitimate business and mass tort has always worked this way. It is not self-help, and counting it as such would flatter the number.

Third: prices are published. Keith posts £1,200. Hello Divorce posts its plan tiers. Boundless posts package pricing. Compare that with the rest of the catalogue, where almost nothing carries a public price and every enterprise product wants a call first. Consumers will not sit through a sales call, so anyone selling to them has to say the number. Published pricing turns out to be the most reliable signal of who a legal AI product is really for.

What is missing is more striking than what is there

The fourteen cluster in divorce, immigration, conveyancing, consumer admin and mass tort. Every one of those involves a transaction with money already moving — a house sale, a visa, a marital estate, a settlement.

Now consider where unrepresented people actually stand in court.

In Virginia's General District Court between 2018 and 2023, the Legal Services Corporation's court data project found that in debt collection cases 79% of collectors had a lawyer and 2% of consumers did. In eviction cases, 68% of landlords had a lawyer and 1% of tenants did.

Those are among the highest-volume civil proceedings in the country, and they are the ones where the imbalance is most extreme.

Products from the 95 funded companies aimed at either: zero.

Courtroom5 covers debt collection and foreclosure, which is precisely the point — the one tool pointed at that ground is the one venture capital did not fund.

Beyond it: nothing for eviction defence. Nothing for custody or child support as distinct from divorce, nothing for benefits appeals, nothing for expungement, nothing for restraining orders. Not underserved. Empty.

Why the gap is exactly where it is

This is not a mystery, and it is not villainy. The fourteen exist where someone can pay. A house sale has a house in it. A green card application has an employer or a family that has been saving. A divorce has an estate to divide.

An eviction defendant is, by definition, someone who could not pay rent. A debt collection defendant is someone who could not pay a debt. The demand is overwhelming and the ability to pay is close to zero, which is the one combination that venture capital cannot serve on its own.

We think that is an honest description of the limit, not an excuse for it. Some of that ground belongs to legal aid, to courts building their own self-help tools, and to whatever the regulators decide to permit next. But some of it is reachable by anyone willing to build for a smaller cheque — and the clearest evidence is that the company doing it raised about a fifteen-hundredth of what Harvey has.

Fourteen products across $6.2 billion, and the fifteenth came in on grants. Whatever else the legal AI boom turns out to be, it is not yet a boom in access — and the part that is, is not being funded like one.

Figures are as disclosed publicly and are not independently audited. The Virginia representation figures come from the Legal Services Corporation's Civil Court Data Initiative and describe one state's general district court. Courtroom5's funding total is a floor estimate from a funding aggregator, and its usage and outcome statistics are company-reported. Nothing here is legal advice, and we have not tested any of these products.