AI-Native and Venture-Backed Law Firms Are Emerging. Here Is How Traditional Firms Should Respond.

Summary of an essay from Dorna Moini's Legal Tech Notes, published on Substack on June 9, 2026. Dorna Moini is the CEO and Founder of Gavel and a former litigator at Sidley Austin. This page gives the argument and the numbers; the full essay is on Substack.

In brief: A new competitor has appeared in the last 18 months: venture-backed firms that hold bar licenses and charge $400 for a contract review instead of $4,000. The essay names them (General Legal, Crosby, Eudia, Norm Law), explains the three legal structures that let outside capital into law, shows which parts of a traditional firm's revenue are exposed, and gives four moves a traditional firm can make without raising $60 million.

Who are the new competitors?

Two kinds. AI-native law firms are built around a model doing the first pass with lawyers as the final check: General Legal, founded by veterans of the team that built CoCounsel, handles startup paperwork and commercial agreements on flat fees with a four-hour turnaround in all fifty states; Crosby does high-volume contract review with a median of 58 minutes to licensed-attorney sign-off, raised $60 million in April at a reported $400 million valuation, and counts Cursor and Clay as clients; Eudia has raised more than $105 million for enterprise contracting and M&A diligence; Norm Law launched in November 2025 with $50 million from Blackstone and two months later hired the chair of Sidley Austin's executive committee.

Private-equity rollups are the second kind. They use a management services organization (MSO), a separate company that owns the firm's back office and that investors can buy into while lawyers keep control of the practice. Massumi + Consoli, a Los Angeles M&A boutique, sold a stake in its back office to private equity; McDermott Will & Schulte confirmed preliminary talks about a similar structure; Quinn Emanuel's founder expects it to become common. Legal tech venture funding went from about $1 billion in 2023 to more than $4 billion in 2025, with $2.34 billion across 103 deals in the first quarter of 2026 alone.

What does an AI-native firm do differently?

Traditional firms add AI to workflows built for humans; AI-native firms build the process around the model. The described workflow has five steps: intake by email or Slack with no human queue, an AI first pass against market-term libraries that produces redlines and clarifying questions, routing by specialty and load, licensed-attorney sign-off, and a feedback loop that retains each client's fallback positions and negotiation history. Lawhive's AI assistant in the UK scored 81% on the Solicitors Qualifying Exam against the 55% a human needs to pass.

Pricing is flat or per document. With the model doing most of the work, the essay estimates five to ten times the output per lawyer, which is how a $400 review can be more profitable than a $4,000 one. The moat is the institutional layer, meaning the license, the insurance, the proprietary data and the switching costs, rather than the software.

Which parts of a traditional firm's work are exposed?

The reassurance that AI cannot do bet-the-company work is mostly true for M&A, high-stakes litigation, novel regulatory strategy and unsettled law. But firms profit mainly from volume work done by associates: document review, diligence, first-draft markup and research memos, a point the essay credits to general counsel Eric Greenberg writing in Bloomberg Law. The question for any firm is how much of its billing is judgment and how much is volume a model plus a reviewer can deliver for a fraction of the price. Blackstone paid Kirkland & Ellis $101.3 million in fees in 2024 and also funded Norm Law to do similar work for less.

How are these firms allowed to operate?

1. Direct outside ownership, permitted in some form in Arizona, Utah, the District of Columbia and Puerto Rico.
2. A dual-entity setup: a technology company takes venture money alongside a lawyer-owned firm that holds the license, linked by a services agreement. This works in all fifty states.
3. The MSO route: investors own the operations company and lawyers keep the practice, a structure that has been compliant for decades in medicine and dentistry. A Texas ethics opinion last year held that the operations company's fee cannot be a cut of legal revenue.

What should a traditional firm do now?

The essay's point is that a focused firm does not need $60 million, because venture-backed firms spend it on engineering talent, multi-jurisdiction lawyers and customer acquisition. Four moves:

1. Choose a narrow, repeatable, pattern-heavy practice you already know, the way Crosby chose sales contracts and Garfield Law chose UK small-claims debt recovery.
2. Refine the stack around orchestration: one trigger that cascades through intake, drafting, calendaring and billing matters more than any single tool's features.
3. Get compliance right: a proper entity and human review of every output.
4. Price flat, so speed increases earnings, and watch client demand; clients expecting AI to lower spend are already routing routine work to firms priced for it.

Mid-sized firms and solos that never relied on the leverage pyramid are well positioned, because AI widens the efficiency and relationship advantages they already have.

Key numbers and sources from the essay

• Crosby: median 58 minutes to licensed-attorney sign-off; $60 million Series B in April 2026 at a reported $400 million valuation; clients include Cursor and Clay.
• Eudia: more than $105 million raised. Norm Law: $50 million from Blackstone, launched November 2025.
• Legal tech venture funding: about $1 billion in 2023, more than $4 billion in 2025, $2.34 billion across 103 deals in Q1 2026.
• AI-native firm directory: 23 firms a year ago, 40 today.
• Lawhive's AI assistant: 81% on the Solicitors Qualifying Exam, versus the 55% pass mark.
• Blackstone paid Kirkland & Ellis $101.3 million in legal fees in 2024.
• Outside ownership permitted in some form in Arizona, Utah, D.C. and Puerto Rico.

What this means for a law firm or legal team

1. List your matter types by revenue and mark each as judgment or volume. The volume column is what a $400 competitor is pricing against.
2. Pick one volume practice and rebuild it end to end: intake, AI first pass against your own playbook, attorney sign-off, and a feedback loop that keeps each client's positions.
3. Move that practice to flat fees before a competitor sets the market price for you.
4. The building blocks exist without a venture round: Gavel Workflows handles intake-to-document automation and client portals, Gavel Exec handles playbook-based contract review in Word, and the Gavel case studies show firms that run this way today.

Frequently asked questions

What is an AI-native law firm?

A licensed law firm built around an AI model doing the first pass of legal work, with lawyers providing review and sign-off, usually priced flat or per document. Examples named in the essay are General Legal, Crosby, Eudia and Norm Law.

How can investors own part of a law firm?

Three routes: direct outside ownership where permitted (Arizona, Utah, D.C., Puerto Rico), a dual-entity structure pairing a venture-funded technology company with a lawyer-owned firm, or a management services organization that owns the back office while lawyers keep the practice.

Does a traditional firm need venture funding to compete?

No. The essay argues a focused firm can compete by choosing a narrow repeatable practice, orchestrating existing tools around it, getting the entity and review structure right, and pricing flat.

Read the full essay: AI-Native and Venture-Backed Law Firms Are Emerging. Here Is How Traditional Firms Should Respond. on Substack, or subscribe to Dorna Moini's Legal Tech Notes.

Related on gavel.io: Gavel Workflows, Gavel Exec, Gavel case studies, The 1 to 2 year window for lawyers.