Moritz Raises $9M, Bleeds Harvey and Legora

A pre-seed should be a quiet thing. You raise a million or two from a couple of angels, you ship something rough, you find out if anyone wants it. Moritz closed nine million dollars in four days after YC demo day, and the round isn't priced like a pre-seed. It's priced like a thesis bet. The thesis is that selling AI software to law firms is the wrong half of the trade. The right half is being the law firm.

Cofounder Pamir Ehsas told Sifted the quiet part out loud: Moritz isn't pitching legal departments, isn't licensing a co-pilot, isn't carving out a workflow. It's building a law firm from scratch where AI does eighty percent of the work and a roster of fifty-plus contracted co-counsel finalises the rest. The headquarters is moving to San Francisco. The book of business already covers Europe, the US, and Australia. Two billion dollars in aggregate contract value, closed since early 2026, with an operations and engineering team of seven.

That number is the entire story.

The Deployment

Moritz launched in early 2026 and has since supported over 100 companies in closing deals worth more than $2B in aggregate contract value, with an average turnaround of four hours. The work is scoped tight: commercial, corporate, and employment matters. Litigation is out. Immigration is out. Tax is out. The company is explicit that the focus is the work that automates well, where the first draft and the intake can be model-driven and the human pass is finalisation, not invention.

The structure is unusual. Seven full-time staff covering operations and engineering. Fifty-plus contracted co-counsel doing the licensed legal work. The firm is set up in the US, with the headquarters move to San Francisco being treated as a talent and customer-velocity decision rather than a cost optimisation. Ehsas is blunt about why: the European legal-talent market is small, the buyers move slowly, and the gravity of the work is in the US.

The expansion plan is anchored: get one large company in a target country to commit, build the country-specific infrastructure jointly with that anchor, then roll out to other firms in the same market once the volume is there. The UK is already live. Spain, Germany, Belgium, Denmark, Italy, France, and Sweden are next.

The capital table reads like a deliberate signal. Y Combinator. 20VC. Urban Innovation Fund. Twenty unicorn founders from Reddit, Instacart, Dropbox, Hugging Face, and Silo AI. None of those names are legal-tech specialists. They're operator-investors who have priced the AI-enabled roll-up thesis before, and they're pricing it again here.

Why It Matters

The picks-and-shovels frame for legal AI is now contested. Harvey and Legora built the canonical version of the bull case: sell software to law firms, ride the adoption curve, become indispensable. Legora's stated mission, in Ehsas's own words quoted to Sifted, is to make every lawyer in the world AI-native. A real business, and it will compound. But Moritz's pitch is that it's the slower business, because the buyer (a partner-track law firm) has the worst incentive structure in professional services to actually deploy AI efficiently. Billable hours and AI-driven productivity are in direct tension. A firm that bills by the hour does not want a tool that reduces hours.

So Moritz skips the buyer. The buyer is the problem. If the law firm is the bottleneck, the move is to be a different law firm.

This is the AI-enabled roll-up thesis that General Catalyst and others have been articulating for two years: the technology builder owns and operates the service business, capturing the full value chain instead of the slim software margin. Until recently it was mostly talked about in healthcare and accounting. Legal was assumed to be too regulated, too licensed, too partnership-locked. Moritz is calling that bluff with co-counsel contractors and a tightly-scoped service menu. Litigation is messy and adversarial; commercial drafting is patterned and automatable. Pick the patterned half. Win the patterned half.

This echoes the one we saw in customer support eighteen months ago. Sierra picked the service tier instead of the tool tier and ate a category that incumbents thought was theirs. The legal incumbents, magic-circle, white-shoe, the regional mid-market, are reading the same playbook now, except they're cast as the loser. The four-hour turnaround is the punchline. A fifty-lawyer mid-market firm cannot match it without rebuilding its delivery model from the floor.

What Other Businesses Can Learn

The strategic move here is portable, and operators outside legal should read it as a template, not a one-off.

First, scope ruthlessly. Moritz did not say "we do all the legal work." It said commercial, corporate, employment, three categories where the work is patterned, the volume is high, and the legal risk-of-novelty is low. Litigation, immigration, and tax are explicitly out. If you are evaluating an AI-enabled service play in your industry, the same scope discipline applies: pick the work that automates because the work is repetitive, not because the model is clever. The model handles the long tail of the patterned. The human handles the short tail of the actually-hard.

Second, the headcount math is the headcount math. Seven full-time staff, fifty-plus contracted licensed professionals, $2B in deal value across three continents. The contractor use is the unlock. You don't hire fifty senior lawyers as employees and pay benefits; you build a co-counsel roster, route the finalisation pass through them, pay per matter, and keep the operating use on the engineering side. Founders trying to roll up an industry should price this carefully: the regulated tier of the work needs licensed humans, but those humans don't have to be on your cap table or your payroll.

Third, the anchor-customer expansion model is doing real work. Moritz isn't going country-by-country with a marketing budget. It's getting one large customer per country to commit, co-investing in the local-jurisdiction infrastructure with them, and then rolling out to subsequent customers on top of that infrastructure. This is how you avoid the legal-tech failure mode where each new market requires twelve months of compliance build-out before revenue starts. The anchor pays for the infrastructure. The follow-on customers pay for the margin.

The vendors selling AI to law firms are betting on a buyer who is structurally incentivised not to use it. Moritz is betting on replacing that buyer entirely.

Fourth, the four-day close is the signal nobody is talking about. Pre-seed rounds in 2024 took six to eight weeks. A four-day close on $9M, post-demo-day, with name-brand co-investors stacking in, tells you that the AI-roll-up thesis is now a priced asset, not an emerging one. If you have a similar play in your category, accounting, consulting, marketing services, recruiting, the capital is faster than the operational build will be. Don't bottleneck on fundraising. Bottleneck on hiring the seven people who can actually deliver the four-hour SLA.

Fifth, geography is now a velocity decision. Moritz is moving HQ from Norway to San Francisco for a stated reason: the European legal talent pool is small and the European customers move slowly. That's a hard call to make publicly, but it's the right call if your product needs both senior US legal-tech talent and customers willing to be design partners on a timeline measured in weeks. If your AI-services startup is sitting in a smaller European market by default, the cost of staying is showing up in your sales cycle whether you've measured it yet or not.

Looking Ahead

The signal to watch is which incumbent legal-tech vendor responds first, and how. Harvey and Legora have two options: defend the picks-and-shovels position by going deeper into the law-firm workflow, or pivot toward operating their own service tier and absorb the channel conflict. The first defends a real but capped market. The second concedes that Moritz read the category right.

The other signal is the second wave. If Moritz hits its second-country anchor inside ninety days and the four-hour SLA holds, expect three or four copycat rounds in the next two quarters, same thesis, different jurisdictions, same investor names on the cap table. The roll-up window in legal services is open, and it doesn't stay open long.

If you're a regional commercial-law partner reading this, the action item is concrete: price your fastest commercial-contract turnaround today. If it's longer than two business days, you're already losing the bake-off you don't yet know you're in.

Sources

• Sifted, Y Combinator alum Moritz raises $9m to build a law firm supercharged with AI, accessed 2026-05-05