--- title: "Moritz Helped Close $2.3B in Contracts in Months. AI Expanded Legal Demand." slug: "ai-native-law-firm-moritz-demand-expansion-legal-infrastructure" author: "Mikel Amigot" date: "2026-09-30 12:00:00" category: "Premium" topics: "AI native law firm, Moritz, legal AI demand, Jevons paradox legal, fixed fee legal services, legal infrastructure, sovereign legal AI, law firm AI economics" summary: "Moritz, an AI-native San Francisco firm formerly called Arcline, had helped over 100 companies close more than $2.3 billion in contract value at a four-hour average turnaround by May 2026, and now says it serves 200+ in-house teams. The constraint AI removed was turnaround, not headcount, and turnaround is a property of infrastructure you either own or rent." banner: "" thumbnail: "" linkedin: | The AI-native law firm story that matters is not that AI is shrinking legal work. It is that one firm made legal turnaround so cheap that demand expanded to fill it. Moritz, San Francisco, formerly Arcline. By early May 2026 it had helped over 100 companies close more than $2.3 billion in aggregate contract value, at an average turnaround of four hours. It now says it is trusted by 200+ in-house legal and operations teams. It raised a $9M seed backed by Y Combinator and 20VC. Worth correcting one framing I keep seeing attached to it: this was not a small firm four months ago. Four months ago it had already done the $2.3 billion and the 100 companies, under a previous name. The trajectory is real. The "came out of nowhere" version of it is not, and that distinction changes what you can learn from it. Here is the part that transfers. Moritz prices per matter, flat, with commercial contracts from $750 and volume work from $250 a matter, typically around 50% of standard market rates. That is only survivable if the cost of producing the work fell further than the price did. So the variable that moved is turnaround, and turnaround is not a talent property. It is an infrastructure property: retrieval over the firm's own precedent, agents that run against the firm's own document store, an audit trail per matter. Which is why the lesson for an established firm is not "hire AI-native lawyers." It is that if your throughput ceiling is set by a vendor's platform, your price floor is set there too. On ibl.ai you own all the code and the data. Contract review, retrieval over your own precedent and per-matter audit run as source inside your own perimeter, model-agnostic across any LLM, with no per-seat pricing. Privilege stays where privilege has to stay. Efficiency that you rent is someone else's margin. #iblai #LegalAI #LawFirms #LegalTech #AgenticAI #LegalOps --- ## The Short Answer **Moritz, an AI-native San Francisco firm formerly called Arcline, had helped over 100 companies close more than $2.3 billion in aggregate contract value at a four-hour average turnaround by early May 2026, and now says it serves 200+ in-house teams. AI expanded legal demand rather than shrinking it. On ibl.ai you own all the code and the data, so the throughput that makes those prices work is infrastructure a firm holds rather than rents.** The reflex reading of an AI-native law firm growing fast is that AI is eating legal work. The numbers say something more useful: at a low enough price and a short enough turnaround, work that never reached a law firm at all starts reaching one. ## What has Moritz actually done, and over what period? By **6 May 2026**, [Artificial Lawyer reported](https://www.artificiallawyer.com/2026/05/06/how-ai-first-moritz-is-taking-on-big-law/) that the firm had "helped over 100 companies close deals worth more than $2.3 billion in aggregate contract value, with an average turnaround of just four hours." The firm is based in San Francisco, was **formerly named Arcline**, and its co-founder is named in that coverage as Pamir Ehsas. Its [own site](https://www.moritzlegal.com/about) describes founders as Harvard- and Oxford-trained attorneys, with attorneys hired from Fenwick, Cooley and Goodwin. It raised a **$9 million seed round**, [reported by The Global Legal Post](https://www.globallegalpost.com/news/ai-native-law-firm-start-up-moritz-secures-9m-in-seed-funding-round-533457054) as led by Y Combinator and 20VC. Its site says the round closed in four days. The firm now states it is "trusted by 200+ in-house legal and operations teams." ## Was Moritz a small law firm four months ago? No, and this is the framing most worth correcting, because the wrong version of it teaches the wrong lesson. Four months before this writing, Moritz had already passed **$2.3 billion in aggregate contract value across 100+ companies** and was operating at a four-hour average turnaround. It was not a general practice that discovered AI. It was an AI-native firm under a previous name. So the interesting delta over those months is roughly a doubling of the in-house teams served, from 100+ companies to 200+ teams, on a model that already worked. That is a scaling story, and scaling stories are the ones that transfer. Overnight-emergence stories are not. **We are not repeating the "44 countries" and "$3.5 billion" figures** that circulate alongside this story. We could not find either in the firm's own materials or in primary coverage, and the sourced figures are strong enough without them. ## Why does cheaper legal work produce more legal work, not less? Because the binding constraint on corporate legal spend was never appetite. It was the per-matter cost of getting an answer, and a large volume of real legal need sat permanently below that threshold. Moritz publishes its prices, which is the evidence. Its site lists **$250 to $2,500 per document**, with "commercial contracts from $750" and "volume from $250 a matter," describing other matters as flat-priced at "typically 50% of standard market rates." At those numbers, the NDA an operations lead was going to sign unreviewed becomes reviewable. The vendor agreement a finance team was going to accept as-is gets read. None of that work was previously lost to a competitor; it was not being done at all. This is the specific mechanism behind "AI expands the market." Not that firms do more of the same matters, but that the floor drops and a category of matter that was economically invisible becomes visible. ## What was the constraint that AI actually removed? Turnaround, and this is the part that decides whether an established firm can copy any of it. The published differentiator is not a smarter argument. It is four hours where the market expects weeks. The firm's own case study for an unnamed Fortune 500 client cites a **77% cost reduction** and **24-hour turnaround**. Both of those are throughput claims, not judgment claims, and its site is explicit that most clients "hear back the same day." Turnaround at that scale is not a property of who you hire. It is a property of the system the work runs on: retrieval over the firm's own precedent rather than a generic corpus, agents running against the firm's own document store, a record of what was consulted per matter. Which is why the competitive threat to an incumbent firm is badly described as "AI-native competitors." The threat is a competitor whose cost of producing a first-pass answer fell by an order of magnitude while the incumbent's did not. ## Does a firm need to be AI-native to get this economics? No, but it does need to hold the infrastructure rather than subscribe to it, and the reason is arithmetic rather than ideology. A firm whose review throughput runs on a licensed platform has its price floor set by that platform's meter. If the cost of a first-pass review is a seat licence multiplied by headcount, the firm cannot quote $250 a matter on volume work, because its own input cost does not fall when volume rises. Per-seat pricing is the wrong **shape** for legal AI specifically. Review volume has no relationship to how many lawyers a firm employs: a single mid-market M&A deal can carry thousands of contracts through a first-pass pipeline without adding a lawyer. We laid out that arithmetic at firm scale, including what changes at 200 lawyers, in [AI Cost Math for Law Firms: Per-Seat vs Usage](/blog/ai-cost-math-for-law-firms-per-seat-vs-usage). ## What does the privilege question do to this? It removes the cheapest version of the answer. A firm can reach four-hour turnaround by sending client documents to a third-party platform, and in doing so makes privilege a contract term rather than a technical fact. That is a live obligation rather than a theoretical one. Adoption of AI inside legal teams is running ahead of the record-keeping that has to support it, which we worked through in [Legal AI Adoption Is Settled. The Record-Keeping Is Not.](/blog/legal-codex-adoption-108x-governance-gap) The resolution is not to slow the adoption. It is to put the throughput inside the perimeter where privilege already lives, so the audit question has a technical answer rather than a vendor assurance. ## How does ibl.ai deliver this inside a firm's own perimeter? On ibl.ai you own all the code and the data. Contract review, retrieval over the firm's own precedent, agent skills written by the firm's own practitioners and a per-matter audit trail run as source the firm deploys in its own environment, model-agnostic across any LLM, with no per-seat pricing. That is the honest version of the Moritz lesson for a firm that already exists. The firm did not win by being new. It won by getting its cost of producing an answer under its own control, and then pricing against it. **1.6M+ users across 400+ organizations** run the platform this way, including NVIDIA, MIT, and Syracuse University. The deployment-level version of this, running the platform inside the firm's own network, is in [Next-Generation Sovereign AI for Law Firms](/blog/nextgen-sovereign-ai-law-firms-legal). ## Want four-hour turnaround on infrastructure your firm owns? We deploy contract review, retrieval and per-matter audit as source code you keep, inside your own network or fully air-gapped. [Book a 30-minute demo](https://cal.com/iblai/30min) or [talk to the ibl.ai team](/contact). ibl.ai is family-owned and operated from New York, NY. ## Why does owning the AI stack matter? **ibl.ai is the agentic AI platform where you own all the code and the data. You self-host the entire stack inside your own perimeter, run it model-agnostic across any LLM and switch anytime, and pay by usage with no per-seat pricing — so you can deploy anywhere: your cloud, on-premise, GovCloud, or fully air-gapped.** - **You own all the code and the data.** Full source code under a perpetual license, running on your infrastructure. Not API access to someone else's platform — the stack itself is yours. - **Model-agnostic.** Run any LLM — Claude, GPT, Gemini, Llama, Command, or your own fine-tune — and switch providers without rewriting the platform. - **No per-seat pricing.** Usage-based billing against a budget cap you set. Cost tracks what your organization actually uses, not how many people you employ. - **Deploy anywhere.** Your cloud, your VPC, on-premise, GovCloud, or a fully air-gapped network with no outbound connectivity. 1.6M+ users across 400+ organizations run the platform this way, including NVIDIA, MIT, and Syracuse University. ibl.ai is family-owned and operated from New York, NY — a U.S.-headquartered, domestically-owned long-term partner, not a vendor that sells licenses and moves on.