Every Law Firm Is Now a Software Company

Where I explore how the convergence of services, software, and education is rewriting legal practice, right now

When first-year associate Winston Weinberg left O’Melveny & Myers to co‑found Harvey with a former DeepMind researcher, the establishment rolled its eyes. Three years later, Harvey reports $100M ARR, serves 500+ clients, and is valued at $5B after a June raise. Harvey has been on a hiring spree, bringing on former Am Law 100 lawyers to train its systems.

Meanwhile Eudia has taken a different route: it acquired ALSP Johnson Hana with a unique $105M funding arrangement and then opened Eudia Counsel, an “AI‑augmented law firm” under Arizona’s ABS regime, publicly courting the end of the billable hour and naming enterprise clients like Stripe and Duracell.

Across the Atlantic, Lawhive just bought Woodstock Legal Services, a first in the UK for an AI‑native platform swallowing a regulated firm, while its in‑house assistant “Lawrence” scored 81% on the SQE. Oh, and it also purchased an Arizona ABS law firm to expand into the US market as well.

These are not one‑offs. They are signals of a new reality: every law firm is now a software company, not metaphorically, not eventually, but functionally and competitively today.

If this sounds interesting to you, read on…

This substack, LawDroid Manifesto, is here to keep you in the loop about the intersection of AI and the law. Please share this article with your friends and colleagues and remember to tell me what you think in the comments below.

---

The Transformation Triangle

I’ve written about my transformation triangle thesis elsewhere, but now it finds tangible proof. Sustainable businesses increasingly rest on three legs: services, software, and education. Remove one and the model wobbles.

• Services: advice, advocacy, negotiation, the traditional core.

• Education: client counseling, content, CLE, ubiquitous but under‑acknowledged.

• Software: the missing leg at many firms, now becoming the most decisive.

Amazon evolved from retailer to platform; Tesla from manufacturer to software on wheels. The same reclassification is hitting law. Firms that master all three corners win. Those that outsource the software corner to someone else find themselves operating inside someone else’s platform.

Why Software Changes Everything

Bob Ambrogi observed that the Venn diagrams of law and technology, which began to overlap with the advent of personal computers, the Internet and the Cloud, have become entirely congruent with the rise of large language models. Software is now as fluent in the use of language as we are. And, for a profession that thrives on language, argument and definitions, software has become an existential threat.

Software doesn’t just digitize; it reimagines. It turns research and review from activities into systems. It extrenalizes and captures expertise. Smart workflows reshape what counts as “legal work.” Even conservative use cases, document automation, research accelerators, diligence summarization, don’t merely save hours; they change incentives, pricing, and product design.

Harvey is a case study: at A&O Shearman’s pilot, 3,500 lawyers issued ~40,000 queries, pushing generative workflows into daily practice. But also, educating the system about what legal work is and how it’s done. Tools like these migrate knowledge into code paths and checklists, compounding with use.

The revolution isn’t about technology companies serving law firms, it’s about the complete fusion of legal expertise and software capability into something entirely new.

The Law Assimilation Playbook

Richard Susskind frames technology’s impact as automation, innovation, and elimination. Forward‑looking firms must plan for all three.

1. Automation: same tasks, faster. E‑discovery, time capture, intake. Necessary, but not sufficient.

2. Innovation: new services and models (e.g., subscriptions, predictive analytics), enabled by software that orchestrates demand and delivery.

3. Elimination: some work simply disappears because assumptions underlying its necessity no longer hold (routine inquiries handled by agents; self‑executing agreements reduce downstream disputes). This isn’t destruction; it’s evolution.

Susskind’s three modes, automation, innovation, elimination, aren’t menu choices. They’re a conveyor belt. And that belt runs straight across the transformation triangle: Services → Education → Software. The leading legal‑software entities are running this playbook in sequence.

Stage One: Automation Pressures Services

It starts innocently. E‑discovery shrinks review teams. Time capture and intake run themselves. Firms think they’re in control: “tools” helping lawyers deliver the same services, just faster. Harvey’s early pitch: speed up research. Lawhive: strip out admin drag. Eudia: accelerate contract review. It’s the faster‑horse phase: services enhanced by software, with lawyers still educating clients and each other. The triangle looks stable because the software still needs human operators.

Stage Two: Innovation Inverts Education

Then the trap springs. The locus of learning shifts. Every prompt, correction, redline, and acceptance becomes training data. Harvey isn’t just answering; it’s absorbing query patterns and preferred reasoning paths. Eudia’s “Company Brain” turns repeated reviews into institutional memory, patterns, preferences, judgment calls. Lawhive’s “Lawrence” doesn’t merely assist; it accumulates playbooks. Education flips: instead of lawyers educating clients or juniors, lawyers educate the software. The visible innovation is new offerings; the real innovation is knowledge transfer from human practice into code.

Stage Three: Elimination Consolidates Software

Finally, repeatable judgment is embedded. Entire categories of routine work don’t just get cheaper; they cease to require humans at the margin. The bottleneck moves from associate bandwidth to model governance and exception handling. Notice what follows: platforms begin owning the services corner, by acquisition, partnership, or gravitational pull, because the expertise they codified now scales best under their control.

The playbook, weaponized against legacy legal economics:

1. Automate the services to gain trust and access.

2. Innovate by capturing education, harvest the tacit know‑how with every interaction.

3. Eliminate human dependence where patterns are saturated, making the software increasingly self‑sufficient.

Each stage feeds the next. Each adoption event strengthens the platform’s corpus. Firms think they’re climbing Susskind’s ladder; many are being moved, rung by rung, onto someone else’s rails.

Bottom line: The triangle isn’t a project for law firms to complete. It’s the blueprint software companies are using to consume the market.

The Convergence of Frameworks

The legal profession is experiencing all three of Susskind’s stages simultaneously, but at different speeds across different firms. While traditional firms are still celebrating their automation wins (“Look how fast we can review documents!”), the software companies have already moved through innovation (“Every interaction teaches us”) and are approaching elimination (“We don’t need you anymore”).

The transformation triangle that firms think they need to complete? They’re not building it, they’re being folded into it. Services become data points. Education becomes training sets. Software becomes everything. The triangle doesn’t describe what law firms need to become; it describes how they’re being assimilated.

Closing Thoughts

You probably read the title of this article, “Every Law Firm Is Now a Software Company,” and thought it meant you need to “hire devs, build apps.” That’s not the point. It’s too late for that.

The point is that it’s already happened. It’s already here. When your lawyers work inside platforms like Harvey, Eudia, or Lawhive’s stack, your knowledge, patterns, and documents feed systems you do not own. The software companies aren’t just competing with firms, they’re integrating them while marketing empowerment.

The next time you hear, “AI won’t replace lawyers, but lawyers who use AI will replace those who don’t,” you’ll know better.

Winston Weinberg left O’Melveny after one year and built a $5 billion company. He understood something his former colleagues didn’t: the future of law isn’t lawyers who use software or even software that replaces lawyers. It’s software that becomes law itself.

Every law firm is now a software company. You just don’t own the code.