Two legal AI companies raised a combined $750 million in fifteen days in March 2026. On March 10, Legora closed a $550 million Series D at a $5.55 billion valuation. On March 25, Harvey confirmed $200 million at an $11 billion valuation. Between them they now represent over $16 billion in enterprise value in a market that barely existed five years ago.
If you run a US law firm, you have almost certainly been pitched one or both. Your partners have seen the demos. Your associates are probably already using one of them, with or without formal approval. The question most firm administrators are asking in 2026 is some version of: which one do we buy, and what does it actually change?
That is the wrong first question. Or rather, it is the second question. The first question is the one almost nobody is asking, and it is the one SRA has spent nearly four decades thinking about: when AI takes over the work that used to train your junior associates and justify their billable hours, how do you develop, evaluate, and retain the people who are supposed to become your future partners?
This guide explains what Legora and Harvey actually are, how they differ, and what each one does well — written for US law firm leaders trying to make sense of the category. It then covers the part the AI vendors are not going to put in their pitch decks: the talent, performance, and retention questions that legal AI creates and cannot solve. That second part is where SRA works, and it is the part that determines whether an AI investment makes your firm stronger or quietly hollows out your talent pipeline.
A note on what this guide is and is not. SRA does not sell legal AI software, and we are not affiliated with either Legora or Harvey. We design and run confidential performance review, evaluation, and engagement programs for US law firms. This is an educational overview from the perspective of the people side of the firm, not a technical software review. For hands-on product evaluation, demo both platforms with your own matters.
What Legora and Harvey actually are
Both are AI platforms built specifically for legal work — not general-purpose chatbots, but domain-specific systems trained and engineered for the way lawyers actually work. Both handle research, document review, drafting, and increasingly, multi-step “agentic” workflows that chain several tasks together. Both are used by major US and global firms. Beyond those similarities, they come from different places and emphasize different things.
Harvey AI
Harvey was founded in 2022 by Winston Weinberg, a former securities litigator, and Gabriel Pereyra, a former research scientist at Google DeepMind and Meta AI. It launched out of early experimentation with OpenAI’s models and has been closely associated with OpenAI and the frontier-model ecosystem since.
By early 2026, Harvey reported it was used by more than 100,000 lawyers across roughly 1,300 organizations, including the majority of the Am Law 100, over 500 in-house legal teams, and dozens of asset management firms across 60 countries. It crossed $100 million in annual recurring revenue in August 2025 and reported $190 million by January 2026. The platform handles document analysis, legal research, contract drafting, due diligence, and complex workflows in a secure enterprise environment, and reported processing over 400,000 agentic queries per day with more than 25,000 custom workflows built by users. Harvey deepened its legal content through an alliance with LexisNexis, integrating statutes, case law, and citations directly into the platform, and announced a Microsoft 365 Copilot integration for Q2 2026.
Legora
Legora (formerly Leya) is a Swedish company that has grown unusually fast. It positions itself as a “collaborative AI platform for lawyers,” with particular strength in research, review, and drafting across complex matters. By 2026 it reported serving over 1,000 leading law firms and in-house teams across more than 50 markets, with customers including Bird & Bird, Cleary Gottlieb, White & Case, Linklaters, Deloitte, Dentons, and Goodwin.
Legora surpassed $100 million in annual recurring revenue and reached a $5.55 billion valuation in its March 2026 Series D, which it raised explicitly to accelerate US expansion, with new offices planned for Houston and Chicago alongside its existing New York presence. Its product is known for “Tabular Review” (turning large folders of contracts into an interactive grid for clause comparison and data extraction at scale) and for natural-language agentic workflows that can incorporate drafting, tabular review, research, and translation. In May 2026 it acquired Qura to build out AI-native legal research.
Both companies are moving fast enough that any specific feature list will be out of date within a quarter. The numbers above are current as of mid-2026 and are drawn from each company’s own announcements and reputable legal-tech reporting. Treat them as a snapshot, not a permanent description.
How Legora and Harvey differ (in practical terms)
For a US firm trying to understand the category, the differences that matter are less about feature checklists and more about positioning, origin, and emphasis. The table below summarizes the practical distinctions as of mid-2026.
The honest summary: these are two well-funded, capable platforms converging on similar territory from different starting points. Harvey leads on raw scale, enterprise penetration, and content partnerships. Legora is praised for collaborative workflow design and its tabular review approach. A firm choosing between them in 2026 should run both against its own matters rather than relying on valuation or ARR as a proxy for fit — the bigger number is not automatically the better tool for your practice mix.
The question neither vendor will put in their pitch deck
Here is what the demos do not cover, and what SRA spends its time on.
For most of the last century, US law firms have run on a pyramid model. Hire a large class of junior associates. Assign them high volumes of routine work — first-pass document review, legal research, first-draft contracts. Bill clients for those hours. Use that work, over five to seven years, to train associates into the judgment, client skills, and substantive depth that eventually makes some of them partners.
Legal AI breaks the bottom of that pyramid. The routine, high-volume work that historically trained junior associates and justified their billable hours is exactly the work Harvey and Legora are best at automating. A Law360 Pulse survey in March 2026 found that AI is no longer experimental in legal — it is operational — and that first-pass document review, contract analysis, and legal research are increasingly handled faster and cheaper than a junior associate can deliver them. Goldman Sachs has estimated that AI could automate roughly 44 percent of legal tasks in the US.
This creates a problem that no AI platform solves and that most firms have not yet confronted: if AI does the work that used to train junior associates, how do those associates develop into senior associates and partners? And if the firm hires fewer juniors because the work is automated, where does the next generation of partners come from?
The efficiency paradox: The same AI that makes your firm more profitable per matter today can quietly erode the talent pipeline that makes your firm exist in ten years. Efficiency is visible immediately. The pipeline damage is invisible until the cohort that should have been your future partners has already thinned out.
Some firms are confronting it directly. In January 2026, Ropes & Gray launched a program requiring newly qualified and trainee associates at its US offices to dedicate one-fifth of their billable time to hands-on AI exploration — counted as billable, not as overhead — a deliberate bet that early AI fluency is now core to associate development. That is a thoughtful response. But it also illustrates the depth of the problem: firms are now having to redesign how associates spend their time, how their performance is measured, and what “good” looks like at each level, because the old answers were built around work that AI now does.
Where SRA fits in the legal AI transition
SRA does not sell AI software, and we are not going to tell you whether to buy Harvey or Legora. What we do is help US law firms answer the questions that AI adoption forces and cannot resolve on its own:
How do you measure associate performance when AI has absorbed the routine work that used to fill their hours? How do you evaluate development and judgment rather than volume? How do you keep associates engaged and learning when the work that used to teach them is automated? How do you retain the people you most want to keep through a transition that is making many of them anxious about their own futures?
These are performance, evaluation, and engagement questions — SRA’s core work for nearly four decades. The AI transition does not reduce the importance of getting the people side right. It raises it.
→ Talk to SRA about performance and engagement in the AI transition → Explore SRA’s review and engagement programs
What legal AI changes for the people side of a US law firm
Whichever platform a firm adopts, four people-side shifts follow predictably. Firm leadership should plan for them at the same time they plan the technology rollout, not eighteen months later when the consequences surface.
The throughline across all four: the metrics, evaluation criteria, and development structures most US firms use were designed around a model of associate work that AI is dismantling. A firm that adopts Harvey or Legora without revisiting how it measures, develops, and retains people is automating its operations while leaving its talent architecture pointed at a world that no longer exists.
How performance evaluation has to change in the AI era
The single most concrete people-side consequence of legal AI is that the traditional associate performance metric — billable hours and volume of output — becomes much less meaningful. If AI drafts the first version of a contract in minutes that used to take an associate hours, billing fewer hours is a sign the associate is using the tools well, not a sign of underperformance. A metric that penalizes efficiency is exactly backwards in an AI-enabled firm.
Firms that are adapting well are shifting associate evaluation toward dimensions that AI makes more important, not less:
Judgment and validation quality. Can the associate critically evaluate AI output, catch errors and hallucinations, and know when the AI is confidently wrong? This is now a core competency, and it is harder to develop than the routine work it replaces.
Substantive depth earlier. With routine work automated, associates are expected to engage with higher-level analysis sooner. Evaluation needs to assess whether they are actually developing that depth or simply passing AI output through.
Client and matter ownership. As technical execution is automated, the distinctively human contributions — client relationships, judgment under ambiguity, strategic thinking — become the basis for advancement. These need to be measured deliberately because they were previously assumed to develop on their own through years of routine work.
AI fluency itself. Ropes & Gray’s decision to make AI exploration billable is a recognition that AI fluency is now a measurable, developable, evaluable skill. Firms are beginning to assess it explicitly in reviews.
This is exactly the territory we cover in What Is the Difference Between a Performance Evaluation and a Performance Review at a US Law Firm? and in Attorney Performance Review: A Complete Law Firm Guide (2026). The AI transition makes the structured, multi-dimensional evaluation those pieces describe more necessary, not less.
A practical checklist for US firms adopting legal AI
If your firm is evaluating Harvey, Legora, or any legal AI platform, run the technology decision and the people decision in parallel. The checklist below pairs the two.
Firms that answer only the left column end up with capable AI tools and a talent architecture that quietly stops working. Firms that answer both columns get the efficiency gains and protect the pipeline that makes the firm worth something in a decade.
Frequently asked questions
Is Harvey or Legora better for US law firms? Neither is universally better. As of mid-2026, Harvey leads on scale, enterprise penetration (majority of the Am Law 100), and content partnerships (LexisNexis, Microsoft 365 Copilot). Legora is praised for collaborative workflow design, tabular review, and AI-native research. The right choice depends on your practice mix, existing systems, and how your lawyers actually work. Demo both against your own matters; do not choose on valuation or ARR.
Will legal AI replace junior associates? Not replace, but reshape. The consensus across 2026 legal-market reporting is that AI is automating the routine work juniors used to do, which compresses the traditional pyramid and changes what firms hire juniors to do. Goldman Sachs estimated AI could automate ~44% of US legal tasks. The firms handling this well are redesigning associate roles around supervised AI use, judgment, and earlier substantive responsibility, not eliminating junior lawyers wholesale.
How does legal AI affect the billable hour? It puts real pressure on it. The Thomson Reuters Institute 2026 Report on the State of the US Legal Market found that AI efficiency gains are creating tension between firm billing models and client expectations. When AI does in 10 hours what took 200, the hours-based revenue for that work drops sharply. This is pushing some firms toward value-based and alternative pricing, though the billable hour remains entrenched in complex, high-stakes matters.
Does SRA sell or recommend legal AI software? No. SRA designs and runs confidential performance review, evaluation, and engagement programs for US law firms. We are not affiliated with Harvey, Legora, or any legal AI vendor. Our role in the AI transition is helping firms answer the talent, performance, and retention questions that AI adoption creates and cannot solve on its own.
How should we measure associate performance now that AI does routine work? Shift evaluation away from billable-hour volume and toward judgment, AI-output validation quality, substantive depth, client and matter ownership, and AI fluency itself. A metric that penalizes the efficiency AI creates is backwards. The structured, multi-dimensional evaluation approach this requires is exactly what breaks down when firms rely on hours and volume as proxies for performance.
What’s the biggest people-side risk of adopting legal AI? The efficiency paradox: the same AI that makes the firm more profitable per matter today can erode the talent pipeline that makes the firm exist in a decade, because it removes the work that historically trained junior associates into partners. The risk is invisible until the cohort that should have become your future partners has already thinned out. The fix is to redesign development, evaluation, and retention deliberately alongside the technology rollout.
How does AI adoption affect associate engagement and retention? It raises both risks, especially among juniors who are anxious about their own futures. Associates watching routine work get automated reasonably wonder where they fit. Firms that measure and address that anxiety directly — through engagement surveys and clear communication about the associate’s evolving role — retain people through the transition better than firms that stay silent and let anxiety compound. We cover the measurement side in
Which Employee Engagement Software Should US Law Firms Actually Use in 2026?.
Sources
- Legora (March 2026). Legora Raises $550 Million Series D to Fuel US Growth. legora.com
- Harvey (March 25, 2026). Harvey Raises at $11 Billion Valuation to Scale Agents Across Law Firms and Enterprises. harvey.ai
- CNBC (March 25, 2026). Legal AI startup Harvey raises $200 million at $11 billion valuation. cnbc.com
- Silicon Republic (March 11, 2026). Legal AI start-up Legora hits $5.55bn valuation with latest raise. siliconrepublic.com
- LawFuel (January 10, 2026). Why This Big Law Firm Is Paying Junior Lawyers to Experiment With AI. lawfuel.com
- Axios (May 2, 2026). AI threatens Big Law’s talent pipeline. axios.com
- Thomson Reuters Institute & Georgetown Law (January 2026). 2026 Report on the State of the US Legal Market. abovethelaw.com
- The Agency Recruiting (May 2026). 2026 Legal Hiring Trends: AI Impact on Law Firm Staffing (citing Law360 Pulse March 2026 survey). theagencyrecruiting.com
Related reading on srahq.com
- → What Is the Difference Between a Performance Evaluation and a Performance Review at a US Law Firm?
- → Attorney Performance Review: A Complete Law Firm Guide (2026)
- → Which Employee Engagement Software Should US Law Firms Actually Use in 2026?
- → Partner Performance Review: How US Law Firms Evaluate Equity Partners in 2026
- → How Should US Law Firms Separate the Coaching Conversation from the Performance Review Record?
- → Which Legal Practice Management Software Are US Law Firms Actually Running On in 2026?
Legora and Harvey are both impressive, well-funded platforms reshaping how legal work gets done. Choosing between them is a real decision worth making carefully. But it is not the most important decision the AI transition forces on a US law firm. The most important decision is what you do about the people whose development, evaluation, and retention the technology is quietly upending.
SRA designs and runs confidential performance reviews, partner evaluations, upward review programs, 360-degree feedback, and firm engagement surveys exclusively for US law firms. We are not a legal AI vendor. We are the people-side partner that helps firms get performance, development, and retention right — which matters more, not less, as AI reshapes the work. Built for US law firms since 1987.
Performance Reviews | Upward Reviews | 360-Degree Feedback | Firm Engagement Survey | Schedule a Consultation
Exclusively serving United States law firms since 1987.
.png){kind=small}