Big Law's outlaw has something to prove

Silicon Valley loves a rivalry, and right now one of its most closely watched showdowns is unfolding in an unlikely arena: corporate law.

Harvey and Legora, two artificial-intelligence startups selling software to law firms, are jockeying for customers, credibility, and bragging rights. With this week's news that Legora raised $550 million in an investment round valuing the company at $5.5 billion and acquired the startup Walter, the Sweden-based company is one step closer to challenging Harvey's dominance.

Max Junestrand, Legora's 26-year-old wunderkind founder with no legal training, isn't going to settle for runner-up.

When he video-calls me from a bare Stockholm apartment that he hasn't had time to furnish, he mentions that he's coming off a 72-hour, 100-degree fever, and it's blunted his usual zip. Still, he's serious about his mantra. "We try to operate like elite swimmers," he says, "trained to look down and swim — not to look at the sidelines."

Whatever Legora tells itself internally, the competition is playing out in public. Even their board members have joined the spectacle, trading barbs online over which company is pulling ahead. In one recent exchange, Sequoia partner Pat Grady argued there was no contest, saying Harvey added more revenue in the fourth quarter than any competitor has in total. Legora's chairman, Benchmark's Chetan Puttagunta, jeered, "the lady doth protest too much, methinks."

The investor sparring is the beef's public version. The private version happens in conference rooms, where lawyers still have to be convinced to trust software with their work.

The Harvey-Legora clash is more than startup theater. It's a high-stakes proxy war for how quickly AI can impact one of the world's more conservative industries. As billions of venture dollars flood into generative AI, rival founders and their investors are racing not just for customers, but for proof that AI can crack the professional services economy. Legal services alone hit $1 trillion in revenue worldwide last year, according to Precedence Research.

In Silicon Valley, these head-to-head combats are increasingly defining the AI era: fast followers challenging early leaders, VCs publicly staking claims, and customers buying more than one product because no one is sure who will lead six months from now.

In the early days of Legora, Junestrand cold-messaged lawyers on LinkedIn, offering to pay their hourly rate for a meeting. He's since channeled his charisma into the competition.

"He's just a fireball," says Michael Gerstenzang, a senior partner at Cleary Gottlieb, one of the marquee firms using Legora.

On paper, though, the Harvey-Legora matchup is pretty one-sided. Forbes reports Harvey could be worth $11 billion in a new fundraise. In the scramble for Big Law's business, the competition looks already won. Of the 100 highest-grossing law firms in the United States, Legora says it counts about 20% of them as customers. Harvey has more than half of these firms using its software. Legora won't share its revenue, while Harvey says last year it booked about $190 million in annual recurring revenue — the income a company collects from subscription fees over a year.

If Harvey showed that large language models could be packaged into software lawyers would actually use, Legora is betting it can out-execute the early market leader. Over the last year, Legora has increased its customers from 250 to 800, and became the fastest startup out of the famed Silicon Valley startup incubator Y Combinator to reach unicorn status.

On the "20VC" podcast recently, Junestrand described his competitive point of view: "There's only winning. Everything else is losing."

Legora arrived after Harvey and, at first, built a familiar kind of virtual paralegal. Now Junestrand argues the roles are reversing and competitors are copying Legora's features. He doesn't name names, but Harvey launched a client-facing portal about a month after Legora rolled out its own. To Junestrand, that's not a threat, but confirmation that Legora is starting to set the pace.

"Imitation is the sincerest form of flattery," he says.

At Legora's first board meeting in New York City, when the company had about 12 employees, Junestrand made a surprising announcement. He told his investors that Legora would not sell its product at all for the next six months. Junestrand explained that, since Harvey was already in the market, Legora had to bring an excellent product. If they delivered a demo that wasn't fully baked, they would "burn that one shot" with firms.

The reaction in the boardroom was mixed. The company had just raised $35 million. Halting sales right after a round was a risk, one that looked like it would starve Legora of the early customer love that it needed to compete. Still, Benchmark's Puttagunta firmly believed that a company doesn't need to be the first to market as long as it is the best. History tends to reward that kind of execution: Google eclipsed Yahoo; Uber outran Lyft. Eventually, the other investors agreed, and the team spent the next few months retooling the code.

It wasn't just the specter of Harvey that Legora had to overcome. Not long ago, selling any new software to law firms looked like a losing business. Lawyers lived in documents. Before large language models, software was lousy at pulling data it could read from messy text, which meant most "legal tech" tools could only search documents by keywords.

There were other limitations. Lawyers have long bristled at the idea of machines doing their work. Clients pay for judgment, for bespoke advice, and for the human who will stand behind a call when it goes sideways. More practically, the billable hour model that fuels law firms' coffers made efficiency unattractive. Software that saves time can also shrink the number of hours a firm can charge for.

When large language models landed four years ago, suddenly the tech could do something that earlier tools couldn't: read, summarize, and rewrite legalese in plain English. This sent up a signal flare that even the most change-averse of industries could actually be disrupted by this technology.

In 2023, Junestrand, along with his cofounders Sigge Labor and August Erséus, began sketching out an idea for a new startup. It would do what lawyers do: comb through data rooms for red flags, compare contracts, research, and draft briefs. If it worked, Junestrand reasoned, the company, then called Leya, stood to become the central platform for all legal work. What Figma is to designers and Salesforce is to sales reps, Legora would be to lawyers.

Junestrand didn't look like the guy to tell elite lawyers how they should work. He grew up in the Stockholm archipelago, where his mother worked in pharma, and his father owned a boat business. He taught himself English by grinding through "World of Warcraft" quests, bought a sailboat at 16, and even modeled briefly as a teen, appearing on posters around Stockholm advertising an ABBA museum.

"My modeling career did not last very long," he said, grinning.

Junestrand juggled courses at the prestigious KTH Royal Institute of Technology and Sweden's top business school while working full-time in venture capital. It only worked because the pandemic pushed classes online, though he eventually dropped out of KTH to focus on Legora.

As Legora's profile rose, so did Junestrand's. He prophesied the end of the billable hour on the tech-friendly talk show "TBPN," made the Forbes 30 Under 30, and hobnobbed with the French president to talk about European technical ambition. Junestrand now splits his time between Stockholm and New York, becoming a kind of emissary for Sweden's Silicon Valhalla.

On a Friday afternoon, I stopped by Legora's office just south of Union Square. Patrick Forquer, the executive in charge of revenue, led me past rows of empty desks to the kitchen for a LaCroix.

A few employees hunched over monitors as lo-fi beats thumped from a speaker. Most workers, Forquer explains, were on the road trying to win over firms. Otherwise, Legora is rigidly in-person, five days a week, he says, on the belief that the most important work happens live and together. To keep people around, catered dinners arrive at 8 p.m. in the Stockholm office.

Forquer said the selling part is relatively easy. Right now, many law firms feel they have to at least look fluent in artificial intelligence, because corporate clients are asking what tools they're using. Gerstenzang, the Cleary Gottlieb lawyer, views the pressure as an existential threat. He warns that if top-tier firms don't figure out how to do the routine work better, faster, and cheaper using technology, clients will simply take that work elsewhere.

Lawyers tell me their firms are buying licenses to many platforms, keeping their options open. Clients may arrive with preferences about their firm's tech, and partners don't want to be locked into a tool that could look second-rate months from now. As one law-firm executive put it: "We are in the fortunate position that we now have the Coke and Pepsi of our industry."

Even as law firms pour money into new tools, many still can't say what they're getting back. The return on the investment shows up in lawyers moving faster and less busywork. What's harder is turning that into something a managing partner can point to on a financial report or a client bill. That raises questions about how long firms will keep paying for "nice to have" productivity.

Junestrand knows this market is just getting started, meaning no lead is stable. If Legora can take market share from Harvey, another company can do the same to Legora.

"In a way," he says, "we're playing the infinite game here."

Have a tip? Contact this reporter via email at mrussell@businessinsider.com or Signal at @MeliaRussell.01. Use a personal email address and a non-work device; here's our guide to sharing information securely.