Perspective.

https://www.bespacific.com/harvey-openai-and-the-race-to-use-ai-to-revolutionize-big-law/

Harvey, OpenAI, and the race to use AI to revolutionize Big Law

Fast Company – no paywall: “…Harvey, which is used by some 250 law firms—including 42% of The American Lawyer’s list of biggest 100 firms in the U.S.—announced in June that it raised a $300 million series D led by Sequoia, bringing its total haul to more than $800 million and driving its valuation to $5 billion. It’s now the highest valued startup in a growing field of AI companies that are all focused on overhauling how the legal profession works. The company’s annualized revenue run rate hit $100 million in August, up from $50 million earlier this year, propelled by an aggressive sales strategy targeting big law firms. Harvey now has 460 employees, 20% of whom are lawyers. The company is also hiring dozens of engineers, sales leads, and account executives as it seeks to increase the moat between itself and its competitors. Whether Harvey ultimately upends the legal procession or winds up burning a lot of cash, time, and effort could reveal that fates of industries as far afield as finance, music, and film. All these sectors—and more—are on a similar curve: exploring whether AI tools can be truly transformative and seeing just how many jobs they’ll reimagine—or disappear altogether… Despite its sizable moat, Harvey’s success is far from assured… Even Harvey’s cofounders have called OpenAI an indirect competitor. But critics also say the company could be steamrolled by OpenAI, pointing out that the LLM could simply build its own legal-focused model. “I’m hearing from more and more attorneys that OpenAI’s Deep Research is the single best research product on the market, and it’s what most attorneys use (even when it’s not a firm approved tool),” former lawyer and legaltech investor Zack Abramowitz wrote on his popular substack Legally Disrupted in June… Harvey’s customers, in the meantime, seem satisfied. “[Harvey is] totally embedded in the workday of our lawyers. It’s really become embedded throughout their daily workflows,” says Gina Lynch, chief knowledge and innovation officer at Paul, Weiss, Rifkind, Wharton & Garrison, the white-shoe law firm with more than a thousand lawyers. As Ethan Mollick, a professor and codirector of the Generative AI Lab at Wharton, wrote on Linkedin, “There was a moment that we thought proprietary data would let organizations train specialized AIs that could compete with frontier models. It turns out that probably isn’t going to happen. The largest frontier models are just much better at most complex tasks than smaller models.”

Another concern is how long Harvey can maintain its lead, given the cost of its product. Harvey’s bespoke services cost $1,200 per seat, per month, with contracts stipulating that large companies need to purchase the service for at least 100 employees and for at least a year. The company justifies its prices by touting the productivity gains it passes onto employees. “You can arm lawyers with tools that make them enormously productive,” says Harvey’s chief business officer John Haddock, who has a law degree from Stanford. And $100,000 a month is still less expensive than the salaries of an army of paralegals and junior associates.