Fashionable authorized AI device Harvey will now be utilizing main basis fashions from Anthropic and Google, transferring past strictly utilizing OpenAI’s, Harvey introduced in a weblog put up on Tuesday.
That is noteworthy as a result of Harvey is among the OpenAI Startup Fund’s most profitable early-backed portfolio firms. The OpenAI Startup Fund is an OpenAI-associated fund to again firms creating merchandise on high of AI applied sciences, primarily OpenAI’s personal. Whereas Harvey says it’s not abandoning OpenAI, merely including extra fashions and clouds, that is nonetheless an enormous coup for OpenAI’s massive rivals.
Harvey is among the first 4 startups that the OpenAI Startup Fund backed, it stated in December 2022. That was again when OpenAI CEO Sam Altman was nonetheless operating the fund. (The others in that first cohort embrace Descript, Mem, and Communicate.)
Harvey, which has grown like mad since then, is now a $3-billion-valuation startup, it stated in February, when it introduced a $300 million Collection D led by Sequoia, with different massive names like Coatue, Kleiner Perkins, and the OpenAI Fund piling in.
Apparently, Google’s enterprise arm, GV, led Harvey’s $100 million Collection C in July 2024 (and the OpenAI Fund participated in that spherical, too). However Harvey didn’t instantly undertake Google’s AI fashions after it put Google’s company enterprise agency on its cap desk. (GV participated in Harvey’s Collection D as properly.)
So, what satisfied Harvey to maneuver past OpenAI’s fashions now? The startup’s internally developed benchmark, dubbed BigLaw, confirmed that all kinds of basis fashions are rising more and more adept at a variety of authorized duties and a few are higher at particular duties than others.
As an alternative of spending its efforts coaching fashions, Harvey figured, it may merely embrace high-performing, reasoning basis fashions from different distributors (e.g. Google and Anthropic through Amazon’s cloud) after which fine-tune them for the authorized market.
Utilizing quite a lot of fashions may also assist as Harvey creates AI brokers, the corporate says.
“In lower than a 12 months, seven fashions (together with three non-OAI fashions) now outperform the initially benchmarked Harvey system on BigLaw Bench,” Harvey wrote within the weblog put up.
Harvey’s benchmark additionally confirmed that completely different basis fashions are higher at particular authorized duties than others. For example, it says Google’s Gemini 2.5 Professional “excels” at authorized drafting however “struggles” with pre-trial duties like writing oral arguments as a result of the mannequin doesn’t absolutely perceive “advanced evidentiary guidelines like rumour.”
OpenAI’s o3 does such pre-trial duties properly, in accordance with Harvey’s testing, with Anthropic’s Claude 3.7 Sonnet following shut behind.
In its weblog put up, Harvey says that it’ll additionally now be a part of the rising ranks of these sharing a public leaderboard of mannequin benchmark efficiency. Its board will rank how main reasoning fashions are doing on authorized duties. And the corporate received’t simply boil rankings all the way down to a single quantity, however may also publish analysis the place “high legal professionals present nuanced insights into mannequin efficiency not captured by single-score benchmarks.”
So, not solely is OpenAI-backed Harvey adopting rivals’ fashions, additionally it is amping up the stress on its backers (together with Google) to maintain proving themselves. Not that OpenAI ought to fear a lot on that rating. Whereas AI benchmarking is rising more and more advanced and considerably political, this can be a world the place OpenAI nonetheless shines.
“We’re extremely lucky to have OpenAI as an investor in Harvey and key collaborator in our product,” Harvey CEO Winston Weinberg instructed TechCrunch in an announcement. “And, we’re energized so as to add to our choices for purchasers as we proceed to serve the wants of our prospects globally.”
{content material}
Supply: {feed_title}
