Law firms raced to adopt AI after developments in technology enabled them to draft contracts, assist with due diligence processes, and draft legal opinions.
The launch of the ChatGPT natural language chatbot in November marked a significant breakthrough in generative AI. Built by Microsoft-backed OpenAI, the bot produces human-like, convincing sentences, using large language models to predict the likely next word in a sequence.
It led to other major tech companies, including Google and Microsoft itself, quickly following suit. Startups have also leveraged the underlying technology used in these products to develop specialized artificial intelligence for legal services.
Law firms and consulting firms are now using the software to automate tasks and drive efficiencies, driven to cut costs by falling revenues amid a corporate drought. Magic Circle law firms and Big Four accounting groups have pioneered AI platforms built for legal tasks such as drafting contracts, translating documents into multiple languages, and proposing legal opinions.
“We get several thousand queries a day and it’s fairly consistent across offices . . . in multiple languages [and] different areas of the law,” says David Wakeling, head of Allen & Overy’s Markets Innovation group, which includes both lawyers and developers. The law firm was the first Magic Circle firm to adopt generative AI and used a homonym produced by a US start-up called Harvey AI from November. “AND 1683416603 an important part of the operating model,” he explains. “We are well beyond the process.”
Harvey was built using GPT language models created by OpenAI, which also invested in the US start-up. In addition to the general internet data that underpins the GPT, Harvey is trained in legal data, including jurisprudence. The system advises A&O lawyers to verify the content it creates, as generative AI is notorious for “hallucinating”: confidently stating things as fact, despite there being no basis for them in reality.
“It’s a blank page, first stab quick,” Wakeling says. “You know you will always change it; it’s never good enough. But [if] you apply it to 3,500 people, which is a significant time-saver; an hour or two a week is no big deal.
Nearly half of all current tasks in the legal profession could be replaced by artificial intelligence, according to a recent study relationship by Goldman Sachs. Automating them would remove the need for humans to do some of the more administrative and mundane jobs, although it could also mean that trainee lawyers and law school graduates may no longer experience this.
“We need to think about how to train young lawyers,” says Kay Firth-Butterfield, head of artificial intelligence at the World Economic Forum. “They may not all immediately be able to advise on the more complex issues or have complex meetings with clients or think of ways to challenge the status quo. This must be learned.
He sees limits to the use of AI: “Because all Generative AI can do is look at historical data for answers, we need human advocates to make sure we continue to expand and advance the law so that it doesn’t atrophies”.
Protagonists argue that AI-powered tools will enable lawyers to do more skilled work and provide strategic advice, saving time and reducing costs for companies and clients.
“It significantly reduces billable hours,” says Richard Robinson, founder and chief executive officer of Robin AI, which launched in 2019 and provides AI-powered legal software. But he points out: ‘The best companies want to get paid for high-level strategic work, things that fundamentally, at least today, no AI is trying to replicate – like high-level negotiations, insights into what happened in other [similar] business on the market”.
$1 billion
PwC investing in AI to automate parts of its audit, tax and advisory businesses over the next three years
Robin AI works with two of the Big Four accounting firms, as well as private equity funds and the law firm Clifford Chance. It sells software and also offers an additional service, where its team of 30 in-house lawyers and paralegals oversee the results generated by the AI. Its technology can also be used to scan legal documents to assess risk exposure.
But Robinson warns that “these tools are basically not ready to be used without people safeguarding them,” stressing that the pure output of such technologies should always be checked and modified by a qualified expert.
Large language models get better as more data is entered and as they are used more. PwC, which uses Harvey for mergers and acquisitions, due diligence and contract drafting, says it has had an influx of clients who say they want to adopt AI but are concerned about data protection.
“Data privacy and security is paramount and really important. . . because the data is sensitive and there is a legal privilege,” says Sandeep Agrawal, partner in legal and tax services at PwC. Agrawal recently met with Harvey executives to discuss how it can isolate data and encrypt information, the which means it’s more secure.Harvey separates all customer data and offers encryption tools to protect access to customer information.
In a sign of growing confidence in the technology, PwC last week pledged to invest $1 billion in AI to automate parts of its audit, tax and advisory businesses over the next three years.
Jerry Ting, chief executive officer of contract management firm Evisort and a lecturer at Harvard Law School, reports on a similar shift towards AI adoption in recent months. Evisort, launched in the US in 2016, offers AI software that allows clients to create and manage contracts, including drafting and signing, through an automated process.
“Before GPT, it was: This is what AI is, this is why it benefits you; it was almost necessary to convince them,” she says. “Now they show up at the door already convinced. The question becomes, ‘how do I use it in a way that is safe, that actually drives my business results, and is within my budget?’”
This article has been updated to indicate that Evisort is a contract management company
—————————————————-
Source link