OM Feature Operations Management

The State of AI and Legal

We take a closer look at the real world implications of AI and its use in the legal industry.

Artificial intelligence (AI) has been discussed — and even used — in legal technology circles for at least a couple of decades. But it’s only since the public release of ChatGPT, less than two years ago, that AI has received significant attention from the legal sector. A wide array of AI-infused software products is being harnessed by an increasing number of legal organizations, ranging from international players to boutique practices.

Alex Heshmaty

The most basic AI tools, such as the autocorrect function of Microsoft Outlook, have been used by the vast majority of firms for many years. And most practice and case management software incorporate certain features that are marketed as using AI (although many of these are also pretty basic).

For purposes of this article, we’ll focus on a more sophisticated form of AI: generative AI (GenAI). Some of the well-known GenAI products currently being deployed in the legal sector include Microsoft Copilot, Lexis+ AI, Thomson Reuters CoCounsel and Harvey. That said, there are many other providers, and new legal AI companies seem to be mushrooming as a result of an investment windfall over the past 12 months.

According to a report from LexisNexis, around a quarter of lawyers regularly use GenAI tools, compared to just over 10% a year ago. Still, almost 40% of legal professionals surveyed said they had no plans to adopt GenAI. For legal administrators and other legal support staff, these figures offer a useful barometer of the sentiment toward AI within the legal sector as a whole — one of clear division.

Nevertheless, many firms are keen to test out the technology, and recent McKinsey research research noted that “the biggest increase in adoption [of AI] can be found in professional services” of which the legal sector plays a big part.

Will Richmond-Coggan, Partner at Freeths, says his impression is that law firms generally have been fairly cautious about rushing to adopt AI technology, particularly generative AI solutions, but he still sees potential.

“Looking ahead to that future, I generally see the opportunity presented by AI as being a positive one for law firms, helping them to run more efficiently and freeing up lawyers to deal with the intellectually demanding and time-consuming work of finding innovative solutions for their clients, with confidence in the tools that they are using for the purpose,” says Richmond-Coggan. “But there are a number of important caveats to that — in relation to the propensity of many tools to hallucinate (or produce inaccurate or incomplete, but highly convincing, outputs); in relation to the security and integrity of client information which might be shared with the solutions; and in connection with the environmental and societal impact of the way such tools are trained and operate, which is frequently very difficult to get transparency about.”

In this article, we’ll consider the impact of AI on the business of law, highlighting some of the advantages as well as these challenges facing those tasked with implementing the new tools. We’ll get insights from law firm insiders about the real-world implications of AI for the legal sector, while looking toward what’s in store for the future.

DOES AI REALLY TURBOCHARGE EFFICIENCY?

A recent report from Thomson Reuters claims that AI could free up around four hours each week for an average U.S. lawyer, equivalent to 266 million hours of increased productivity which “could translate into $100,000 in new, billable time per lawyer each year.” But according to the LexisNexis survey, just over half of in-house counsel expect bills to reduce due to GenAI, in contrast to only 40% of lawyers in private practice.

But these figures raise some questions: Who stands to benefit from the implementation of GenAI in the legal sector? Will efficiency gains boost profits, reduce client bills or provide a better work-life balance for fee earners? Or will the cost of licensing GenAI products and training staff to use them offset any actual advantages for law firms or their clients?

“It is difficult to predict how quickly AI will change the legal sector, but increasingly we won’t be able to ignore its impacts,” says Paul Philip, the Chief Executive of the Solicitors Regulation Authority (which regulates lawyers in the United Kingdom). “So far it has mainly been larger firms using AI. However, with such technology becoming increasingly accessible, all firms can take advantage of its potential. There are opportunities to work more efficiently and effectively.”

“Sooner or later, the technology will have been refined to the point where it becomes genuinely valuable in supporting firms who want to deliver efficient, accurate solutions to their clients. At that point, we will want to be ready to take advantage of those opportunities, which means getting our people familiar with the tools now, the ways that they work, and the risks to be on their guard against, ahead of that point.”

What’s certain is that legal management professionals and those responsible for running law firms will need to weigh up the pros and cons of implementing AI to decide whether there is a genuine business case in their particular circumstances. In other words — will it be profitable (or otherwise beneficial) for their individual firm, or could it end up being more hassle than it's worth?

“Our firm has not been slow to get to grips with the opportunities presented by this new technology — albeit in carefully controlled sandbox conditions and through hackathons and other isolated exploratory projects,” says Richmond-Coggan. “Sooner or later, the technology will have been refined to the point where it becomes genuinely valuable in supporting firms who want to deliver efficient, accurate solutions to their clients. At that point, we will want to be ready to take advantage of those opportunities, which means getting our people familiar with the tools now, the ways that they work, and the risks to be on their guard against, ahead of that point.”

LEGAL AI: A MARRIAGE OF TRADITION AND MODERNITY

Considering the historically reserved and cautious nature of the legal profession — and its relative slow uptake of previous incarnations of technological innovations such as email —  it’s interesting how eagerly many firms are diving into the AI pool. According to analysis from Mordor Intelligence, legal AI software is expected to grow from $2.19 billion in 2024 to $3.64 billion by 2029. Harvey alone was recently reported to be seeking a $2 billion valuation.

But just because many firms are investing in AI, this doesn’t mean it’s being embraced with open arms by all and sundry. A whopping 90% of respondents to the LexisNexis survey admitted to having at least some reservations about AI being adopted by law firms, with over a quarter admitting to significant concerns about the technology.  It’s worth noting that these figures were almost identical to the findings of a similar survey from the previous year, indicating that AI marketing efforts are not winning over the naysayers.

Some of the biggest fears relate to AI hallucinations, security (e.g., data protection), and simply untrustworthy tech. In fact, a recent study from Stanford University found that GenAI software specifically targeted at the legal sector exhibit hallucinations (i.e., where they essentially make up facts) in at least one in every six queries.

Although the commercial legal AI tools that tend to garner the bulk of publicity are those geared toward lawyers, the technology is arguably currently more useful to other professionals working in law firms.

“Given the blandness of the output it is not something we would [use to] replace human ingenuity … we strive to be just that little bit different from the rest of the pack.”

Legal administrators who are responsible for maintaining client records can gain valuable insights — particularly in relation to new business opportunities — if they use AI software to help analyze the firm’s troves of data. Meanwhile, legal marketers can harness GenAI for inspiration when creating blogs and social media content. (Although they would still be wise to have a legal copywriter check the resulting copy for hallucinations and make sure data they are putting in is info they don’t mind being out in the public sphere.)

Aside from GenAI being used within firms to assist lawyers with drafting documents (the aforementioned LexisNexis report notes that document drafting is the main use case for 90% legal sector users), law firm marketing and PR managers will be interested to know that general communication activities — such as content marketing and press releases — accounts for 73%.

Brian Inkster, of Inksters Solicitors in Scotland, is less optimistic about the future of AI within the legal sector, particularly as it relates to content creation. “Given the blandness of the output, it is not something we would [use to] replace human ingenuity … we strive to be just that little bit different from the rest of the pack.”

Instead, he says they are focusing on what they have and ensuring it works efficiently for them.

“We are not seeking to harness AI in any specific way. Instead, we are concentrating on our already planned roadmap for making our existing legal technology (case management and document automation systems) work to their full potential,” says Inkster. “There is much for us still to do on that front without distraction by shiny new toys. On the face of it, such toys are unlikely to take us anywhere near where our existing systems will, via consistent implementation and improvement.”

HOW SHOULD LAW FIRMS PREPARE FOR AI?

Despite all the AI hype over recent months, it’s probably wise for law firms to adopt a cautious approach in the short to medium term. As Inkster notes above, there are already many highly effective software tools that are already licensed by firms, with further abilities to be unlocked by exploring their potential.

Having said that, business managers in firms may well be able to maximize efficiencies by deploying AI effectively in a targeted approach — and Richmond-Coggan believes that it will gradually reduce much of the routine grunt work. So perhaps a certain amount of cynicism mixed with a dash of hope and an open mind to genuinely helpful AI tools is the best way to approach this emerging technology.