AI is beginning to change the business of law - Ars Technica

Overview

Attorneys are finding uses for AI apart from generating fake case quotations.

In spring 2024, two days after undergoing complex cardiac surgery in the Midlands, a man in his mid-seventies unexpectedly deteriorated and died.

Details

The hospital referred the death to the coroner’s service, as is protocol when a cause is unknown, and clinical negligence barrister Anthony Searle was instructed by the man’s devastated family to represent them.

To try to get to the bottom of what had happened, Searle knew he would need to ask the surgeons some probing questions. So when the coroner declined his request for an independent expert report, Searle was frustrated.

Instead, he turned to another resource that is proving increasingly useful in the chronically underfunded coroners’ courts: AI.

“Deaths that go to inquests are there because they are a shock. What families want is to increase their understanding as to how their loved one has died,” says Searle. “My use of Chat GPT allowed my questions to be more focused on the technical aspects of the surgery and help fill the gaps left by not having experts to call upon.”

Searle, 35, is at pains to emphasize that he does not put any client data or information into the AI tools he uses, and vets all of the information and citations the bot spits out.

Yet his caution belies his status as an early adopter tentatively putting AI to use at the centuries-old Bar, work that could push the institution at least partially into the modern era.

Potential uses for the technology may eventually include assisting barristers’ clerks, who negotiate fees and arrange their diaries, so they can make better use of advocates’ time. The possibility of AI helping to draft skeleton arguments, the case summaries presented in court, has also been mooted.

In addition to research, Searle is using the technology to create bespoke AI tools, including an app for calculating damages in clinical negligence claims. The app analyses data from the actuarial tables English courts use to work out a person’s future losses as a result of injury, and calculates more precise estimates taking into account factors such as age and lost pension contributions.

Searle’s initiative in taking up the technology means he has got involved in developing broader AI governance strategies for expert witnesses in clinical negligence cases and at his top London chambers, Serjeants’ Inn. “This is an ancient profession,” he says. “We’ve been around for hundreds of years, and I think much like the common law we like things to develop only incrementally.”

Despite the traditionally slow pace of change, Searle is far from the only lawyer experimenting with new technology. In England’s underfunded justice system, AI is increasingly being offered as an answer to problems like court backlogs and a lack of resources for those involved in trials.

Proposed government reforms, which would mark the biggest overhaul of the criminal justice system in modern times, include plans to roll out AI in courts for listing cases, translation and transcripts, among other uses.

In a sign of how fundamental AI is to the government’s vision, deputy prime minister and justice secretary David Lammy delivered his key speech on court reforms last month at a Microsoft AI event in London—a setting that would have seemed incongruous only a few years ago. Courts minister Sarah Sackman has described AI pilots as “game changing.”

Where Google searches and You Tube videos used to be the go-to resource for Searle to get to grips with complex medical procedures, AI offers a juiced-up version that can help improve the types of questions lawyers like him need to be asking in such cases.

Chat GPT and custom programs in the medical space, such as Pub Med in Anthropic’s Claude, allow Searle to search quickly for clinical studies and articles from medical journals.

The tools are a way of more efficiently understanding and gathering information on the complex topics he has to litigate every day, he says. He has noticed when speaking to experts that some appear more impressed by questions since he has taken on his automated research assistants.

Like many sectors, however, AI has also become a buzzword in the legal world. Almost all large law firms are drafting grand plans for its use yet show little evidence of how it is in fact disrupting the multibillion-pound sector.

“If there were a ranking for the most au courant and widely discussed topic among law firm leaders over the last few decades, AI would win hands down. But so far the evidence is that that’s all it is: discussed,” says Bruce Mac Ewen, president of New York law consultancy Adam Smith, Esq. “How can firm leaders commit to anything so unproven and so consequential? Clearly, no one knows where this is going.”

UK law firm Shoosmiths added £1 million to its bonus pot last year as a reward for staff hitting 1 million prompts on Microsoft Copilot, while US firm Ropes & Gray is pushing junior lawyers to spend a fifth of their billable hours on the technology, using it for research and experimenting with how it can speed up tasks such as contract drafting. Elite outfit Clifford Chance cut 10 percent of its back office at the end of last year, partly in anticipation of AI taking over some tasks.

However, any great cull of lawyers appears years away. Client confidentiality and data protection limit how much AI can currently be used in reality.

The industry is also on high alert after London’s High Court handed down a scathing judgment last year citing two cases containing false information in which barristers had, or were suspected of, using AI. In one case, as many as 18 fake case citations were used.

“Artificial intelligence is a tool that carries with it risks as well as opportunities,” the judges said in their ruling last June. “There are serious implications for the administration of justice and public confidence in the justice system if artificial intelligence is misused.”

In a January 2026 survey, 50 percent of barristers said they were using AI for legal work, up from 25 percent in 2024, yet only 2 percent said it was embedded in their legal operations and strategy, according to legal data analytics provider Lexis Nexis, which ran the questionnaire.

For Searle, who acts both for and against bodies like the NHS and litigates everything from maternal care failures to surgeries gone awry, it is the medical knowledge provided by AI that is particularly worth having.

“When the legal profession . . . are talking about the risks of AI it’s always in the context of legal research and applying the law,” he says. “But most of my work is not about the law, the law is settled… the law is not controversial. Really what it’s about is the medicine and the arguments that can be made in relation to this particular diagnosis or treatment or surgery.”

Searle uses the same custom medical programmes that AI platforms have created for medical students and professionals. He says such custom tools have become integral to his practice as a lawyer.

“I think we’re probably already at the stage where AI can challenge people’s judgments and opinions,” says Searle.

But, he cautions, “the legal system and the rule of law is about humans… There still needs to be that degree of empathy and judgment from a human to ultimately make the decision.”

1.          Once again, ULA can't deliver when the US military needs a satellite in orbit
   

2.          There can (still) be only one: Highlander is 40
   

3.          You're likely already infected with a brain-eating virus you've never heard of
   

4.          We keep finding the raw material of DNA in asteroids—what's it telling us?
   

Ars Technica has been separating the signal from the noise for over 25 years. With our unique combination of technical savvy and wide-ranging interest in the technological arts and sciences, Ars is the trusted source in a sea of information. After all, you don’t need to know everything, only what’s important.

Key Takeaways

• Attorneys are finding uses for AI apart from generating fake case quotations
• In spring 2024, two days after undergoing complex cardiac surgery in the Midlands, a man in his mid-seventies unexpectedly deteriorated and died
• The hospital referred the death to the coroner’s service, as is protocol when a cause is unknown, and clinical negligence barrister Anthony Searle was instructed by the man’s devastated family to represent them
• To try to get to the bottom of what had happened, Searle knew he would need to ask the surgeons some probing questions
• Instead, he turned to another resource that is proving increasingly useful in the chronically underfunded coroners’ courts: AI