Driving widespread AI adoption across the legal sector

Winning strategies for knowledge management teams at law firms to drive AI engagement, mitigate risk, and prove return on investment.

AI use is accelerating inside law firms, but the heat is now on legal knowledge managers to ensure widespread adoption.

We look at the top strategies knowledge managers are deploying to encourage AI use, mitigate risk, and prove return on investment.

Our September 2024 survey of legal professionals in UK & Ireland on their use of generative AI found:

AI adoption accelerates across firms

Four out of five private practice lawyers use or plan to use generative AI.

AI adoption amongst law firms is accelerating at pace. The number of private practice lawyers in the UK using generative AI for work purposes has almost quadrupled in a little over a year, a new LexisNexis survey has found.

AI users at firms jumped from 11% in July 2023 to 42% in September 2024. However, the remaining majority have yet to touch AI tools in a work capacity.

This is likely to change in the next few months. The percentage of private practice lawyers planning to use AI sits at 44%. Only 13% said they have no plans to use AI.

86% of lawyers now use or plan to use AI for work

Jonathan Kewley, Partner and Co-Chair of the Global Tech Group at Clifford Chance, says the level of adoption taking place across the legal sector is extraordinary.

"The kind of applications we're seeing – the way it's been rolled out, the innovation, the investment – it's off the scale."

While AI is a useful tool in the medium term, Kewley believes its use will only accelerate in the longer term. 

"It's important that lawyers are familiar with AI technology in the same way they are with the internet now. Those that don't use it will face a bit of an existential risk."

Laura Hodgson, Generative AI Lead at Linklaters, believes technology's disruption of the legal sector was only a matter of time.

"The legal profession has remained largely unaffected by new technology for several decades and was overdue for a shake-up."

Hodgson added that while AI tools are still in the early stages, they're already adding real value to how lawyers work.

Encouraging AI adoption

The most effective change and innovation techniques at leading firms.

The survey revealed 60% of the legal industry have made at least one internal change to implement generative AI. This reflects the growth of this rapidly evolving technology.

The most common change was offering an AI-powered product to staff, which rose from 15% in January 2024 to 36% by September 2024.

There was also significant growth in developing policies on the use of generative AI (11% to 24%) and providing AI-related training for staff (11% to 18%).

2/3 of organisations in the legal sector have made a change to implement generative AI

As expected, large law firms are the most likely to have made a change to drive AI forward, at 78%.

"We have been early adopters for many of the next-generation solutions we use and have regular roadmap discussions with our key vendors" says Greg Baker, Global Head of Practice Innovation at Linklaters.

Kewley reveals they've made the decision to share AI tools with everyone at Clifford Chance, and actively encourage them to use it.

"If people aren't using AI, we're starting to follow up and ask them why, because we really want them to benefit from it."

Encouraging AI adoption needs to come from the top down, says Joe Cohen, the Director of Innovation at Charles Russell Speechlys.

"We have been encouraging all of our staff to use AI. Senior leadership have done a great job of speaking about their own, often daily, use of it."

This top-down approach is worth investigating. Law firm leaders are significantly more likely to be using generative AI than associates. Our survey revealed 49% of law firm leaders are using AI, compared to only 31% of associates.

Cohen told us nearly half of their 1,200+ staff are using AI monthly, and the technology has provided answers to tens of thousands of queries over the last couple of months.

"Firmwide training, regular communications and a comprehensive engagement plan to support awareness and adoption has helped us greatly."

I want to view the Artificial intelligence (AI) resource kit

Kewley at Clifford Chance says that the firm also actively encourages its employees to share noteworthy use cases for AI.

"We've got a hub where people can submit their use cases and we're finding really great examples of diverse and innovative AI deployment."

This collaborative approach was true with most of the firms we spoke to, with training sessions, use-case or prompt engineering workshops, and open forums all now commonly practiced.

The use cases for AI in the practice of the law are already immeasurable. Keeping pace with this change will mean more time, resources and training.

I want to learn more about Lexis+ AI

Mitigating the risks of AI

Human verification is crucial to mitigating risk.

The risks for lawyers relying on incorrect information are enormous. Just ask the US legal team who cited fake cases in court.

Despite increasing adoption rates, three-quarters (73%) of private practice lawyers are concerned about inaccurate or fabricated information from public-access generative AI platforms.

Relying on fabricated information (hallucinations) is the biggest concern to lawyers using AI

Knowledge managers need to tackle this topic head on, empowering lawyers to feel confident navigating the risks that come with AI.

Beckhaus from Freshfields says the firm uses a host of technologies, processes, and controls to proactively mitigate and minimise AI risks, including automated tools to manage and monitor the AI model lifecycle.

"The most important element of our approach, however, is the ‘lawyer in the loop’ principle and human centered legal AI," he continues.

Can the use of generative AI lead to trade mark infringement?

AI can help speed up research and drafting, but the human lawyer is equally responsible for the output, says Dr. Katy Peters, a Law Lecturer in Professional Legal Practice at the University of Surrey.

"Whilst it may no longer be necessary to spend hours in a library or searching an online database, it will still be necessary to create appropriate prompts, review responses, adapt templates and challenge discrepancies."

I want to learn about AI and the need for explainability

We asked respondents how confident they would be using a generative AI tool that was grounded on legal content sources, with hallucination-free, linked citations to the verifiable authority used to generate the response.

Three-quarters (72%) said they would feel more confident using a generative AI tool grounded in legal content sources with linked citations to verifiable authorities, up from 65% in January 2024. 

Three-quarters of lawyers feel more confident using AI grounded in legal content sources

Interestingly, as adoption rates and usage continues to climb, over-reliance is a growing concern. Over-reliance on AI is playing on the minds of half (50%) of lawyers.

I want to know how Lexis+ AI speeds up legal research safely

Unsurprisingly, leaking confidential information (49%) and accidental bias (43%) were also valid concerns.

Transparency is crucial, believes Patel from Eversheds Sutherland. "AI needs to provide clear indications of where responses are derived from, enabling experts to review and verify them."

Tart-Roberts from Macfarlanes says training staff to use AI is also critical to minimising risk.

"Educating users about the types of tasks that are best suited to generative AI and how to write effective prompts goes a long way towards mitigating risk."

LexisNexis has now launched Lexis+ AI in the UK, which searches, summarises, and drafts using LexisNexis content.

Grounded in LexisNexis' enormous repository of accurate and exclusive legal content, Lexis+ AI combines the power of generative AI with proprietary LexisNexis search technology and authoritative content. Results are always backed by a verifiable, citable authority or source.

This tool was built with the RELX responsible AI principles in mind, says LexisNexis' Senior Director of Segments, Stuart Greenhill.

"Everything we do considers the real-world impact of the solution. Lexis+ AI proactively prevents the creation or reinforcement of bias, we ensure that we can always explain how and why our systems work in the way they do, human oversight is built in and we respect and champion privacy and data governance in all that we do."

Future-proofing AI into the legal toolkit

While lawyers are currently using generative AI as a standard practice for everyday matters, Hodgson at Linklaters says the long-term impact AI will have is still to be fully-recognised.

"It is only at the point that generative AI is built into tools across the lawyer’s tech workbench that clients will notice a significant change in service delivery."

Freshfields has a cross-functional team in place that seeks to match the firm's AI needs with the legal capabilities of new AI developments.

"We collect the requirements from our colleagues and clients and scan the market globally for AI products which meet business needs," says Beckhaus from Freshfields.

"We also explore potential partnerships for our GenAI journey, and work with several large language models on defined use cases ourselves."

Final thoughts

Knowledge managers at medium and large sized firms have clearly gone above and beyond when it comes to enacting their AI adoption strategies. The high AI use rates across respondents from those firms is evidence of this.

But to prove However, there are concerns about inaccurate or fabricated information, over-reliance on AI, leaking confidential data, and unintended bias.

To address these risks, firms are implementing policies, training programmes, and using AI tools that are grounded in authoritative legal sources with verifiable citations.

Additionally, the shift towards AI-driven efficiency may lead to changes in pricing structures, with a potential move away from the billable hour model towards value-based billing.

The legal market is undergoing a significant transformation, with firms embracing AI to streamline operations and enhance service delivery while grappling with the challenges of maintaining accuracy, confidentiality, and ethical standards.

Those firms that effectively navigate this transition and leverage AI's capabilities while upholding professional integrity are likely to gain a competitive edge in the evolving legal landscape.

Survey methodology

The survey was conducted across 803 lawyers and legal support workers in the United Kingdom and Ireland from August to September, 2024. Surveys were conducted in English.