Legal technology 
and AI

Chapter 1: The current state of AI in legal professions

Many firms can see that embracing AI is a journey they need to go on, helping them complete legal work faster and develop their business when they do. But often, they are waiting for others to make the first move.

When it comes to how the legal profession sees the potential of AI, the second annual Future of Professionals Report from the Thomson Reuters Institute reveals a striking finding. More than three-quarters of over 2,200 respondents (77%) said they believed AI would have a high or transformational impact on their careers, up from 67% who agreed with either sentiment when asked in 2023. And those who said transformational were 42% of respondents, up from 34% in 2023.

Regulations, staffing issues, the potential of a recession, generational leadership changes — these are all areas of concern for respondents, as per the report. But it’s AI that has most captured the industry’s attention, and not only that of working lawyers. Zachary Warren, Technology & Innovation Insights Lead for the Thomson Reuters Institute, has seen substantial interest in GenAI across the board, even from those who hadn’t before shown great interest in technological developments, “We’re seeing people from the C-suite, and law firm COOs and CFOs, all talking about generative AI for the first time.”

As per the Future of Professionals Report, AI is expected to affect the strategy of nearly all (94%) corporations. Its most significant areas of influence are likely to be in operational strategy (59% of C-suite respondents) and in product/service strategy (53%).

When asked what their highest priority is for the next 18 months, about half of legal respondents said it would be exploring AI’s potential and implementing it.

The rise of “fast followers”

The ground began to shift in 2022, with the widescale release of large language models (LLMs) powered by AI — that’s when, essentially, AI “went public.”

Before the rollout of these LLMs, “In the legal industry, if you wanted to use AI, you needed to develop a point solution,” Andrew Fletcher, Director of AI Strategy and Partnerships at Thomson Reuters Labs said, “You had to look at the data you had and use that to train a model to solve a particular problem. If you were using a general-purpose solution, it probably wouldn't have worked that well in your specific legal use case.”

Now lawyers are getting quality results even from out-of-the-box AI solutions. This is leading them to consider just how much more they could improve their efficiencies if they adopt professional-grade, legal-oriented AI systems.

“We’re seeing a lot of transformational change in a short period of time,” Warren said, “[AI] is going to have some shake-up effects on how professionals do their daily work, and it may change what it means to even be a lawyer, in some cases.”

To be sure, many legal professionals aren’t fully on board with AI yet. About 50% of respondents to the Future of Professionals Report said, that while they believe AI-powered technologies will typically provide a basic starting point, they also believe professionals will still need to do much of the work themselves. (Another 28% said AI was a strong starting point).

When asked what their highest priority is for the next 18 months, about half of legal respondents said it would be exploring AI’s potential and implementing it.

The rise of “fast followers”

The ground began to shift in 2022, with the widescale release of large language models (LLMs) powered by AI — that’s when, essentially, AI “went public.”

Before the rollout of these LLMs, “In the legal industry, if you wanted to use AI, you needed to develop a point solution,” Andrew Fletcher, Director of AI Strategy and Partnerships at Thomson Reuters Labs said, “You had to look at the data you had and use that to train a model to solve a particular problem. If you were using a general-purpose solution, it probably wouldn't have worked that well in your specific legal use case.”

Now lawyers are getting quality results even from out-of-the-box AI solutions. This is leading them to consider just how much more they could improve their efficiencies if they adopt professional-grade, legal-oriented AI systems.

“We’re seeing a lot of transformational change in a short period of time,” Warren said, “[AI] is going to have some shake-up effects on how professionals do their daily work, and it may change what it means to even be a lawyer, in some cases.”

To be sure, many legal professionals aren’t fully on board with AI yet. About 50% of respondents to the Future of Professionals Report said, that while they believe AI-powered technologies will typically provide a basic starting point, they also believe professionals will still need to do much of the work themselves. (Another 28% said AI was a strong starting point).

There’s also concern about to what extent AI should be used in legal practices. In a recent report on generative AI and professional services, a survey found that, while 85% of respondents believe GenAI could be used for legal work, only about 55% believe it should be used for legal work.

Yet when asked if they anticipated AI becoming a part of their workflow within 3 years, roughly 50% said yes, citing both open-source GenAI programmes like ChatGPT as well as legal-specific GenAI solutions.

Questions remain as to the accuracy, reliability, and security of AI systems, and how a lawyer using AI will remain in compliance with ethical and regulatory policies (see Chapter 2). And there also remains a lack of awareness in the legal sector about AI technologies and their potential. Only about a quarter of lawyers said they have used open-source solutions like ChatGPT, and only around 10% said they are actively using legal GenAI systems.

One way to describe these more reluctant law firms is as “fast followers.” Many aren’t eager to be the first among their peers to leap into AI. But they also want to move fast once the market shifts in that direction.

It’s a race to be second. It makes strategic sense for smaller law firms to keep an eye on their deeper-pocketed rivals to see how they adopt AI and what results they’re posting. The problem, however, comes when a “fast follower” waits too long to start its adoption. That’s why lawyers should be getting familiar with the practical usage of AI now. The more fluent they are with AI-fuelled technology, the easier their firm’s widescale adoption of the technology becomes — because the learning curve will get steeper every month.

Chapter 2: Ethical and regulatory considerations

Areas of concern for many legal professionals include whether using an AI-powered technology for a legal matter could compromise client confidentiality and data security, or expose lawyers to the risk that AI models could generate inaccurate results, contain inbuilt bias, or even “hallucinate” (i.e. make up answers). Questions over how AI models are trained and where liability lies if lawyers rely on AI as the basis for their advice, loom large.

It’s a murky picture, as in the UK, there is as yet no specific legislation around AI. Policy-makers have signaled that their cross-industry approach will be light-touch and revolve around the way AI is used, rather than focussing on the technology specifically — a stance that is echoed by the Solicitors Regulation Authority, which says:

“Our regulation focuses on the outcomes firms’ actions produce, not necessarily the tools they use to reach them. However, the use of advanced technologies can help firms meet their consumers’ needs more effectively and affordably. As such, supporting innovation and technology is one of our strategic priorities. We want to help firms and consumers safely gain the benefits that AI can bring.”

However, its latest Risk Outlook Report does provide some guidance, which includes:

• Firms must ensure that sensitive information is protected when using AI systems
• Firms will be held accountable for any errors in the advice given if it relies on AI outputs – monitoring and supervision are essential
• Staff must understand and should be able to explain how AI systems work and how they make decisions
• Clients should be made aware of how AI is being used in their matters
• Firms must stay abreast of different rules and regulations in all jurisdictions where they operate and be aware that they could be subject to rapid change
  • For example, the European Union has a more formal, risk-based legal framework of requirements. Its recent AI Act lists “assistance in legal interpretation and application of the law” as falling into the “high risk” AI usage category — one that will need to be registered in an EU database. Other EU laws could have a bearing on AI applications, notably the GDPR.

In the U.S., on the legislative front, there have been no notable Congressional actions to date. Guidance is slowly emerging on a state-by-state basis, while some courts have considered AI-related actions as well. This includes the U.S. District Court for the Eastern District of Texas, which adopted a local rule mandating that lawyers verify AI-generated content and attest that they have maintained “independent legal judgment.”

Developing AI policies

Questions also abound as to what constitutes an “ethical” AI usage policy in a legal context. Is a lawyer who uses a GenAI programme to advise a client employing a service that they and their clients can trust? Can a firm that uses GenAI charge a client the same rate to do work they had formerly done manually?

Legal departments are developing AI usage policies which can be summarised roughly as this: AI should be regarded as, “Very much a legal assistant. It’s there to be your co-counsel, working alongside you,” Fletcher said, “But you, as the legal professional, are the one who’s in control and ultimately making the decisions. You are the one who has built a relationship with your client, and you understand what's best for them.”

That’s why verifying AI-generated results for accuracy is essential. There’s the potential for a professional negligence claim if a law firm uses any GenAI programme that “hallucinates” —uses inaccurate or fabricated case law or citations for its findings. Having a strong peer review and fact-checking system is paramount for any legal operation using GenAI. A good way to regard an AI system is as: “People plus technology,” Warren said, “Not as a technology that’s in any way designed to replace critical thinking in lawyering.”

Each law firm will have to establish its red lines. What are the things that your firm doesn’t want AI to do? What would you consider to be a step too far for AI?

In a recent survey, Thomson Reuters found some consensus agreement that AI should not be used as a final arbiter for any legal action, while considered as being appropriate for lower-stakes work like document generation and research. Respondents said they felt secure about using GenAI tools for non-legal work within a law firm, such as question-and-answer services or administrative tasks. Almost 68% of respondents said they felt that GenAI should be applied to non-legal work within a firm, compared to 55% who said they felt it should be applied to legal work within the firm.

Chapter 3: Embracing technological evolution

In the next 3 to 5 years, it’s predicted that at least half of all legal professionals will be using some kind of GenAI-powered tool. So, it’s fair to say that there could be a cultural shift in this timespan as to how GenAI is regarded. The technology will go from being seen as an interesting option to being considered a must-have for a legal business — a primary means by which some types of legal work will get done.

A good comparison is the internet search engine. Back in around 1997, someone who used a search engine to look up facts, addresses, and businesses was experimenting with an interesting next-generation technology. By the mid-2000s, using Google had become a daily fact of life for every professional.

Law firms that have undertaken successful AI rollouts have several things in common. For one thing, they typically let their professionals take the lead in exploring AI systems. They encourage their legal teams to log all notable improvements in efficiency and productivity and share them with their colleagues. They’re letting their lawyers set the pace, not pushing them to use AI to meet any efficiency quotas.

For example, CoCounsel Drafting from Thomson Reuters uses professional-grade GenAI to help legal organisations overcome drafting challenges. Users have found that it enables them to cut drafting time by up to 50%. “There is substantial time savings when using CoCounsel Drafting,” J.J. Ball, legal counsel at Systemiq, said, “When we are operating on an average turnaround time of three to four business days for a response, we can cut that down to one to two business days by utilising the tool to get drafting work done.”

When it comes to ROI on time with CoCounsel Drafting, Ball adds, “The contract summary and clause summaries can be a massive time saver in higher-risk legal requests if leadership or GC wants a summary of a particularly large contract or contract with clauses.”

A good question is how law firms are going to verify and fact-check AI-generated sources of information. In some cases, there still isn’t enough quality data for fine-tuned legal AI training, as many firms face privacy restrictions in terms of using client data for training purposes. This will likely be one of the bigger challenges to the development of more sophisticated, legal-centered GenAI programmes.

Chapter 4: Uncovering AI potentials

AI technologies in the legal industry address the challenge of managing vast and complex information efficiently. Its various capabilities extend far beyond legal drafting tasks, offering innovative solutions that transform traditional practices into streamlined operations. Through AI processing and analysing large datasets, lawyers can quickly access critical insights, make more informed decisions, and navigate modern legal complexities with greater ease.

AI’s diverse use cases in legal work include:

• Summarising: AI tools condense lengthy documents into concise summaries, allowing lawyers to rapidly grasp essential information when dealing with extensive legal texts, contracts, or court opinions
• Legal research: AI's natural language processing capabilities allow it to search vast legal databases and identify patterns and trends, potentially predicting case outcomes and highlighting emerging legal issues
• Contract analysis and management: AI-powered tools review contracts swiftly. By identifying risks, inconsistencies, or non-standard clauses. This ultimately enhances accuracy and reduces human error
• Predictive analytics: by analysing historical case data, AI provides insights into the likelihood of success for legal outcomes and aids in the development of effective legal strategies
• Compliance and risk management: AI systems monitor regulatory changes in real-time, helping organisations stay compliant and mitigate risks proactively

The integration of AI into legal practices offers a wide range of significant advantages. Law firms can not only enhance their operational capabilities but also position themselves competitively in a challenging market. Due to the substantial time savings, they can have the bandwidth to focus on more strategic, creative, and interpersonal tasks. The future of law is undoubtedly intertwined with AI, promising a more efficient, accurate, and innovative legal sector.

The savings potential

Respondents to the Future of Professionals Report said they expected AI could save them about 12 hours per week in the next five years, or 4 hours per week over the upcoming year — that is, 200 hours annually. That’s the equivalent of adding a new colleague for every 10 team members on staff.

This could mean a substantial return on investment; for a UK lawyer, something along the lines of an estimated £75,000 in additional billable hours. But more value could lie in intangible improvements to the lives of legal professionals.

When asked what they wanted to do with those 200 saved hours, some respondents said they hoped to devote more time to strategic work and planning. Some wanted to use the time for business development and marketing. But the top answer (24%) was to achieve a better work-life balance. Many legal professionals simply would like to not have to work 80- or 90-hour weeks, and they see AI as a way to eliminate and streamline tasks that consume too much of their time in the office.

While the potential ROI, in terms of proposed cost savings, is impressive. Ultimately, AI could have the most value for a law firm with the firm’s personnel:

• More efficient means of onboarding new hires, both laterally and for newly qualified lawyers, particularly if the latter are already versed in using AI technology
• Reduced junior lawyer burnout, as associates won’t be drowned in repetitive, unproductive tasks
• More energised senior lawyers, who devote more of their time to client advisory and helping to grow the firm’s client base

Chapter 5: Strategic AI integration

Once a law firm commits to infusing its operations with GenAI technology, the next stage is critical to ensure that the technology is used properly and regularly. There are two essential elements to achieving this.

Integrating the technology into the firm’s existing tech stack, without experiencing substantial delays or causing excessive user confusion and frustration.

At the same time, getting the buy-in from everyone from senior partners downwards. Legal professionals need to be on board with incorporating the technology into their day-to-day work, not paying lip service to using it or regarding AI as a costly new imposition from above.

Also, integrating AI can pose a logistical challenge. Most law firms have tech infrastructures that their IT staff have built up over the past decade or longer. That’s why a law firm’s IT staff needs to be involved at the start of the integration process. The IT team needs to have all its questions answered: 

• How does the GenAI technology operate? 
• What sorts of resources will it require?
• What are its security protocols? 
• How do you best fold the technology into the firm’s existing tech stack?

There’s a core element of trust needed between the law firm and its technology providers. As Andrew Fletcher notes, “Law firms can be reticent to turn on some AI features because they want to be sure about what they’re going to do. There’s a need to have confidence in how those solutions are being delivered.”

Technology providers should be as transparent as possible about the role that AI is playing in their products. A good provider won’t obfuscate how its systems use AI — in fact, the very opposite is true. “Adoption becomes easier because of that trusted relationship and the fact that [AI] is sitting within the tools that you already are using,” Fletcher said.

Doing a seamless strategic integration is the best way to achieve a wide buy-in on GenAI among legal professionals. A GenAI system should be considered a vital piece of a legal professional’s desktop or laptop. What’s not ideal is for lawyers to regard “AI” as a stand-alone system they need to access separately from; for example, the small cognitive shift required whenever a professional must open a different window on their desktop to access an AI system may end up limiting their AI usage. 

The more that AI prompts are nested within a legal professional’s sidebar and considered as being as easy and intuitive as using spellcheck, the more that AI will become part of a lawyer’s daily work routine.

Chapter 6: The future of legal work

GenAI will shake some bedrock concepts in the legal industry, including what constitutes legal work and how law firms operate. Once GenAI usage becomes widespread, it will lead to a shift in mindset in how law firms hire and train staff.

Traditionally, a firm hires a junior lawyer. The entry-level lawyer must handle the firm’s time-consuming grunt work for 2 to 3 years, often needing to work long hours on repetitive tasks. Only after that time will they start to get involved in strategic decision-making, becoming part of deal teams, and working on more litigation, among other, more advanced, types of work.

The challenge is this: law firms will always have newly qualified lawyers that they want to groom into future partners. But if much of the “grunt work” becomes automated and substantially handled by GenAI systems, what is the role of these entry-level lawyers? Will the work of the entry-level lawyer of the future centre on being an evaluator and double-checker of AI-generated information? How will a law firm better incorporate its junior lawyers into more value-added client services?

There are many questions to consider, and not much consensus among law firms at the moment.

But the concept of considering GenAI as something to think about using down the road may no longer fly with a law firm’s clients. Roughly three-quarters of corporate clients surveyed said that they want their law firms to be using AI-powered solutions.

“What they want is for their litigators, their transactional lawyers, to put their critical thinking on top of the AI output, in order to get the best results possible,” Warren said, “By and large, clients are beginning to expect their law firms to be using these tools.”

This means a change in philosophy for law firms. For a long time, legal operations could afford to downplay and even ignore some technological developments. After all, their job was to serve clients, often using tried-and-true methods that hadn’t changed greatly in decades. The attitude among some managing partners was: “We're a law firm. We're not a technology company. We’re not in the business of technology.” The question has now become: “Are you sure about that?”

“Clients will always want expertise. And they will always want their work to be done cheaper, faster, and better,” Warren said, “How are you going to do that? Through technology. As a result, technology starts to become a differentiator. Who has the best use of technology to get to what their clients want?”

One reason why there are professionals across the board in the legal space interested in this technology is that: “There’s a growing recognition that for the law firm of the future, if they’re going to keep up with what their clients want and do it better than the firms they’re competing with, technology will, by necessity, play a big role,” Warren added.

Whether they want to or not, law firms are going to be in the technology business. It’s only a question of how.

Conclusion

AI is in the process of reshaping the legal landscape. As the technology continues to evolve, it has the potential to change relationships between law firms and their clients, between managing partners and junior lawyers, and between competitors in the legal space.

Law firms will have to be proactive. The future is coming, whether they like it or not. It makes sense for a firm to get on board before its competitors do. Solutions like CoCounsel offer one way for a law firm to confidently embrace and adapt to what is going to be a generational change in legal operations.