A survey conducted by Lexis Nexis found that 95% of UK lawyers anticipate Artificial Intelligence (AI) will significantly influence the law, with 38% considering it significant and 11% transformative. The influence of AI, particularly ChatGPT, extends across all sectors, including the legal field, which despite its values of nobility and tradition, demonstrates a willingness to embrace technological advancements. However, some legal professionals continue to remain sceptical about adopting AI.
Law firms are actively exploring the potential of generative AI in performing menial tasks like drafting, summarising, and comparing documents, aiming to enhance services, cut costs, and meet client needs. Even judges are embracing AI chatbots, as Lord Justice Birss recently disclosed using ChatGPT to summarise an area of the law for a judgment. However, there are instances, like that of Steven Schwartz, who infamously relied excessively on ChatGPT, resulting in a court filing containing fabricated cases, rulings, and quotes. In my view, the responsibility and liability rested with the lawyer for failing to verify before filing considering AI is merely a tool, not a substitute for legal expertise.
In his book, “The End of Lawyers?”, Richard Susskind, a leading figure in legal tech, argues that the widespread adoption of AI won't replace lawyers but will compel them to master skills in risk management, design thinking, and data science. Currently, most lawyers lack training in these areas, but in the future, they will need to grasp at least the fundamentals and collaborate closely with AI specialists. This possible shift suggests that technological requirements may become more prominent than client needs and the law at large, potentially complicating rather than simplifying the legal profession.
Another issue with integrating AI into the legal sector is the absence of regulation. Currently, there are no specific regulations in the UK governing AI or generative AI, nor are there statutory requirements for generative AI technology firms to verify the factual accuracy of their output, rendering them unreliable and insecure. Furthermore, non-legal organisations such as JPMorgan and KPMG have already prohibited the use of ChatGPT due to privacy and ethical concerns.
Adding complexity is the human element inherent in lawyer-client interactions which AI cannot replicate. Clients seek not only legal knowledge and expertise but also trust, empathy, and understanding from their lawyers, particularly in sensitive criminal and family law cases. As Giulia Gentile suggests, the relationship between lawyers and clients could be compromised by robot lawyers or excessive reliance on AI tools, as they lack the empathy, human connection, and legal creativity of human lawyers.
Many law firms have started developing their own dedicated generative AIs tailored to their specific legal needs, enhancing reliability. However, access to such resources remains limited, particularly for small to medium-sized firms, which struggle to keep pace with these advancements despite efforts by the Law Society to promote the legal-tech ecosystem. The long-term impact of AI on the legal sector remains uncertain, as lawyers are just beginning to explore its vast potential with many of the current AI products, including the widely used ChatGPT, still in their early stages of development.