As AI continues to shape the legal profession, its integration must be approached with caution and a clear understanding of the limits and responsibilities involved. The legal industry is one of the most heavily regulated professions, and that fact alone places constraints on how generative AI can be used in legal practice — particularly if lawyers are to uphold their duties to clients.

The use of advanced technologies in such a highly regulated environment will inevitably require increased regulation to ensure transparency, fairness, and compliance. Once generative AI became widely available, it became clear that it could not be 'put back in the box', and regulators have since been working to keep pace with its rapid development.

As the technology landscape evolves, legal professionals will need to stay informed about emerging tools and concepts. This will likely include some retraining, with legal education and professional development shifting to focus more on technology, including AI, data analytics, and cybersecurity. In this context, AI literacy may soon be as essential as proficiency in traditional tools like Microsoft Office.

Despite the growth of AI, the value of core human skills — such as communication, empathy, and client management — remains undiminished. These skills are central to effective legal practice and cannot be replicated by machines.

Some law firms have responded to the emergence of generative AI by restricting its use by staff — for example, international firm Hill Dickinson did so at the start of 2025. However, such restrictions may prove counterintuitive and ultimately unworkable. Clients and other stakeholders are increasingly drawn to firms that are embracing modern technologies. In reality, staff may resort to using AI tools on personal devices if official channels are closed. The more sustainable approach is to encourage responsible, compliant use of AI and to leverage insights from its usage to improve workflows.

To implement AI effectively and ethically, law firms must introduce robust internal policies to safeguard sensitive data and prevent misuse. Clients deserve transparency about how AI is being used and why — they should understand what they are paying for. In addition, anyone using AI in legal contexts must be adequately trained. This includes an understanding of concepts like AI bias and hallucinations — what they are, how they arise, and how to mitigate them. AI literacy, in this sense, is not optional — it is foundational.

For that reason, the remainder of this section will take you through four core areas of AI literacy: bias, hallucinations, environmental concerns and whether or not we can trust AI.