SRA Issues Warning Notice on the Misuse of AI in Legal Practice

Against a background of increasing judicial scrutiny of artificial intelligence (AI) generated errors in legal proceedings, the Solicitors Regulation Authority (SRA) has published a warning notice (link: SRA | Misuse of AI – Warning notice | Solicitors Regulation Authority) highlighting the risks associated with the misuse of AI in the delivery of legal services. Published on 17 August 2026, the warning notice applies to all SRA-regulated firms and individuals and serves as a reminder that professional and regulatory obligations remain unchanged when AI tools are used. The SRA stated that it had received 42 self-reports relating to potential misuse of AI between July 2025 and July 2026 (link: SRA | SRA cautions profession about safe and responsible use of AI in legal sector | Solicitors Regulation Authority).

The SRA identifies two key areas of concern. First, the growing number of instances where AI-generated content has contained inaccurate information, including fabricated case citations and legal authorities, commonly referred to as ‘AI hallucinations’. The SRA makes clear that solicitors and regulated individuals remain accountable for the accuracy of all documents and submissions, regardless of whether AI was used in their preparation. It has stated that AI has “no separate legal personality”. Courts have also made it clear that reliance on AI is not a defence for presenting false information. In fact, the SRA has warned that placing false information before the court may be considered contempt of court, potentially constituting  interference with the administration of justice. Consequently, it is more important than ever for regulated individuals and firms to implement robust verification procedures and ensure that all authorities, citations and legal references are independently checked before being relied upon in legal advice, correspondence or court submissions.

Secondly, the SRA warns of the risks to client confidentiality. Information entered into the public domain or inadequately secured AI systems may be retained, processed, or used by third-party providers, creating potential breaches of confidentiality, legal professional privilege, and data protection obligations. The warning notice states that misuse of AI may result in legal professional privilege being permanently waived. Firms are expected to understand how AI tools handle data and to ensure appropriate contractual, technical, and organisational safeguards are in place before using them with client information.

The warning notice emphasises that solicitors must continue to provide competent legal services, effectively supervise staff, maintain appropriate governance and risk management systems, and uphold their duties to clients and the courts. While AI may enhance efficiency and productivity, it does not transfer or reduce professional responsibility.

The legal sector is continuing to embrace AI but innovation must be balanced with professional responsibility. The SRA’s warning notice serves as a timely reminder that firms should regularly review their use of AI tools, governance arrangements and internal controls to ensure they remain compliant with their regulatory obligations while continuing to benefit from technological advancement.