SRA Issues Formal Warning Notice on AI Misuse After Receiving 42 Reports of Potential Breaches in 12 Months

The Solicitors Regulation Authority (SRA) published a warning notice to the solicitors' profession, disclosing that it received 42 reports of potential AI misuse between July 2025 and July 2026, with a number of investigations currently ongoing. The notice addresses two categories of risk: AI-generated hallucinations in legal research, advice, and court submissions; and confidentiality breaches arising from entering client information into AI tools without adequate safeguards. The SRA cited the case of Pinsent Masons, which self-reported after a junior solicitor used AI to draft two misleading letters in an insolvency application. The case is referred to by name in the warning notice as *Cork and another v Smith*, in which Insolvency and Companies Court Judge Mullen publicly admonished the firm. Sullivan and Cromwell is also referenced in coverage of the notice for a separate incident in which AI-hallucinated citations appeared in a filing to a New York judge. The notice states that solicitors remain accountable for their work regardless of how it was prepared, and that firms must have effective governance structures and controls in place to manage AI risks. On confidentiality, the SRA warned that entering client information into open-source tools such as ChatGPT would amount to a waiver of privilege, and that client data should only be entered into AI systems where appropriate contractual, technical, and organisational safeguards exist. The Bar Standards Board (BSB) had earlier in 2026 issued a similar notice warning barristers that free AI tools would generally be unsuitable for legal work. Law Society of England and Wales vice president Brett Dixon welcomed the notice.