Pinsent Masons Self-Reports to SRA After Junior Lawyer Used AI to Generate Fabricated Law in Court Letters, as London Judge Declines to Initiate Contempt Proceedings
Pinsent Masons has referred itself to the Solicitors Regulation Authority (SRA) after one of its junior lawyers used artificial intelligence to generate fictitious legal citations that were then included in letters sent to court. Judge Mark Mullen of the Insolvency and Companies Court (High Court of Justice, England and Wales), on being informed of the incident, declined to initiate contempt of court proceedings. The case in which the fabricated citations appeared was a block transfer application, where the AI cited an Insolvency rule from 2016 that had nothing to do with block transfer applications. The self-referral is a significant act of regulatory transparency from a Magic Circle-adjacent firm. Proactive engagement with the SRA — reporting the incident before being required to do so — is consistent with the obligations on regulated entities to cooperate with their regulator and may be a factor the SRA weighs in deciding how to respond. The firm's decision to self-report also reflects a broader awareness that AI-generated hallucinations (fabricated legal references produced by AI tools that present false information as real) represent a professional conduct risk requiring immediate escalation when discovered. The incident at Pinsent Masons mirrors a pattern seen in other jurisdictions — most notably in the US, where several firms have faced sanctions after AI-generated citations were submitted to federal courts — but this appears to be one of the first instances of a UK firm taking the formal step of self-reporting an AI misuse incident to the SRA. The London judge's decision not to initiate contempt proceedings suggests the court was satisfied by the firm's corrective response, though any SRA investigation and its outcome remain pending.
Why this matters
Pinsent Masons' self-referral to the SRA over AI-generated fabricated law is one of the most consequential professional conduct developments in UK legal AI to date. It establishes a precedent that even junior-level AI misuse that reaches a court — rather than being caught internally — will be treated as a regulatory event requiring disclosure. For law firms, this story underlines the inadequacy of purely aspirational AI usage policies: without technical controls, supervision protocols, and training that creates genuine awareness of hallucination risk, similar incidents at other firms are likely. The SRA's response will be closely watched as a signal of how strictly it intends to regulate AI-assisted legal work.
On the Ground
A trainee working on AI governance matters for a firm following an incident like this would be reviewing and marking up AI governance policy drafts against the SRA's published guidance, assisting with regulatory notification submissions, and preparing vendor due diligence questionnaires for the AI tools in use across the firm.
Interview prep
Question you might get
“What professional conduct obligations does a UK solicitor have when they discover that AI-generated content in a court document contains fabricated citations, and what steps should a firm take to manage the regulatory consequences?”
Sign up free to see the full answer
A model answer you can lift into an interview — how to frame this story for a partner.
Sign up freeSources
My notes
saved