Bar Standards Board publishes AI guidance for barristers after fabricated-citation case

TL;DR:

• The Bar Standards Board has published guidance on AI adoption among barristers, framed around a tiered risk model: low-risk administrative tasks, medium-risk legal research, and high-risk court submissions or work with vulnerable clients.
• The guidance follows the Ayinde v London Borough of Haringey case, in which a barrister submitted court documents containing fabricated case citations and an inaccurate legislative summary generated by AI.
• Barristers are now expected to maintain “a basic level of general technology and AI competence” and tell clients when AI “materially impacts the nature or scope of your legal service”.

The BSB intervention completes the picture of UK legal-sector AI regulation begun by the Solicitors Regulation Authority and the Law Society. Each branch of the profession now has formal guidance, the underlying risks identified by each regulator are broadly the same, and the duty of competence in existing professional rules has been extended to cover AI tool selection and oversight. The Ayinde judgment provided the political momentum; the practical effect will be felt in chambers’ supervision systems and continuing-professional-development programmes through the rest of 2026.

A risk-based framework, not a prohibition

The BSB has not banned AI use, nor has any UK legal regulator. The tiered approach mirrors the Information Commissioner’s Office framing and the SRA’s existing risk-management orientation: barristers can use AI for diary management or initial document drafting, but court submissions and work involving vulnerable clients sit at the top of the risk pyramid where verification expectations are non-negotiable.

The guidance also asks barristers to evaluate potential bias in AI systems, ensure data-governance protocols are appropriate, and disclose AI use to clients when it materially affects scope. Ewen MacLeod, the BSB’s director of strategy, policy and insights, said the guidance is designed to help barristers “adopt technology responsibly while preserving professional standards” and noted that “effectively procuring, adopting, and using technologies takes skill and resources” — a quiet acknowledgement that smaller chambers lack the procurement capacity larger sets enjoy. UK legal regulators are converging on substantively similar positions: tool selection sits inside the existing duty of competence, verification of AI outputs is the barrister’s responsibility, and client communication thresholds apply to material impact.

Looking forward

Expect chambers to revise their AI-use policies and supervision arrangements before the next bar-practising-certificate cycle. The SRA’s parallel guidance applies to solicitors, the BSB’s covers barristers — the gap that remains is among litigants in person, where fabricated-citation incidents are likely to multiply faster than any regulatory body can address. For UK businesses procuring legal services, the practical change is that AI disclosure expectations are now embedded in regulator guidance; ask the question in instructions. The fee implications of pre- versus post-AI drafting will become a contested point in legal-services pricing through 2026.