Not using AI could make lawyers negligent, taskforce says

TL;DR:

• The UK Jurisdiction Taskforce says professionals can be sued for negligence for failing to use AI where a competent peer would have.
• Its new legal statement concludes existing English law already resolves most AI liability disputes, so no new legislation is needed.
• Liability can equally arise from careless use: skipping due diligence, not telling clients AI was involved, or not checking output for errors and bias.

Lawyers and other professionals can be sued for negligence for failing to use artificial intelligence, according to a new legal statement from the UK Jurisdiction Taskforce (UKJT). The finding reframes AI not as an optional efficiency but as a professional tool whose non-use may itself fall short of the required “reasonable care and skill”.

A duty that cuts both ways

The UKJT’s Legal Statement on Liability for AI Harms treats AI like any other tool available to a professional: the question of whether it should be used, and how, is judged against what a reasonable practitioner of comparable rank and specialism would do. That test turns on professional bodies’ regulations and guidance, and on expert evidence about competent practice in the field. The same duty of care runs in the other direction. The taskforce set out examples where careless use creates liability: failing to carry out proper due diligence, failing to explain to clients how AI was being used, or failing to check output for errors and biases, the familiar problem of confidently wrong “hallucinated” results.

Looking forward

Crucially, the UKJT concluded that existing English law already provides the framework to resolve most AI liability disputes, and that no new legislation is required, a notably restrained position as governments elsewhere reach for bespoke AI statutes. That restraint carries weight: the taskforce noted its earlier statements on cryptoassets and smart contracts were “adopted directly by English judges” and relied on by international investors choosing English law for technology deals. For UK firms, the practical message is uncomfortable but clear. The debate has moved past whether to adopt AI; competence itself is starting to assume it. That mirrors warnings elsewhere in professional services that trust, not technology, is the real barrier to scaling AI, only now the risk of standing still has a legal name attached.