Caroline Hearn of Pinsent Masons raised the prospect after a special committee of legal experts issued its view on liability for AI harm.
The draft legal statement published by the UK Jurisdiction Task Force (UKJT) is open for consultation until February 13, but is not binding on courts. The UKJT aims to provide some clarity on how emerging technologies interact with current legislation. No views are expressed on areas where there is room for reform. In this case, that statement is designed to help you understand the circumstances in which your company may be liable for, or be able to bring claims against others, for AI-related harms under private law in England and Wales.
This statement is premised on the basis that AI systems do not themselves have legal personality under UK law and therefore cannot be held legally responsible for physical or economic harm. However, according to the UKJT, individuals and businesses can be held liable for harm related to the use of AI in a variety of circumstances.
Some of the situations in which the Task Force believes liability for AI damages may arise relate to professional negligence. Professional negligence is a concept that describes when professionals, such as lawyers, architects, accountants, and doctors, fail to perform their obligations to others with reasonable skill and care. If this happens, you may be subject to liability.
Mr Hearn said: “As AI is widely deployed in professional services, professionals are seeing an evolution in the terms of engagement to reflect how these tools are used in client matters. While some clients are accustomed to the controlled use of AI in a secure environment, others prefer to prohibit it. UKJT’s draft legal statement highlights issues that are consistent with our experience, particularly around confidentiality, privilege and the secure deployment of AI tools.”
“The ability to use AI effectively in professional situations depends on many factors, including robust due diligence, high-quality prompts, appropriate oversight, and clear internal processes. When companies develop standardized prompts and workflows, they improve efficiency and consistency, but they also come with risks. As with any standard format, errors can quickly spread throughout the issue if the approach is not properly designed and monitored,” she said.
Hahn said expectations regarding the professional standards that apply to the use of AI in litigation processes may change over time, and failure to meet those standards could have cost implications.
“In high-volume exercises, such as disclosure, where speed and proportionality are central, parties may agree to an AI-based review methodology,” Hahn said. “If both parties agree to a particular approach, this may influence how the professional standards of reasonable skill and care are applied in the circumstances.”
“UKJT has also raised the possibility that not recommending the use of generated AI for disclosures to provide a more efficient process could be considered negligence in the future, but this remains an issue.In many cases, however, this is due to clients reluctance to pay for time-consuming and costly manual reviews, resulting in legal complications. Beyond efficiency, there is a further question, which remains highly speculative, of whether, in some circumstances, lawyers could face claims because AI may have identified fraudulent documents. “These will be difficult discussions to implement in practice, but they illustrate how expectations around the use of AI will change,” she said.
“More broadly, UKJT highlights that reasonable skills and professional standards of care will continue to change as AI tools advance. What is reasonable now may change rapidly as new capabilities become mainstream. “It is important to keep abreast of industry guidance,” Hearn added. “However, it is important to remember that even if a court later decides that the approach falls below the required standards, widespread market practice does not necessarily insulate professionals.”
“As AI becomes more integrated into professional workflows, courts are likely to increasingly look to expert evidence to assess whether AI was or should have been appropriately used in a particular matter. The nature of these experts, and how they evaluate that evidence, is likely to evolve as both technology and professional practice continue to evolve,” she said.
