With the increase in the use of artificial intelligence (AI) in everyday situations, more and more our clients are supplementing their instructions to us or responding to our advice with the assistance of AI.
Whilst there are obvious benefits to efficiency by incorporating AI in this way, judges are increasingly warning parties of the potential dangers and unintended consequences of crossing litigation with open-source AI platforms.
Wake Smith Solicitors’ associate Andrew Donaldson, who has a particular interest in tech and AI matters, looks at how client’s use of free, publicly accessible platforms can have serious legal and practical consequences.
Andrew is keen to spread the word about the increased risk of clients using open/free AI platforms when analysing or responding to legal correspondence and advice. He advises clients about this regularly.
He said: “It is crucial to use AI tools responsibly and, with the relevant guardrails in place, to avoid compromising the confidentiality and accuracy of legal advice.
“For example, litigation privilege exists to protect clients from having to disclose material created for the purpose of advising on a dispute and communication with their advisors regarding the same. Unfettered use of free AI platforms might compromise this right for certain documents and advice.”
Andrew added: “The bottom line is solicitor clients ought to be considerate of what they are inputting into free open-source AI platforms.
“Whilst there is no binding judgment which deals with this activity, the court has indicated the risks around preserving confidentiality, data protection and legal privilege. It is therefore prudent to either use paid-for platforms which preserve privacy of prompts/attachments, or exercise high levels of caution on what content you are feeding into a free, non-confidential platform.”
To conclude, uploading legal advice, confidential or privileged documents or personal data into open AI systems is generally not recommended because of:
- Waiver of privilege: AI systems do not protect legal advice as they may be equivalent to publishing it and waiving the party’s right to avoid its disclosure.
- Breach of confidentiality: A party with an obligation to keep documents or information confidential may unintentionally breach their obligation to keep that material confidential.
- Data risks: Public AI systems can process and store inputs, bringing concerns about data privacy and confidentiality.
Controversially, back in November 2025, judges in the Upper Tribunal case of UK v Secretary of State for the Home Department [2026] UKUT 81 (IAC) noted: “Uploading confidential documents into an open-source AI tool, such as ChatGPT, is to place this information on the internet in the public domain, and thus to breach client confidentiality and waive legal privilege, and any such conduct might itself warrant referral to the regulatory body and should, in any event, be referred to the Information Commissioner’s Office.”
Andrew added: “The Upper Tribunal’s view is not binding, but a fair warning of the risk of waiving or breaching confidentiality, privilege and data protection that free/open-source AI large language models pose.”
To understand the range of comprehensive legal advice on business and personal matters Wake Smith Solicitors’ offer call 0114 266 6660 or click here.