Artificial intelligence is now part of everyday business life.
Business owners, directors and professionals increasingly upload contracts, emails, legal advice, witness statements and confidential communications into platforms such OpenAI’s ChatGPT or Anthropic’s Claude to:
- summarise documents;
- analyse disputes;
- prepare responses;
- draft correspondence;
- review contracts; or
- obtain quick explanations.
Many clients do not realise that uploading legal advice or confidential documents into AI systems may create serious legal and commercial risks.
In some circumstances, it could potentially:
- compromise confidentiality;
- waive legal professional privilege;
- expose commercially sensitive information;
- create regulatory or data protection issues; and
- complicate future litigation or arbitration.
Recent UK and US cases show that courts are beginning to examine these risks far more closely.
For businesses involved in disputes, investigations, shareholder conflicts, arbitration or regulatory matters, this is becoming increasingly important.
A common scenario
A client receives advice from their solicitor regarding:
- a shareholder dispute;
- a threatened claim;
- an arbitration;
- a settlement proposal;
- allegations of misconduct; or
- a contractual dispute.
The client then copies that advice into ChatGPT or Claude and asks: “Please summarise this advice.” Or “Tell me if my solicitor is correct.”
At first glance, this may appear entirely harmless. However, from a legal perspective, an important question arises on whether the client has potentially disclosed privileged and confidential information to a third party.
That question matters because legal professional privilege depends heavily on confidentiality. If confidentiality is lost, privilege may also be lost.
What is legal professional privilege?
Under English law, communications between a client and their solicitor for the purpose of legal advice are generally protected by legal professional privilege.
Privilege is extremely important because it allows clients to:
- speak openly with their lawyers;
- discuss sensitive issues honestly;
- prepare litigation confidentially; and
- seek legal advice without fear of disclosure.
Privileged documents are usually protected from disclosure to opponents, regulators or courts.
However, privilege is not unlimited, and one of the core requirements is confidentiality.
If privileged material is voluntarily disclosed to third parties in circumstances inconsistent with confidentiality, privilege may potentially be waived.
This is where AI platforms create significant legal and commercial concern.
Recent judicial decisions in both the UK and the United States demonstrate increasing scrutiny of AI usage within legal and commercial contexts.
In UK and R (on the application of Munir) v Secretary of State for the Home Department [2026] UKUT 81 (IAC), the Upper Tribunal criticised the use of AI-generated material containing hallucinated legal authorities.
Importantly, the Tribunal also commented on confidentiality risks associated with uploading legal documents into open-source AI systems.
The Upper Tribunal stated that uploading client letters and Home Office decision letters into an “open source AI tool” effectively places the information into the public domain, potentially breaching client confidentiality and waiving legal privilege.
The Tribunal distinguished between:
- open or public AI systems that may retain, process or utilise uploaded data; and
- restricted or enterprise systems with contractual protections and closed environments.
This distinction is increasingly important.
Many businesses and professionals assume that using an AI platform is equivalent to using a standard software tool. That assumption may be dangerous. Remember that the fact that a platform is marketed as “secure” or “enterprise-grade” does not automatically resolve privilege concerns.
Businesses that adopt AI without proper governance may expose themselves to avoidable legal, regulatory and evidential risks.
At the same time, organisations that understand the technology, implement appropriate safeguards and maintain proper human oversight are likely to gain significant operational advantages.
How we can help
At IMD, we advise businesses on complex commercial disputes, arbitration, technology-related risk and contractual matters.
We assist clients with:
- technology and AI-related disputes;
- confidential information issues;
- commercial litigation and arbitration;
- SaaS and technology contracts; and
- dispute prevention strategies.
As AI adoption accelerates, businesses increasingly require legal advice that combines commercial understanding with practical knowledge of emerging technology risks.
This article is for general information only and does not constitute legal or professional advice. Please note that the law may have changed since this article was published.
