By Brodie Pearson

Solicitor

Corporate & Commercial Law

28 October 2026

Article Blog

Artificial intelligence (AI) is becoming an increasingly common feature of everyday business, from drafting documents and analysing data to automating customer services and generating content. While businesses are embracing the opportunities AI presents, their existing commercial contracts may not adequately address the risks that come with it.

Contracts negotiated before the widespread adoption of AI may contain familiar confidentiality, intellectual property (IP) and liability provisions, but do these provisions still provide sufficient protection?

Confidentiality and data: where is your information going?

Many AI tools rely on user inputs to generate outputs, and some may retain or use submitted information to improve their models. Where a supplier uses AI to deliver services, businesses should consider whether the contract permits customer information to be inputted into third-party AI systems.

Existing confidentiality clauses may not expressly address this practice. Businesses should consider whether restrictions on disclosure extend to AI tools and their providers, whether customer data can be used to train models, and what safeguards apply to personal or commercially sensitive information. Contracts should clearly establish when AI use is permitted, what information may be processed and what security measures are required.

Intellectual property: who owns the output?

AI-generated content raises questions about ownership, permitted use and infringement. A contract may state that a customer has a license to work produced by a supplier, but that wording may not resolve every issue concerning AI-generated material, particularly where the output incorporates third-party content or the legal protection available for AI-generated works is uncertain.

Businesses should review ownership and licensing provisions to ensure they address AI-assisted deliverables and provide sufficient rights for the customer’s intended use. They should also consider whether existing IP warranties and indemnities cover infringement arising from AI-generated outputs, including whether exclusions or limitations undermine the protection expected.

Liability: is the cap still fit for purpose?

The financial consequences of an AI-related failure may extend well beyond the immediate cost of putting things right. These could include data breaches, IP infringement claims, regulatory exposure or losses caused by inaccurate AI-generated information.

A standard liability cap may not reflect the scale or nature of these risks. Businesses should consider whether the overall cap is appropriate, whether specific risks warrant separate or higher caps, and whether any exclusions from liability could leave them exposed. Any revised allocation should reflect the services, the potential impact of failure and the parties’ respective ability to manage the risk.

What should businesses do?

Businesses do not necessarily need to renegotiate every contract simply because AI is involved. However, they should identify where AI is being used by their own teams and suppliers, assess the sensitivity of the information involved, and review relevant contractual protections.

Where appropriate, contracts should expressly regulate AI use, establish clear responsibilities and address ownership, infringement and liability. Businesses should also ensure that their internal AI policies align with their contractual commitments.

AI may be changing how business is done, but the fundamental principle of commercial contracting remains the same: risk should be identified, understood and allocated deliberately. Businesses should not assume that contractual protections negotiated before AI become commonplace will adequately address the risks its use now presents.

If you have any queries about any of the contents of this article, or if you would like to speak to a member of our team, you can contact us here.

Disclaimer: General Information Provided Only

Please note that the contents of this article are intended solely for general information purposes and should not be considered as legal advice. We cannot be held responsible for any loss resulting from actions or inactions taken based on this article.