Generative artificial intelligence (AI) is prompting a re-evaluation of traditional approaches to intellectual property (IP), particularly in the context of the IP contract. As businesses in creative, media, and technology sectors integrate AI tools into their workflows, legal professionals must adapt IP contracts to address emerging risks and regulatory uncertainty. For IP specialists like Briffa Legal, this evolution presents both challenges and opportunities.
What Are the Key IP Risks Arising from AI Use?
The adoption of generative AI introduces several risks that directly affect the IP contract. Key concerns include:
- Training data: Many AI systems are trained on datasets that may include copyrighted works without permission. This raises questions about whether such data use constitutes infringement.
- Outputs and derivative works: Generated content can sometimes replicate elements of original works, potentially infringing copyright and impacting IP rights.
- Attribution and moral rights: When AI assists or fully generates creative output, questions arise about who holds authorship, whether creators must give attribution, and how moral rights apply.
- Ownership and licensing: Current UK law has not settled who owns rights to AI-generated content, so the IP contract must address it.
- Contractual liability: When outputs infringe third-party rights, identifying which party is responsible becomes critical.
These issues necessitate more nuanced and comprehensive IP contracts to mitigate risk.
What Does UK Law Currently Say About IP Contracts and AI?
Under Section 9(3) of the UK’s Copyright, Designs and Patents Act 1988, when no one can identify a human author, the law assigns authorship to the person who arranged the creation of the computer-generated work. Many experts now view this provision as outdated, especially in light of modern AI capabilities.
In 2025, the UK government launched an expert working group to explore copyright reform. A notable proposal is an “opt-out” model, which would allow AI developers to use copyrighted materials for training unless rightsholders explicitly object.
These developments affect how IP contracts are drafted and enforced, particularly in relation to data use, licensing, and ownership.
What Should an AI-Resilient IP Contract Include?
To address these new risks, IP contracts should be updated to include:
- Definitions: Clearly define “AI-generated content”, “training data”, and “derivative works” to avoid ambiguity.
- Scope of licence: Specify whether AI-generated or AI-assisted works are included in the licensing arrangement.
- Warranties and representations: Parties should warrant that they have the right to use training data and that outputs do not infringe third-party IP rights.
- Indemnity clauses: Allocate responsibility for legal costs and damages in the event of an infringement claim.
- Ownership clauses: Establish ownership of AI-generated outputs clearly in the IP contract.
- Audit and transparency provisions: Allow for oversight of data sources and model behaviour.
- Compliance with evolving law: Include terms to ensure compliance with future legal and regulatory developments.
These contract features are essential to ensure legal clarity and reduce exposure to future disputes.
Why the IP Contract Matters for AI-Adopting Businesses
For businesses incorporating AI, a well-drafted IP contract is essential to manage exposure. Without clear terms, organisations may face liability for infringements linked to training data or AI outputs.
Creative professionals and rightsholders also rely on strong IP contracts to control the use of their works in AI development. Some UK licensing bodies have proposed collective licensing models, though these remain in development.
Until a standardised solution is available, IP contracts remain the primary tool for asserting and enforcing rights in AI contexts.
FAQs
What is an IP contract in the context of AI?
An IP contract outlines how intellectual property rights are created, assigned, licensed, and enforced. In AI contexts, it must also address training data, output ownership, and liability.
How does AI affect IP contract drafting?
It introduces complexities related to data usage, authorship, output rights, and legal risk, all of which must be reflected in updated contract terms.
What clauses are essential in an AI-related IP contract?
Key clauses include ownership, warranties, indemnities, licence scope, audit rights, and compliance provisions.
Is UK law changing to address AI and IP?
Yes. Ongoing consultations are examining whether copyright law and contract frameworks need reform to accommodate AI developments.
How can law firms assist with AI IP contracts?
By providing audits, drafting updated templates, offering guidance on regulatory trends, and helping clients understand their legal exposure.
The Role of IP Law Firms in Drafting AI-Ready IP Contracts
Firms like Briffa Legal are well-positioned to advise clients on AI-related updates to the IP contract. As case law and regulation evolve, there is growing demand for:
- IP contract audits
- AI-specific licensing agreements
- Risk assessments focused on copyright and AI
- Template clauses for IP contracts
Keeping pace with these developments will allow IP law firms to support clients effectively in an increasingly complex legal environment.
