Are You Using Generative AI in Your Business? A Guide on Intellectual Property

Generative AI is a type of artificial intelligence that creates new content by learning patterns and structures from existing data. Unlike traditional AI systems designed for specific tasks, generative AI produces outputs such as text, images, music, and even code. It works by processing large datasets and generating content based on what it has learned, using techniques like machine learning and neural networks.

This technology powers tools that businesses and individuals use every day, offering opportunities for creativity, automation, and innovation. However, as mentioned in a previous blog it also raises important questions about intellectual property and data security.

Generative AI is has brought about lots of positives, in many cases it is revolutionising how businesses operate, offering new ways to innovate, create content, and deliver services. But while the opportunities are immense, so are the risks – particularly when it comes to intellectual property (IP) and confidentiality.

There are many specific AI tools being used in business today, but over the last few years there has been a real increase in generative AI tools being used in a non-industry specific way.  Here are a few tools regularly used by many businesses today:

1. ChatGPT (OpenAI)

• Purpose: Generates human-like text, making it ideal for customer support, content creation, and brainstorming ideas.
• How It’s Used: Writing blogs, drafting emails, answering queries, and even coding assistance.

2. DALL·E (OpenAI)

• Purpose: Creates images from textual descriptions, allowing for customised visual content.
• How It’s Used: Producing artwork, visual marketing materials, or mock-ups for design projects.

3. Jasper AI

• Purpose: A writing tool focused on marketing and business content, including social media posts, advertisements, and blogs.
• How It’s Used: Streamlining content creation with tailored, professional outputs.

4. MidJourney

• Purpose: Generates high-quality art and images based on user-provided prompts.
• How It’s Used: Branding, conceptual art, and design projects.

5. Adobe Firefly

• Purpose: Enhances creative workflows by generating text-to-image visuals and transforming existing content.
• How It’s Used: Graphic design, video production, and marketing collateral.

6. Canva Magic Write

• Purpose: Integrates generative AI into Canva’s design platform for creating text content directly in your projects.
• How It’s Used: Writing captions, designing presentations, and crafting marketing campaigns.

7. Codex (OpenAI)

• Purpose: Specialises in code generation and assists developers with writing, debugging, and understanding code.
• How It’s Used: Automating coding tasks and accelerating software development.

Whether you’re using generative AI for marketing, product development, or customer service, understanding these risks is essential to protect your business and avoid costly legal disputes. Here’s what you need to know.

The Intellectual Property Risks of Generative AI

Infringement of Third-Party IP Rights

Generative AI doesn’t create content from scratch. Instead, it produces outputs based on the data it’s trained on – data often sourced from the internet or other copyrighted materials.

Under UK law, copyright automatically applies to original works, such as text, images, and music. If the AI’s output incorporates or reproduces copyrighted material, your business could inadvertently infringe on third-party IP rights.

Case Example: Infopaq International A/S v Danske Dagblades Forening (C-5/08) established that even small parts of a copyrighted work could be protected if they express the author’s intellectual creation. This principle highlights how even minor elements in AI outputs could lead to infringement claims.

What You Can Do:

• Vet your AI-generated outputs carefully.
• Use tools trained on datasets with clear IP permissions.
• Consult legal experts when using AI-generated content in public-facing or commercial projects.

Who Owns the IP in AI Outputs?

Under current UK copyright law, a work must be created by a human to attract copyright protection. Outputs from generative AI do not automatically qualify as original works, leaving them in a legal grey area.

While human input into the AI process might create grounds for copyright, the extent of this protection is unclear and case law in this area is still being developed.  If you provide AI-generated content to clients, be cautious about assigning IP rights—you may not legally own them to assign.

Case Insight: In the Copyright, Designs and Patents Act 1988 (CDPA), Section 9 specifies that authorship requires a human creator. The lack of provision for non-human creators in UK law underscores the complexity of ownership in AI-generated works.

What You Can Do:

• Clearly define ownership of AI-generated content in contracts with clients.
• Avoid making guarantees about IP rights unless you’re confident in your position.
• Stay informed on legal developments, as this area of law is rapidly evolving.

The Confidentiality Risk: Protecting Client Data

AI-generated models often retain information from their inputs to improve their learning. If you’re using client data to generate outputs, this raises significant confidentiality concerns.

Example: If you input commercially sensitive or personal data into a generated AI tool, that information might be retained by the AI provider and used in future outputs for others to use. This could breach data protection laws, such as the UK’s Data Protection Act 2018 and the GDPR.

What You Can Do:

• Review the terms of service of the AI provider. Ensure the data you input will not be retained or shared.
• Obtain explicit consent from clients before using their data with AI tools.
• Use AI tools with robust privacy safeguards.

Best Practices for Businesses Using Generative AI

1. Audit Your AI Use: Identify where AI is being used in your business and evaluate the associated risks.
2. Work with Legal Experts: Ensure your use of generated AI complies with IP and data protection laws.
3. Define Roles and Responsibilities: Make sure your contracts and agreements clearly outline who owns the IP in AI outputs.
4. Train Your Team: Educate employees on the risks of using AI, particularly around IP and confidentiality.

Looking Ahead: A Changing Legal Landscape

The UK Intellectual Property Office (IPO) and courts are actively considering how to address the challenges posed by AI.  Updates via changes in statute and case law are going to follow, but these things take time. While these solutions are in development, businesses using generated AI must navigate a complex and uncertain legal framework.

By taking proactive steps now, you can protect your business and position yourself to adapt as the law evolves.

 

---

Q&A on Legal Updates to AI and IP

How does UK law currently define authorship and ownership of AI-generated works?

Under UK law, the authorship and ownership of AI-generated works are primarily governed by the Copyright, Designs and Patents Act 1988 (CDPA). Section 9(3) of the CDPA addresses computer-generated works, stating:

“In the case of a literary, dramatic, musical or artistic work which is computer-generated, the author shall be taken to be the person by whom the arrangements necessary for the creation of the work are undertaken.”

This provision implies that when a work is generated by a computer without a human author, the individual who made the necessary arrangements for its creation is considered the author and, consequently, the first owner of the copyright. This approach is designed to ensure that such works receive copyright protection, even in the absence of direct human authorship.

It’s important to note that the legal landscape concerning AI-generated works is evolving, and future cases may provide more specific guidance on this matter.

Are there proposals to update the Copyright, Designs and Patents Act 1988 to address AI?

Yes, there have been several proposals to update the Copyright, Designs and Patents Act 1988 (CDPA) to address the challenges posed by artificial intelligence (AI). The UK government has recognised the need to clarify the relationship between intellectual property and AI, particularly concerning copyright.

Government Initiatives:

• Code of Practice on Copyright and AI: In June 2023, the UK government announced the development of a code of practice aimed at making licenses for data mining more accessible. This initiative seeks to overcome barriers faced by AI firms and users while ensuring protections for rights holders. The government expects parties to adopt this code voluntarily but has indicated that legislation could be considered if an agreement isn’t reached. GOV.UK
• Consultations on AI and IP: The Intellectual Property Office (IPO) has conducted consultations to gather views on how generated AI intersects with copyright and patents. These consultations explore issues such as the use of copyrighted works by AI systems, the authorship of AI-generated works, and potential reforms to existing laws. GOV.UK

Parliamentary Discussions:

• Calls for Legislative Updates: In May 2024, Baroness Stowell of Beeston, chair of the Communications and Digital Committee in the House of Lords, urged the government to update copyright law to clarify the rights of copyright holders and AI developers. She emphasised the need to address uncertainties in the current legal framework to prevent problematic business models from becoming entrenched.

Recent Developments:

• Abandonment of Voluntary Code Initiative: In February 2024, the government abandoned plans to broker an industry-led agreement on a new AI copyright code of practice. Despite this, the government continues to engage with stakeholders and has indicated that alternative non-statutory solutions may emerge, focusing on greater transparency over the data used to train AI models.

These initiatives reflect the UK’s proactive approach to adapting its copyright laws in response to the evolving landscape of generated AI and its implications for intellectual property.

In cases of AI-generated outputs, what factors determine whether human input qualifies for copyright protection?

In the context of AI-generated outputs, the determination of whether human input qualifies for copyright protection hinges on the extent and nature of human involvement in the creation process. Key factors include:

1. Degree of Human Creativity and Control: The human contributor must exercise creative choices that significantly influence the final work. Mere operation of an AI tool without meaningful creative input is generally insufficient for copyright protection.
2. Originality: The human input must result in an original work, reflecting the author’s personality and free creative choices. This aligns with the European Court of Justice’s (ECJ) standard for originality, which requires a work to be the author’s own intellectual creation.

Case Law and Commentary:

• European Court of Justice (ECJ): The ECJ has established that for a work to be protected by copyright, it must be the result of the author’s own intellectual creation, reflecting their personality and free creative choices. This standard emphasises the necessity of human creativity and decision-making in the creation process. Cambridge University Press
• UK Perspective: In the UK, Section 9(3) of the Copyright, Designs and Patents Act 1988 (CDPA) addresses computer-generated works, stating that the author is “the person by whom the arrangements necessary for the creation of the work are undertaken.” This provision suggests that the individual who makes the necessary arrangements for the creation of a computer-generated work is considered the author, provided there is sufficient human input. In cases where no human author exists for example in computer-generated works, the person overseeing the creation process is recognised as the author because they are deemed to be closest to the creation of the work being undertaken.

This then gives rise to the question, whether literary, dramatic, musical or artistic works created by AI can meet the “ author’s intellectual creation” originality test, and thereby, whether AI can be classed as an ‘author’ in the first instance. Without any legislative intervention, it is likely to be difficult to argue that a work created by AI could be ‘original’ under this test.

• Commentary: Legal scholars have noted that the degree of human intervention required for a work to qualify for copyright protection must involve more than merely selecting a single work from a number of works created by an AI. There must be a significant level of human creativity and decision-making involved in the creation process.

In summary, for AI-generated outputs to qualify for copyright protection, there must be substantial human creative input that results in an original work, reflecting the author’s personality and free creative choices. The mere use of AI tools without meaningful human involvement is generally insufficient to meet the threshold for copyright protection.

Infringement and Liability

If AI generates content that infringes third-party IP, who is legally responsible – the user, the AI provider, or both?

What safeguards should businesses implement to avoid unknowingly infringing IP when using AI-generated content?

To avoid unknowingly infringing on intellectual property (IP) rights when using AI-generated content, businesses should implement a combination of technical, legal, and organisational safeguards. These measures reduce the risk of liability and ensure compliance with IP laws.

1. Due Diligence on AI Tools and Providers

• Understand Licensing Terms: Review the terms of service and licensing agreements of AI tools. Ensure they specify how generated content can be used and clarify liability for infringement.
• Evaluate Training Data: Use AI systems with transparent training data sources. Avoid tools trained on datasets with copyrighted material if usage rights are unclear.
• Reputable Providers: Choose AI providers with a strong track record of legal compliance and clear IP policies.

2. Contractual Protections

• Indemnity Clauses: Negotiate contracts with AI providers that include indemnity provisions, holding the provider accountable for legal claims arising from copyright or trademark infringement.
• Warranties on Non-Infringement: Require assurances from providers that the AI-generated outputs are free from third-party IP infringement.

3. Content Review and Vetting

• Human Oversight: Implement review processes where legal or creative teams vet AI-generated content before use. This helps identify potential IP conflicts.
• Content Moderation Tools: Use additional software to cross-check AI outputs against copyrighted or trademarked material.

4. Implement Clear Use Policies

• Define Acceptable Use: Establish internal policies for using AI tools, including guidelines on permissible inputs and outputs.
• Restrict Inputs: Avoid feeding copyrighted material into AI systems unless the necessary rights or permissions are secured.

5. Invest in IP Training and Awareness

• Employee Training: Educate employees on IP laws, emphasizing the risks and responsibilities of using AI-generated content.
• Legal Team Involvement: Involve legal counsel early in the development or use of AI-generated materials.

6. Monitor for Infringement

• Trade Mark and Copyright Checks: Use reverse image searches, plagiarism detection tools, and copyright monitoring platforms to ensure outputs do not unintentionally replicate existing works.
• Audits: Conduct regular audits of AI-generated content to ensure ongoing compliance.

7. Secure Usage Rights

• Obtain Licenses: Where appropriate, obtain licenses for the datasets or models used, especially when outputs may overlap with existing copyrighted materials.
• Public Domain and Open Licenses: Prefer models trained on public domain or openly licensed data, ensuring compliance with their terms.

8. Monitor Legal Developments

• Stay Updated: Monitor changes in IP law related to generated AI to anticipate new compliance requirements.
• Participate in Industry Discussions: Engage with industry bodies to align practices with evolving standards.

Example Practices

• Case Study (Getty Images v. Stability AI): This case highlighted the risks of using copyrighted material in AI training. Businesses should avoid tools implicated in similar controversies.
• Fair Use Analysis: Conduct fair use evaluations if relying on potentially copyrighted material, especially in jurisdictions with strict IP rules.

By implementing these safeguards, businesses can reduce the risk of IP infringement while leveraging AI-generated content responsibly and effectively.

How does UK IP law on AI compare with laws in other jurisdictions, such as the EU or US?

The treatment of intellectual property (IP) in relation to AI-generated works differs significantly across the UK, EU, and US. These differences reflect each jurisdiction’s approach to copyright, authorship, and IP protection.

1. Authorship of AI-Generated Works

UK

• Copyright, Designs and Patents Act 1988 (CDPA):
  • Section 9(3): In the case of computer-generated works, “the author shall be taken to be the person by whom the arrangements necessary for the creation of the work are undertaken.”
  • This provision allows AI-generated works to be protected under copyright, with authorship attributed to the individual responsible for making the arrangements (e.g., the AI user or developer).
  • Unique Approach: The UK is one of the few jurisdictions to explicitly address computer-generated works in copyright law, assigning authorship even in the absence of human creativity.

EU

• Focus on Human Originality:
  • EU copyright law requires that a work reflects “the author’s own intellectual creation,” as established in the Infopaq International A/S v. Danske Dagblades Forening case.
  • AI-generated works generally do not qualify for protection unless significant human creative input is involved in shaping the output.
  • The EU prioritises the originality of human intellectual effort, excluding purely AI-generated works from copyright.

US

• Strict Human Authorship Requirement:
  • The US Copyright Office and courts maintain that only human-authored works are eligible for copyright.
  • This principle was affirmed in Naruto v. Slater (the “monkey selfie” case) and Thaler v. Perlmutter (denying copyright for AI-generated art with no human involvement).
  • Generated AI content may only be protected if a human’s contributions meet the originality threshold.

2. Copyright Ownership and Licensing

UK

• Copyright is assigned to the person who “undertakes the necessary arrangements” for AI-generated works, typically the AI user or commissioner, unless otherwise contractually agreed.
• AI developers and providers are not automatically entitled to copyright unless explicitly specified.

EU

• Copyright ownership depends on the human author’s creative input.
• For AI-assisted works, copyright belongs to the human contributor who adds originality, while purely AI-generated content may fall outside protection entirely.

US

• Similar to the EU, copyright attaches to the human creator of original elements. Outputs from purely AI systems, without substantial human creativity, are not protected.
• Licensing agreements often determine how AI-generated content can be used, but these do not confer copyright protection for the AI outputs themselves.

3. Liability for IP Infringement by AI

UK

• Liability is shared between the user and the provider, depending on the circumstances:
  • User Liability: If a user intentionally employs AI to create infringing content, they may be liable.
  • Provider Liability: Providers may be held accountable if the AI system predictably generates infringing content or if the training data was improperly sourced.

EU

• The EU leans toward holding users accountable, as they control how AI tools are used.
• Providers may share liability if they fail to implement safeguards or if the AI’s training data infringes IP rights.

US

• Infringement liability focuses on the party that caused the infringement.
  • Users who knowingly generate or distribute infringing content are held liable.
  • Providers may face liability if their AI systems were trained on copyrighted materials without authorisation (e.g., the lawsuits against OpenAI and Stability AI).

4. Policy Developments and Future Trends

UK

• The UK is exploring reforms to address AI and IP issues, including consultations and proposals for clearer guidance on copyright ownership and liability.

EU

• The EU is considering stricter regulations on AI, including transparency obligations for datasets used to train AI models (e.g., the proposed AI Act).

US

• The US is grappling with how to address AI in copyright law. Recent lawsuits against AI companies for training models on copyrighted data without authorisation may set precedents.

Comparison Table

Aspect	UK	EU	US
Authorship	Arranger of AI work is the author	Human originality required	Human originality required
Purely AI-Generated	Copyright possible under CDPA Section 9(3)	No copyright protection	No copyright protection
AI-Assisted Works	Copyright if human adds originality	Copyright if human adds originality	Copyright if human adds originality
Liability	Shared between user and provider	Primarily user, potentially provider	Primarily user, potentially provider
Legal Framework	Explicitly addresses AI works	Human-centric	Human-centric

Conclusion

The UK offers more explicit guidance on authorship for AI-generated works than the EU or US, making it relatively unique in recognising such outputs under copyright. The EU and US remain aligned in requiring significant human originality for protection. However, all jurisdictions are actively revisiting these issues, and future legislation may align their approaches further.

Are there international efforts to harmonise laws around AI-generated content and IP?

Yes, there are ongoing international efforts to harmonise laws concerning generated AI content and intellectual property (IP). These initiatives aim to address the complexities introduced by AI in the realm of IP rights and to establish consistent legal frameworks across jurisdictions.

World Intellectual Property Organisation (WIPO):

WIPO has been at the forefront of facilitating global discussions on the intersection of AI and IP. Through its “WIPO Conversations on IP and Frontier Technologies,” the organisation provides a platform for stakeholders to discuss the impact of emerging technologies, including AI, on IP rights. These conversations aim to bridge information gaps and foster international cooperation in developing harmonised legal standards. ArXiv

European Union (EU):

The EU is actively working on regulations to address AI’s implications for IP. The proposed Artificial Intelligence Act includes provisions that would require AI systems to disclose the use of copyrighted material in their training data. This move seeks to enhance transparency and ensure that AI development respects existing IP rights.

United Kingdom (UK):

The UK government has recognised the need for international collaboration in adapting IP laws to the challenges posed by AI. In its response to consultations on AI and IP, the UK Intellectual Property Office (UKIPO) emphasised the importance of engaging in international discussions to develop harmonised approaches, particularly concerning AI inventorship and authorship.

Industry Initiatives:

Beyond governmental efforts, industry stakeholders are also contributing to harmonization. For instance, the Dataset Providers Alliance (DPA), formed by companies involved in content licensing for AI training data, aims to promote ethical data sourcing and advocate for consistent legal standards across borders.

These collective efforts underscore a global recognition of the need to harmonise IP laws in response to the rapid advancements in AI technology. While challenges remain, such as differing national legal frameworks and the pace of technological change, these initiatives represent significant steps toward establishing coherent and consistent international standards for AI-generated content and IP rights.

Are there any active government initiatives or consultations addressing AI’s role in IP?

Yes, the UK government has actively engaged in initiatives and consultations to address the intersection of artificial intelligence (AI) and intellectual property (IP). These efforts aim to adapt the IP framework to the evolving technological landscape and ensure that AI developments are effectively integrated into existing legal structures.

1. National AI Strategy

In September 2021, the UK government launched the National AI Strategy, outlining a 10-year plan to position the UK as a global leader in AI. This strategy emphasises the importance of a robust IP framework to support AI innovation and includes commitments to review and potentially reform IP laws in light of AI-generated advancements. GOV.UK

2. Consultations on AI and IP

The UK Intellectual Property Office (UKIPO) has conducted several consultations to gather insights on how AI interacts with IP laws:

• Artificial Intelligence and IP: Copyright and Patents Consultation (2021): This consultation sought views on the relationship between AI and copyright, as well as patents. It addressed issues such as authorship of AI-generated works and the patentability of AI-devised inventions. GOV.UK
• Government Response to AI and IP Consultation (2022): Following the 2021 consultation, the government published its response, outlining planned actions and areas for further consideration. The response highlighted the need for ongoing evaluation of IP laws to accommodate AI developments. GOV.UK

3. AI Regulation White Paper

In March 2023, the government released a white paper titled “A pro-innovation approach to AI regulation,” proposing a framework to regulate AI technologies. This document discusses the role of IP in fostering AI innovation and seeks feedback on potential regulatory measures. GOV.UK

4. AI Safety Institute

The UK has established the AI Safety Institute to focus on the safe development and deployment of AI technologies. While its primary focus is on safety, the institute’s work intersects with IP considerations, particularly concerning the use of copyrighted materials in AI training datasets. Financial Times

5. Ongoing Engagement with Stakeholders

The government continues to engage with industry stakeholders, legal experts, and the public to assess the impact of AI on IP. This includes exploring issues related to data mining, licensing, and the protection of AI-generated works. Such engagements are crucial for developing policies that balance innovation with IP rights. IPO Blog

These initiatives reflect the UK’s proactive approach to ensuring that its IP framework evolves in tandem with advancements in AI, fostering an environment conducive to innovation while safeguarding IP rights.

What case law developments should businesses watch for regarding AI and IP disputes?

Businesses should closely monitor several key legal developments concerning artificial intelligence (AI) and intellectual property (IP) disputes, as these cases will significantly influence the legal landscape and inform best practices.

1. Getty Images v. Stability AI

• Overview: Getty Images has filed lawsuits against Stability AI in both the UK and the US, alleging unauthorised use of millions of its images to train the AI model, Stable Diffusion.
• Implications: These cases will address critical issues such as the legality of using copyrighted materials for AI training without explicit permission and the potential infringement resulting from AI-generated outputs.

2. Artists’ Lawsuit Against Stability AI and Midjourney

• Overview: In January 2023, artists Sarah Andersen, Kelly McKernan, and Karla Ortiz filed a class-action lawsuit against Stability AI, Midjourney, and DeviantArt, claiming that these companies infringed on their copyrights by using their artworks to train AI models without consent.
• Implications: This lawsuit will explore the boundaries of fair use in AI training and the rights of artists over their works in the context of generated AI art.

3. The New York Times v. OpenAI

• Overview: In December 2023, The New York Times filed a lawsuit against OpenAI, alleging that the company used its articles without authorisation to train AI models, resulting in copyright infringement. This week however, there has been an update in the case: OpenAI engineers accidentally deleted data potentially relevant to the case.

OpenAI agreed to provide two virtual machines so that counsel for The Times and Daily News could perform searches for their copyrighted content in its AI training sets. (Virtual machines are software-based computers that exist within another computer’s operating system, often used for the purposes of testing, backing up data, and running apps.) Lawyers for the publishers say that they and experts they hired have spent over 150 hours since November 1 searching OpenAI’s training data.

However, on November 14, OpenAI engineers erased all the publishers’ search data stored on one of the virtual machines.  OpenAI tried to recover the data and was mostly successful. However, because the folder structure and file names were “irretrievably” lost, the recovered data “cannot be used to determine where the news plaintiffs’ copied articles were used to build OpenAI’s models.

• Implications: The outcome of this case could set precedents regarding the use of news content in AI training and the application of fair use doctrines in such contexts.

4. Music Industry Lawsuits Against AI Companies

• Overview: In June 2024, the Recording Industry Association of America (RIAA), along with major music labels, filed lawsuits against AI developers Suno and Udio. The suits allege that these companies used copyrighted music to train AI systems, leading to the generation of songs that mimic the work of well-known artists.
• Implications: These cases will examine the legality of using copyrighted music in AI training and the potential infringement arising from AI-generated music.

5. DABUS AI Inventorship Cases

• Overview: The DABUS cases involve attempts to patent inventions created by an AI system named DABUS, with the AI itself listed as the inventor. Courts in multiple jurisdictions, including the UK, US, and EU, have ruled that only natural persons can be recognised as inventors under current patent laws.
• Implications: These decisions highlight the challenges of attributing inventorship to AI systems and may prompt legislative reviews to address AI’s role in innovation.

6. OpenAI’s Legal Victory Over Progressive Publishers

• Overview: In the US OpenAI secured a legal win this month against Alternet and Raw Story when a judge dismissed the publishers’ copyright case. The publishers had argued that OpenAI violated the Digital Millennium Copyright Act (DMCA) by scraping news articles and omitting copyright management information.
• Implications: This ruling could influence future IP cases, potentially affecting publishers’ ability to claim damages for model training without compensation.

7. Microsoft CEO’s Call for Copyright Law Revisions

• Overview: Microsoft CEO Satya Nadella has advocated for a revision of copyright laws to enable tech companies to train AI models without risking IP infringement. He highlighted Japan’s flexible approach and urged governments to establish legal frameworks defining “fair use” to support innovation. The Times
• Implications: This call reflects a broader industry push to balance creative rights with technological progress, emphasising the need for updated laws to prevent stifling innovation.

8. Legal Counsel Navigating AI Risks

• Overview: Legal professionals are increasingly leveraging AI technology for various tasks, raising ethical and accuracy concerns. Experts emphasize the necessity of governance frameworks to manage AI-related risks, including data privacy and output reliability.
• Implications: This highlights the importance of involving AI experts and thoroughly evaluating AI vendors to mitigate risks, as well as staying informed about ongoing and upcoming AI-related litigation and legislation.

Monitoring these developments will help businesses navigate the evolving legal landscape surrounding AI and IP, enabling them to implement strategies that mitigate risks and ensure compliance.

How might AI’s role in creating counterfeit goods (e.g. using AI to replicate designs or branding) shape future IP enforcement strategies?

AI’s increasing role in creating counterfeit goods, such as replicating designs, branding, or product models, poses unique challenges to intellectual property (IP) enforcement. This evolving landscape is likely to shape future enforcement strategies in several significant ways:

1. Enhanced Detection Technologies

• AI for Enforcement: Enforcement agencies and rights holders will increasingly use AI to detect counterfeit goods by analysing patterns, designs, or branding across digital and physical markets.
  • Example: AI tools can scan e-commerce platforms for listings that infringe trade marks or copyrights by comparing images, descriptions, and keywords against authentic product databases.
• Blockchain Integration: Combining AI with blockchain could provide secure authentication and tracking of genuine goods, making it easier to detect counterfeit items.

2. Stricter Digital Platform Regulations

• Platform Accountability: Governments may introduce laws holding digital platforms (e.g., e-commerce sites, social media platforms) accountable for hosting AI-generated counterfeit goods.
  • Example: The EU’s Digital Services Act already requires platforms to take proactive measures against illegal content, including counterfeit goods.
• Content Moderation Technologies: Platforms will likely be required to implement more sophisticated AI moderation tools to identify and remove listings of counterfeit products created using generative AI.

3. Legal and Legislative Reforms

• Expanded IP Definitions: Laws may be updated to include provisions specifically targeting AI-generated counterfeits.
  • Example: New legislation could classify the act of training AI models on copyrighted or trade marked designs without authorisation as a form of infringement.
• Criminal Liability for AI Use: Laws could establish liability for entities or individuals using AI to intentionally replicate and sell counterfeit goods.

4. Dynamic IP Monitoring

• Continuous Scanning: Businesses will invest in AI-powered systems to continuously monitor for counterfeit goods on global platforms.
  • Example: AI tools like image recognition software can detect unauthorized use of a brand’s logo or product design in real-time.
• Automated Takedown Mechanisms: IP owners may automate the submission of takedown requests using AI, reducing the time and cost of enforcement.

5. Education and Awareness Campaigns

• Consumer Awareness: Governments and businesses may launch awareness campaigns to educate consumers about the risks of AI-generated counterfeits, including safety concerns and economic impacts.
• Vendor Training: Training for manufacturers and retailers will emphasise identifying and avoiding counterfeit supply chains, especially those enabled by AI.

6. Collaboration Between Stakeholders

• Public-Private Partnerships: IP enforcement agencies, tech companies, and rights holders will collaborate to create AI-driven solutions to combat counterfeiting.
  • Example: Joint AI databases for sharing counterfeit detection algorithms or datasets.
• International Cooperation: Counterfeiting is a cross-border issue, and AI-powered counterfeiting will drive more collaboration between jurisdictions for harmonized enforcement efforts.

7. Preventive Design Measures

• AI-Resistant Designs: Companies may create products or branding elements resistant to replication by AI, such as incorporating unique, difficult-to-replicate features.
  • Example: Holograms, microprinting, or dynamic patterns embedded in product packaging or design.

8. Ethical AI Standards

• Regulating AI Development: Policymakers may require AI developers to embed safeguards in their systems to prevent unauthorised use for counterfeiting.
  • Example: Licensing agreements or terms of use that explicitly prohibit training AI models on copyrighted or trademarked content without permission.

9. Increased Focus on Supply Chain Integrity

• AI-Powered Authentication: Manufacturers may embed AI-readable markers, like QR codes or NFC chips, in products to ensure their authenticity throughout the supply chain.
• Enhanced Customs Enforcement: Customs agencies will use AI to better identify counterfeit shipments at borders, analysing cargo manifests and images for red flags.

10. Evolution of Litigation Strategies

• Targeting AI Providers: Rights holders may increasingly hold AI developers liable if their models or tools are used to create counterfeits, especially if they failed to implement safeguards.
• Challenging AI Outputs: Businesses may need to establish legal precedent for classifying AI-generated replicas as counterfeit under IP law.

AI’s role in creating counterfeit goods will drive a shift in IP enforcement strategies toward technology-enabled solutions, stronger regulatory frameworks, and closer collaboration between stakeholders. Rights holders and enforcement agencies will need to adapt quickly to address the scale, sophistication, and cross-border nature of AI-generated counterfeiting.

Need Support with Intellectual Property?

At National Business Register, we specialise in helping businesses navigate the complexities of intellectual property. Whether you need advice on protecting your own IP or understanding ownership options for you, our team is here to help.

📞0800 069 9090
📧info@nbrg.co.uk

Don’t leave your IP to chance – let’s protect your business together.