AI's relentless advance is redefining innovation, creativity and economic value in the world we live in. Algorithms are evolving in sophistication every day and generative AI models are raising new challenging questions in the field of intellectual property (IP) rights and their enforcement. AI is both a source of new creative tools and a disruptive force challenging long-standing business models and requiring that we develop new legal standards and norms.

As a result, its impact is weighing on lawmakers, creators and industry stakeholders.

Overview

AI is a broad catch-all for a spectrum of tech enabling machines that simulate aspects of human intelligence, including reasoning, learning, perception, and decision-making.

The relationship between AI and IP is complex and rapidly evolving. On the one hand, AI tools are a driver for efficiency, accelerating and enhancing processes, leading to an explosion in the volume and diversity of outputs. But on the other, AI-generated works challenge the very definitions on which IP law is predicated—posing numerous questions around inventorship, authorship, ownership and the enforceability of rights.

A snapshot is unpacked below.

AI as a creator and the dilemma of authorship and ownership

Among the most intriguing impacts of AI on IP is its ability to autonomously generate images, text, code, and other content, but also to propose solutions to challenging technical and scientific problems. The legal frameworks underpinning IP rights—copyright, patents, trademarks, and designs—are historically built on human authorship and inventorship. But the introduction of non-human outputs requires a re-evaluation of core legal concepts.

  • Copyright: Traditionally, it is granted to the author of an original work. But who is the author when a machine generates an image, composes a melody, or writes a story? Opinion by jurisdictions differ. UK legislation suggests that copyright in computer-generated works could be granted to the person who made the "arrangements necessary" for creation. Elsewhere, like in the US, the consensus is generally to not recognize copyright in works created by non-human agents.
  • Patents: Patent law hinges on a requirement for a named "inventor" who contributed to the conception of an invention. But the rapid emergence of AI-generated inventions—such as new molecule designs, software coding or algorithms—has led to contentious legal battles. 

    Most notably, the so-called DABUS case has been litigated and / or administratively reviewed in many jurisdictions including the UK, US, Australia and Canada, which tested whether an AI system could be listed as an inventor. Courts and Patent Offices have so far insisted on human inventorship, but the question lingers since AI is increasingly contributing to scientific advancement. 
  • Trademarks and designs: AI can autonomously develop logos, product designs, and branding elements.  

AI as a driver for creative capacity

AI's ability to analyse vast amounts of data, learn patterns, and generate outputs is supporting the creation of new technical and research tools, and enables inventors to rapidly deliver and design complex products. 

Infringement and detection

From an enforcement standpoint, AI is a powerful tool for detecting infringers and managing IP portfolios. Machine learning systems can monitor vast digital landscapes, spot unauthorised use of copyrighted works, patents, and trademarks. They can also identify counterfeits, track the misuse of brand names, and even predict potential infringing behaviour. In fact, it's an area Gowling WLG has already pioneered thanks to its global Saturn brand protection platform.

  • Copyright enforcement: AI-powered content recognition systems are widely used to scan websites, social media, and e-commerce platforms for pirated material. 
  • Patent and trademark monitoring: Algorithms are used to search patent databases and track new filings, helping identify potential overlap or infringement. They also scan online marketplaces for counterfeit products, enabling more effective brand protection.
  • Automation in litigation: AI can assist legal practitioners by predicting case outcomes, summarising legal documents and managing evidence. It streamlines enforcement actions and reduces costs, although concerns have been voiced that it may also introduce new biases and complexities.

Challenges posed by AI to IP enforcement

Despite the benefits, AI introduces significant challenges to IP enforcement:

  • Mass generation and anonymity: AI can produce vast numbers of works and inventions with minimal human oversight, complicating attribution and ownership. The relative ease of anonymous or pseudonymous production makes tracking infringers far harder.
  • Training data and fair use: Many AI systems are trained on copyrighted works without explicit permission. It's an ongoing debate that raises questions about infringement, fair use exceptions and the moral rights of original creators. Several lawsuits against AI companies concern the unlicensed use of images, text, and music in training data, are currently in-flight.
  • Cross-border enforcement: AI-generated content is global by default, spreading rapidly across jurisdictions. Differences in national IP laws, especially relating to non-human authorship, hinder harmonised enforcement and create uncertainty for creators and investors. Additionally, AI may be trained in one jurisdiction, yet used in another, complicating the determination.
  • Evolving nature of infringement: AI can mimic creative styles, generate deepfakes, and produce derivative works that blur the line between inspiration and infringement. It again poses critical questions about the threshold of originality and scope of derivative rights.

Opportunities for legal reform

Challenges outlined have already prompted calls for reform in IP law and practice. Policymakers and industry leaders are considering a range of options.

  • Clarification of authorship: There are calls from some quarters for laws to be amended to clarify the status of AI-generated works.
  • Regulation of training data: There are also calls for clear guidance on the use of copyrighted works for AI training. It may include licensing schemes, opt-out mechanisms, or new fair use exceptions tailored for generative AI.
  • Global harmonisation: Efforts to align national IP laws on AI-generated works could help reduce uncertainty and facilitate cross-border enforcement.
  • Enhanced AI tools for enforcement: Investment in AI-powered enforcement systems offer the potential for rights holders to better protect and monetise their IP assets in a digital-first world.
  • Ethical and societal considerations: IP law must balance the interests of creators, consumers, and society. As AI becomes more autonomous, questions of moral rights, attribution, and the social value of creative works become more pressing.

Future direction and notable development

The evolving cross-over between AI and IP is only going to increase. But it's worth keeping an eye on emerging precedent and legislation in the meantime.

  • Subject matter eligibility: As AI becomes increasingly integral to modern innovation, the question of subject matter eligibility for AI-related inventions is rapidly evolving. Notable developments include expanded discussions of whether new patent frameworks, legislation and guidelines are needed to better address inventions generated or facilitated by AI, for example in relation to how claims involving AI should be construed.

    Looking ahead, the future direction is expected to focus on balancing the recognition of genuine human creativity and inventive input with the realities of AI-assisted and AI-generated inventions. This may involve new guidance from patent offices and increased international harmonization to ensure clarity for innovators worldwide.

  • AI-generated art and copyright registration: The US Copyright Office has stated that works created entirely by AI do not qualify for copyright registration and is keen to continue emphasizing the need for human authorship.
  • Collaboration will be critical from here on: Lawmakers, courts, and stakeholders must collaborate to ensure that IP law remains fit for purpose in the era of AI. It is likely to require updated definitions, processes, and enforcement tools that can recognise the unique capabilities of AI, while also safeguarding the rights and interests of human creators.
  • Reform of IP law in response to AI: It's not simply a technical exercise—it is going to be a profound societal challenge, inviting us to reconsider the meaning of creativity, ownership, and value in a world where machines are partners in invention.

The road ahead promises to be interesting, delicate and dynamic. But with thoughtful policy, robust enforcement, and ethical stewardship, AI and IP can co-exist to foster a vibrant, innovative and fair creative economy.