AI and Intellectual Property

In the rapidly evolving domain of artificial intelligence (AI), the intersection with intellectual property (IP) law presents a complex tapestry of legal, ethical, and commercial considerations. As AI systems, particularly generative AI, become more sophisticated, they raise unprecedented questions about the use and protection of intellectual creations. The use of content by creators for training AI systems, including the scraping of data from the internet, has ignited a debate on the extent to which such practices infringe upon copyright and other IP rights. This is particularly pertinent when AI is trained on datasets that may contain copyrighted material without the express permission of the rights holders. The legal framework in the Netherlands, as well as the broader European context, including the proposed AI Act, is grappling with these issues to ensure that innovation is not stifled while protecting the rights of creators.

Legal Issues with AI-Generated Content

• Use of Creator Content for AI Training: The legality of using creator content for AI training, especially when it involves generative AI, hinges on whether such use constitutes copyright infringement. The practice of scraping data from the internet for this purpose can potentially violate copyright laws if the data includes protected works and is used without authorization. This infringement risk is heightened with generative AI, such as text and image generators, which can produce new works that closely resemble the source material.
• IP Protection for AI-Generated Content: Determining whether content or materials produced by AI can be protected under IP laws is a contentious issue. Currently, there is no consensus on whether AI-generated works, such as texts or images, can be attributed with copyright since they lack a human author. However, the evolving legal landscape may redefine the parameters of protection for such works. We see that the legal framework is adapting to these technological advancements, with considerations varying by jurisdiction.
• Protection of AI Systems, Tools, and Software: AI systems, tools, and software may be protected under various IP regimes, including patents for novel inventions or copyright for original software code. However, the protection is subject to the fulfillment of traditional IP law requirements.

Ownership and Legal Risks

• Ownership of AI-Produced Materials: The question of who owns the results from content or materials produced by AI or an AI-assisted system is still under debate. Ownership can be influenced by the input data, the AI's design, and the purpose for which the AI is used. The proprietary rights may rest with the AI developer, the user, the one inputting or selecting the data, or potentially the AI itself (or no ownership might be available at all), depending on the jurisdiction and the specific circumstances. Trade secret protection is another critical aspect of IP law in the context of AI. Protecting the data used to train AI systems can be essential for maintaining a competitive edge.
• Legal Risks: The legal risks associated with AI and IP include potential infringement claims, challenges in enforcing rights against AI-generated content, and uncertainties in ownership which can lead to complex litigation. To mitigate these risks, parties must be proactive in understanding the legal landscape and implementing robust IP strategies. Contractual agreements and (online) disclaimers are a powerful tool for parties to prepare for and implement rules governing AI and IP. These agreements can define the scope of use, ownership rights, and responsibilities of each party, providing clarity and legal enforceability.

Preparation and Implementation in the business

The intersection of artificial intelligence (AI) and intellectual property (IP) rights presents both unique opportunities and challenges for companies. It is crucial to approach this matter with a comprehensive strategy that encompasses legal, technical, and business considerations. The complexity of AI's relationship with IP rights is significant, and while a comprehensive strategy can be outlined, the nuances of specific cases may be beyond a general advisory capacity. It is essential for companies to remain agile and informed, as the legal landscape surrounding AI and IP is continually evolving.

Some of the points that are relevant in implementing a sound IP/AI strategy are the following:

• Initial Assessment and Strategy Development
  • Risk Identification: The first step is to conduct a thorough risk assessment to identify potential IP issues that may arise from the use of AI. This includes understanding how AI might infringe on existing IP rights, create new IP, or expose the company to IP theft or misuse.
  • Opportunity Analysis: Concurrently, assess the opportunities AI presents for creating new IP and enhancing the value of existing IP assets. AI can be a source of innovation, leading to new products, services, or processes that can be protected by IP rights.
• Legal Framework and Compliance
  • IP Audit: Perform an IP audit to catalog all existing IP assets and evaluate their relevance in the context of AI deployment. This will help in understanding the scope of the company's IP portfolio and its potential expansion with AI-related innovations.
  • Regulatory Compliance: Ensure compliance with the relevant laws and regulations governing AI and IP in the Netherlands, such as the AI Act.
  • Contractual Agreements: Review and update contractual agreements, including licenses, joint development agreements, and non-disclosure agreements, to address AI-specific considerations and IP ownership issues.
• Protection of AI-Generated IP
  • Patent Strategy: Develop a patent strategy for AI-generated inventions, considering the patentability criteria and the nuances of AI as an inventor. This may involve seeking legal clarity on whether AI-generated works can be patented and who owns the rights to such patents.
  • Copyright Considerations: Evaluate the copyrightability of AI-generated works, including software, databases, and potentially creative works, and determine the appropriate form of protection.
  • Trade Secret Management: For AI innovations that are not patentable or publicly disclosed, consider protecting them as trade secrets. Implement robust confidentiality measures to safeguard these assets.
• Licensing and Monetization
  • Licensing Models: Develop flexible licensing models that account for the dynamic nature of AI technologies and the various ways they can be commercialized.
  • Revenue Streams: Explore new revenue streams through the strategic licensing of AI-generated IP, while also ensuring that such licensing does not infringe on third-party rights.
• Employee and Third-Party Relations
  • Employee Agreements: Ensure that employment contracts clearly stipulate the ownership of IP rights, particularly for IP created using AI tools or by AI systems.
  • Collaboration Policies: Establish clear policies for collaborations with third parties, such as research institutions or other companies, to prevent disputes over IP ownership and ensure mutual benefit.
• Continuous Monitoring and Adaptation
  • Market and Legal Developments: Stay informed about market trends and legal developments related to AI and IP, as this is a rapidly evolving field.
  • Policy Updates: Regularly update company policies to reflect changes in the legal landscape and the company's own use of AI.
• Education and Training
  • Staff Training: Provide training for staff on the importance of IP rights in the context of AI, including how to identify and protect new AI-generated IP.
  • Legal Team Development: Ensure that the company's legal team is well-versed in AI-related IP issues and is capable of addressing them effectively.

With the proposed AI Act, the European Union is taking steps to regulate AI systems, introducing new legal requirements and standards. Our firm is closely monitoring these developments, ensuring that our clients are prepared for the impact of this legislation on their AI-related activities and their IP portfolio. At CMS, we combine our deep knowledge of IP law with a keen understanding of AI technology, positioning us to help clients navigate the legal intricacies of this dynamic field. Whether you are developing AI technology, using AI in your business, or dealing with AI-generated content, we provide the legal expertise to secure your IP rights and advance your interests.