AI and Intellectual Property (Pt. 1): Safeguarding Creations in the Digital Age

 

Carlos Casado

IP advisor at Latin America IP SME Helpdesk

 

AI's Recent Boom

In the past decade, we have witnessed a technological revolution that is reshaping every aspect of our daily lives – it is the rise of Artificial Intelligence (AI). From virtual assistants on our smartphones to self-driving cars, it has become a disruptive force that redefines how we interact with technology and the world around us.

AI is a science that seeks to develop systems/machines capable of performing tasks that require human intelligence. This includes the ability to learn from data, recognize patterns, make autonomous decisions, and in some cases, even emulate human cognitive processes such as thinking and problem-solving.

The three fundamental pillars for understanding how AI learns and makes decisions are Machine Learning, Deep Learning, and Reinforcement Learning.

First, Machine Learning focuses on creating algorithms and models that enable machines to learn from data and make decisions based on that learning, rather than explicit instructions. An example is Netflix's recommendation algorithm, which suggests movies based on user preferences and past viewing behavior.

Second, Deep Learning is a subset of Machine Learning that employs interconnected neural networks, mimicking the human brain, to autonomously learn from vast amounts of data and algorithmic processing. For instance, facial recognition on Facebook, where the neural network automatically identifies and labels people in photos.

Third, but equally important, Reinforcement Learning is based on a machine's interaction with its environment to maximize a reward, in other words, to help the system understand which decisions are good and which are not based on its current surroundings. One of the most well-known examples is AlphaGo, developed by DeepMind, which learned to play the board game Go and defeated the world champion human player in 2016.

 

AI and Intellectual Property (IP)

One of the most intriguing and challenging aspects of this technological advance is its impact on intellectual property. AI can create music, art, literature, and digital content with astonishing originality and quality, raising questions such as: who is the author of the work, the programmer, or the machine? How can I protect the content that AI creates for me?

In addition, it has proven to be a valuable tool in detecting and preventing copyright and trade mark infringement by scanning vast amounts of online content.

It is worth mentioning that the creation of works by AI has important implications for copyright. Traditionally, works are protected if they are original, although the CJEU has stated on occasion (Case C-5/08, Infopaq International) that copyright only applies to original works and that originality must reflect the author's "own intellectual creation". This is generally understood to mean that an original work must reflect the personality of the author, which clearly implies that there must be a human author for a work to be protected by copyright.

However, the latest AI systems are no longer mere tools but make many of the decisions associated with the creative process without human intervention.

The protection of AI-created content is a complex and controversial issue that has no single definitive answer. It depends on factors such as the type of content, the degree of human intervention, the applicable legal framework, and the interests at stake. Nevertheless, some recommendations to protect against this legal uncertainty are:

• Enter into an AI development contract. When working with software developers, we should establish clear contracts defining the ownership rights of any content generated, not forgetting to address copyright ownership and any licensing required to use the content.
• Another interesting option is to associate a brand with the AI-generated content. Once we are sure that the content is unique and has commercial value, which could be anything from a software application to an online service, we should register a trade mark for that product or service through the EUIPO or the National Intellectual Property Office, depending on the degree of protection we seek.

 

AI Protection in Latin America: Costs and Strategies

In the context of Latin America, the protection of Artificial Intelligence (AI) is an issue that involves both international agreements and national intellectual property regulations. To safeguard AI creations in this region, it is essential to consider three main approaches: the Patent Cooperation Treaty (PCT), the Madrid System for the International Registration of Marks and the Hague Agreement on the International Registration of Industrial Designs, as well as copyright protection at the national level.

International protection systems (PCT, Madrid, Hague)

The PCT is an international agreement that facilitates the process of applying for and protecting patents in several countries with a single application. The system is administered by WIPO and currently has 153 contracting states, including most Latin American countries, such as Argentina, Brazil, Chile, Colombia, Mexico, Peru, and Uruguay. Registering a patent through this system would cost around €1,378.

If your AI is associated with goods or services that you wish to market in Latin America, the Madrid System provides an efficient route to register and protect trade marks in several countries in the region, thus ensuring the identity and recognition of your brand in the Latin American market. This system is also administered by WIPO and has 108 contracting states, including Colombia, Cuba, and Mexico. The basic fee for registering a trade mark is around €600 to €825.

For those whose AI has generated product-related industrial designs, such as user interfaces or aesthetic elements, the Hague Agreement offers the possibility to protect these designs in multiple Latin American countries. This can be crucial to ensure the exclusivity and ownership of the visual elements of your product. The basic fees for registration are around €364 for one design and €18 for each additional design included in the same application.

 

National IP protection

In addition to international approaches, it is necessary to consider the national copyright laws in each Latin American country. Although some works generated by AI may not qualify for copyright protection due to the absence of a human author, it is important to understand the specific regulations of each country and assess whether AI can be considered a "creative extension" of its human team.

If we take Brazil as an example, according to the INPI, the fees for registering a patent range from €13 to €308 per application, while registering a trade mark would cost between €13 and €62. All of this does not include additional costs such as examination, granting, or annual fees. It's important to highlight discounts of up to 90% for patents and 60% for trade marks may be applied to individuals, microenterprises, educational institutions, and research entities, among others.

Regarding Mexico, the IMPI sets the patent registration fees at around €143 (excluding VAT), and for trade mark issuance, it amounts to approximately €113 (excluding VAT).

 

Contact us

If you have any further questions after checking out this article on AI & IP, don’t hesitate and contact us!

You can also address all your questions directly to the European IP Helpdesk, regarding AI protection in Europe.