Intellectual property as a tool for AI governance and protecting sports assets

By Dr. Amira El-Shinawy

Intellectual property in the modern sports environment is no longer limited to protecting works, trademarks, or content from traditional imitation. Rather, it has become one of the most important legal tools for governing the relationship between sports and artificial intelligence. It allows determining what content, data, and sports identities may be used, who has the right to use them, for what purpose, under what conditions, and what legal responsibility results from exceeding those limits.

This function is becoming increasingly important as sports organizations increasingly rely on artificial intelligence for performance analysis, talent identification, tactical strategy development, content production, marketing, player evaluation, and audience experience enhancement. As algorithms become more capable of learning from sports assets, the need to define the legal nature of these assets and the rights governing their collection, processing, reuse, and exploitation grows.

In this context, protection is not limited to raw sports data as mere numbers or isolated information. Rather, it may extend—subject to the necessary legal conditions—to the innovative structure of the data, databases, the methods used to select, arrange, organize, and process their elements, and the legally or commercially valuable investment and effort involved in their creation.

Furthermore, the confidential aspects of the data, methods of collecting and analyzing it, and algorithms or methodologies of competitive value may be protected as confidential information or trade secrets, provided the necessary protection conditions are met and the required measures are taken to maintain their confidentiality.

This is linked to protecting the athlete's commercial and personal identity; the name, image, voice, trademark, logos, and distinctive features of a player or club are now susceptible to being recreated by artificial intelligence technologies, including deepfakes, augmented reality, and synthetic content. Therefore, intellectual property rights, along with related personal and contractual rights, play a role in regulating the use of these elements and preventing the transformation of a sports identity into a commercial asset exploited by intelligent systems without legal basis.

Equally important is the sports content itself; player images, match footage, audio recordings, sports programs, designs, editorial materials, audio-visual content, and their associated trademarks and databases can constitute integrated economic assets. Therefore, using these materials to train, reproduce, or generate outputs based on artificial intelligence models should not be viewed as a purely technical matter, but rather as an activity where copyright, trademarks, database rights, confidential information, contractual rights, and other forms of protection may intersect, depending on the nature of the asset being used and the applicable law.

Contracts and licenses here serve a complementary function to the rights established by law. Through them, it is possible to move from subsequent protection to proactive protection by defining the scope of access to data, the entities authorized to use it, the purposes of use, the duration and geographical scope, and prohibiting the relicensing, resale, or reuse of data in training artificial intelligence models without explicit consent, as well as establishing mechanisms for tracking, review, notification of breaches, determining liability, and compensation in the event of unlawful exploitation.

This issue takes on particular significance in light of the Getty Images v. Stability AI case before the High Court of England and Wales, which rendered its judgment on 4 November 2025 in case number [2025] EWHC 2863 (Ch). The dispute addressed multiple claims relating to copyright, databases, and trademarks within the context of the Stable Diffusion AI model.

The significance of the case does not lie in claiming that the court ruled generally on the legality or illegality of training AI models on copyrighted works; that conclusion would go beyond the scope of the judgment. Rather, the case reveals the complex layers of protection that can intersect when creative and commercial assets enter the AI development cycle. Furthermore, the court established liability in specific instances related to certain watermarks appearing in the model's output, but this was not a blanket ruling in Getty's favor on all its claims.

From a sports industry perspective, this case offers a crucial lesson: the legal value of a sports asset does not end with its creation, but extends to how it is collected, stored, processed, made available, and fed into intelligent models, as well as the nature of the resulting outputs. A sports image may be subject to copyright, may also bear a trademark, may be part of a licensed database, and its use may be governed by a commercial contract imposing additional restrictions. Therefore, risk analysis should not be based on a single right in isolation from the other elements of protection.

Hence the importance of building an integrated system for managing sports assets in the age of artificial intelligence, starting with inventorying assets and identifying rights holders, classifying data and content according to their legal nature, identifying what is protected by copyright, trademarks, database rights, or trade secrets, and then surrounding access, processing, and use operations with appropriate licenses and contractual restrictions.

This approach is not limited to protecting clubs, federations, broadcasting bodies, and sports analysis companies, but extends to protecting the athlete himself, so that his physical or tactical performance data, or his image, voice, name, or personal brand, do not become unlimited inputs for commercial models over which he has no control over how they are used or the outputs generated from them.

Therefore, intellectual property in the age of artificial intelligence does not only perform a defensive function; it does not wait for an infringement to occur in order to act, but it can act as a proactive legal structure to manage the lifecycle of the sports asset: from its production or collection, to its preservation and classification, then its licensing and availability, up to its use in intelligent models and the outputs produced by those models.

In this sense, intellectual property moves from its traditional concept as a means of protecting creativity from copying, to a broader concept as a tool for managing the economic value of data, content, and sports identity, controlling their transfer to the algorithmic environment, and balancing the availability of technological innovation with the protection of investment, rights, and commercial integrity of the sports industry.

The real question is no longer: Who owns the sports content? Rather, it has become: Who has the right to convert this content into algorithmic knowledge, who has the right to invest in it, and who bears responsibility for the outputs built upon it? Here, specifically, intellectual property emerges as one of the central tools for formulating the rules governing the relationship between sport and artificial intelligence.

Photo credit: Franco Monsalvo | Source: Pexels