Navigating Intellectual Property Risks of AI-Generated Works Through Corporate Governance

[Rishi is a student at National Law University Jodhpur.]

The rapid advancements in artificial intelligence (AI) have significantly transformed the technological landscape worldwide, fostering innovation while simultaneously raising substantial legal and governance challenges. According to the World Intellectual Property Organization (WIPO), there has been a steep rise in AI-related academic publications, global corporate AI investment [Figure 1], and global artificial intelligence market size; however, AI and specifically ‘generative AI tools’ are substantially affecting the creative market and raising AI-related intellectual property (IP) issues such as lack of copyright protection for machine-generated works, infringement concerns over training data, and ambiguity in patentability due to unclear inventorship.

Figure 1

This article examines the critical role of corporate governance in addressing AI-related IP Issues, emphasizing its importance in mitigating risks, ensuring compliance, and fostering sustainable innovation. By exploring legal perspectives and governance frameworks, the discussion aims to provide insights into how companies can navigate the complexities of managing AI-driven IP assets in a rapidly evolving regulatory environment. 

Understanding AI and Intellectual Property

IP rights generally include copyrights, patents, trademarks, trade secrets, etc. Similarly, IP in the realm of AI refers to the legal protections granted to the works, inventions and creations produced by AI technologies, such as general-purpose technologies, etc. Accordingly, automation of AI creation involves machine learning and its varied process such as supervised learning, unsupervised learning, reinforcement learning. Majorly, the generation involves the input of ‘training data’, formulation of ‘learning task and networks’, and output of generated data based on the prediction rule.

In Beijing Film Law Firm v. Baidu Network Technology Company Limited, the Beijing Internet Court gave some primary observations. First, a work that should be protected under the copyright law should meet two requirements, i.e., the work being created by a human or ‘natural person’ rather than a machine and the originality of the work. Second, an AI system, i.e. deemed to be a software machine does not satisfy to be an author of works and thus, shall not be protected under the copyright law. Additionally, the court held that, “although the software developer could not sign as an author on the works, it could developer (owner) can charge software use fees or resort to other means to pursue benefits, for which the input in the software development has already been rewarded.”

Accordingly, the final patent rulings in UK, USA, and Europe, regarding the DABUS AI system have held similar views that “an invention or creation is patentable only if a ‘natural person’ is named as the inventor”. Furthermore, in most jurisdictions, algorithms or AI-generated works do not qualify as ‘patentable’ as they lack technical character and are termed as vague systems, and computer programs without technical effect. European Patent Convention, Article 52(2)(c) also excludes ‘computer programs’ from being patented; however, the European Patent Office allows the patentability of a work if it holds a “technical character” to serve a “technical purpose”.

While the boundaries of IP in the realm of AI-generated works are yet not clear, the test of “technical character” and “human involvement” in the creative process is critical. Although, the development of a clearer model of authorship in AI-generated works is required, herein, the question of IP subsistence highlights the need for corporate governance mechanisms to manage this issue.

Corporate Governance and Way Forward for AI IP Disputes

Corporate governance

Corporate governance refers to “a combination of rules, practices, processes, and laws…that governs the relationships between a company’s board of directors, management, shareholders and stakeholders”. Courts, in different jurisdiction have recognized the principles of good corporate governance as; accountability, compliance, equity, reliance, security, transparency, risk management, etc. Accordingly, the structure of corporate governance involves 3 key players; namely, Board of Directors (BOD), management, and shareholders.

Herein, primarily, the BOD takes up the primary responsibility for corporate governance, and the directors take crucial decisions to attain the company’s long-term business objectives. Similarly, the directors are elected by the shareholders or the board, who rely on them for tracking corporate affairs on daily basis. Courts have held that the BOD is responsible for mitigating company’s risk appetite, allocating capital, approving corporate strategy, and reviewing risk management processes. Herein, effective corporate governance ensures management of AI-related IP issues, as it affirms a given structure through which companies can manage risks and ensure legal and ethical standards.

Relationship of corporate governance and AI-related IP issues

The WIPO director has stated that “global regulatory frameworks for AI and IP will be increasingly vital in the coming years”. The current IP framework is aimed at facilitating human creativity and innovation toward an economy that provides a sustainable development basis for inventing and creating. It is in this context that there is growing talk towards subjecting AI technologies under IP protection for firms operating technology, mainly in order to create a vibrant economic environment for the content generated by AI. But this nexus of AI and IP raises a number of concerns to which corporate governance must also pay focus.

From a regulatory standpoint, key issues include:

• Full transparency in AI processes yet not at the cost of biases, thereby not inhibiting the competitive advantage that fuels AI innovation and business success.

• Copyright implications for AI-created works: Navigating the complexities of copyright law when AI generates new content, particularly on questions of authorship and ownership.

• Copyright issues and, consequently, the questions of fair use, licensing agreements, and probability of infringement arise if copyrighted material is used during training of artificial intelligence.

• Branding and trademarks: Anticipating how AI's role in the creation of content and as a decision-maker will change branding strategies and trademark protections. 

• Role of AI in IP enforcement: Regulating the methods by which AI enforces IP rights, safeguards inventions, and contributes to the development of new business models.

The mismanagement of AI-related IP assets introduces increased financial risks for the members of a company, and protection of AI-generated IP assets is governed by complexed web of multiple national and international jurisdictions. In addition, as noted above, since patents cannot be granted to creations attributed to AI systems, principles of corporate governance can be used to ensure strategic-oversight, policy-making, risk management, etc. and that the produced work has a “technical character”.

Additionally, while a cohesive system is being finalized to effectively resolve these potential issues, an effective corporate governance plan on AI-related IP issues will, in the meantime, ensure the accuracy of the AI system being used, the use of past data being used, compliance with the laws, risk-aversion policies, etc. Since the primary focus of governance is to promote better decision-making processes and improve accountability rate, leading companies are developing some proposals and regulation on governance of data based on the above-mentioned corporate governance principles.

Several regulatory guidelines and frameworks such as EU’s Ethics Guidelines for Trustworthy AI, OECD’s Recommendation on Artificial Intelligence, and Process on Generative AI, and UNESCO’s Recommendation on the Ethics of Artificial Intelligence, are based on the principles of corporate governance such as accountability, privacy, fairness, etc.

Majorly, incorporation of principles of corporate governance would include assessment processes into existing internal governance mechanisms [Figure 2]. Herein, the recommendation includes involvement of top management and the operational level stakeholders.

Level	Relevant Roles
Management and Board	The top management evaluates the AI system on different levels of deployment, procurement, development, etc.
Corporate Responsibility Department	Evaluates the regulatory and technological compliance of the AI system.
Product and Service Development	Evaluates the final AI-based product and discusses the results at the management level.
Quality Assurance	Monitors and reports on the failures and performance issues of the AI system, ensuring quality standards are met.
Human Resource	Ensures the appropriate training data and methodologies are used for the AI system to maintain high training standards.
Procurement	Oversees the trustworthiness and compliance of the procurement process for AI-related components and services.
Day-to-day Operations	Involves developers and project managers, who prepare an assessment list of the day-to-day AI-system results.

Figure 2.

Together, these governance touchpoints form an accountability matrix that not only promotes compliance and transparency but also generates a quasi-legal backbone to protect corporate interests in the absence of clear statutory IP rights for AI-generated content.

Conclusion

The intersection of corporate governance and AI-related intellectual property presents a complex yet evolving opportunity which demands adaptive governance frameworks. As AI continues to redefine creative and inventive processes, existing frameworks struggle to accommodate with the machine-learning based generative works, particularly with regards to the authorship, ownership, and patentability. Herein, effective governance frameworks and practices, such as creating clear and comprehensive IP policies, implementing strong IP management systems, conducting regular IP audits, and importantly, engaging in continuous training for effective and innovative compliance management are required to be incorporated in the existing systems. Furthermore, incorporation of principles of corporate governance, such as transparency and explainability, accountability, ensures responsible disclosure of AI systems for fostering general understanding, awareness of stakeholders in their AI system interaction, traceability of AI inputs, and their AI system lifecycle. By prioritizing these principles, corporations can not only safeguard their intellectual property assets but also contribute to the ethical and transparent evolution of AI technologies in a rapidly advancing world.