Introduction:
The capabilities of Artificial Intelligence (AI) today are nothing short of transformative. Since the onset of the COVID-19 pandemic in 2020, AI has experienced significant global growth, driven by an increased reliance on automation, digital transformation, and data-centric innovation. By 2025, global annual revenue from AI technologies is estimated to be nearly USD 400 billion, with projections indicating a fivefold increase by 2030. Tools like ChatGPT have become integral to workplaces, with over 100 million monthly users as of 2025. Further, studies by Ernst & Young suggest that productivity gains from generative AI (GenAI) could raise global GDP by USD 1.2 to 2.4 trillion over the next decade.
Despite these developments, many jurisdictions, including Kenya, have yet to develop a comprehensive legal framework for AI. This regulatory gap creates significant challenges, particularly in the realm of Intellectual Property (IP), where questions arise regarding the ownership, authorship, and protection of works created or assisted by AI systems. In Kenya, institutions like the Kenya Industrial Property Institute (KIPI) and the Kenya Copyright Board (KECOBO) play key roles in safeguarding IP rights, but the statutory framework is still silent on AI-generated outputs. This article explores the legal challenges surrounding AI and IP in Kenya, particularly the recognition of AI as an inventor, and compares these issues with emerging international jurisprudence.
THE LEGAL LANDSCAPE: KENYA
The Industrial Property Act, Cap 509 Laws of Kenya, governs patent law in Kenya. Section 2 defines an inventor as “the person who actually devises the invention” and includes the legal representative of such a person. The language here is unmistakably anthropocentric, using gendered pronouns like “he” and “she,” and excluding any reference to non-human entities.
Further, Section 22 of the Act stipulates that a patentable invention must be new, involve an inventive step, and be industrially applicable. These criteria, particularly novelty and inventive step, imply a level of human ingenuity and intentionality that current AI systems do not independently exhibit. AI systems function by processing pre-existing data, making it difficult to argue that their outputs meet the originality threshold required under patent law.
Similarly, the Kenyan Copyright Act (Cap. 130 Laws of Kenya) defines authorship in a manner that presupposes human creativity. KECOBO currently does not recognize non-human authors, and there is no precedent or regulatory guidance on whether works created entirely by AI (without human input) can be copyrighted.
THE DABUS CASE: A GLOBAL TEST CASE
Perhaps the most prominent test of AI inventorship has come from the case of DABUS (Device for the Autonomous Bootstrapping of Unified Sentience), an AI system created by Dr. Stephen Thaler. Dr. Thaler filed patent applications in several jurisdictions, naming DABUS as the inventor of a fractal container and a neural flame device. These applications were denied in multiple countries, including the UK, US, and the EU, on the basis that only natural persons can be named as inventors.
- United Kingdom
In Thaler v Comptroller-General of Patents, Designs and Trademarks [2023] UKSC 49, the UK Supreme Court held that the UK Patents Act 1977 requires an inventor to be a “person”—interpreted as a natural human being. Lord Kitchin emphasized that all references to inventorship in the Act point to human agency and intention, and that there was no parliamentary intent to broaden this to include machines.
- United States of America
The United States Federal Circuit in Thaler v. Vidal, 43 F.4th 1207 (2022), similarly concluded that the Patent Act permits only human beings to be inventors. The court affirmed that “Congress has consistently used terms that connote human agency,” including “individual” and “whoever.”
- South Africa
Interestingly, South Africa diverged from this trend by accepting the DABUS patent application in 2021, becoming the first country to recognize AI as an inventor. However, this decision was made administratively without substantive examination and is not seen as persuasive jurisprudence in most common law jurisdictions.
DOCTRINAL CHALLENGES AND POLICY GAPS IN KENYA
Kenya’s existing IP framework does not accommodate AI-generated inventions or works. The core doctrinal challenges include:
- Personhood Requirement: The Industrial Property Act confines inventorship to human persons. This excludes AI systems even where they function autonomously.
- Assignment and Ownership: Even if AI-generated inventions were recognized, questions arise about who holds the rights—the owner of the AI, the programmer, the user, or the AI itself.
- Enforceability: Enabling enforceable rights over AI-generated IP would require amendments to statutory definitions, including terms like “author,” “creator,” and “inventor.”
These challenges underscore the urgency for legislative reform to maintain Kenya’s relevance in global IP governance, especially as AI becomes a key economic driver.
LEGAL AND POLICY DEVELOPMENTS IN KENYA: EMERGING RECOGNITION OF AI AND IP INTERSECTIONS
While Kenya has yet to enact formal legislation addressing the role of Artificial Intelligence in intellectual property (IP) law, there are encouraging signs of institutional awareness and preliminary action. Several governmental and non-governmental actors have begun to acknowledge the legal challenges posed by AI technologies, particularly in the context of innovation, IP ownership, and data governance.
- Digital Economy Blueprint and Innovation Policy
In 2019, the Government of Kenya launched the Digital Economy Blueprint, a policy document intended to guide the country’s transition into a digitally driven economy. Although it does not directly address AI and IP, it identifies “Digital Government,” “Innovation-Driven Entrepreneurship,” and “Infrastructure” as key pillars. These pillars implicitly support the development of a regulatory environment that can accommodate emerging technologies, including AI.
Moreover, in 2022, the Kenya National Innovation Agency (KENIA) published the Draft Kenya Innovation Policy, which explicitly recognizes the transformative potential of AI and calls for mechanisms to safeguard the rights of innovators—including those using AI tools. The policy acknowledges the need for legal frameworks that evolve with technological advancement, though it stops short of proposing detailed legislative amendments to IP law.
- Engagement with WIPO and Regional Institutions
Kenya is an active participant in the World Intellectual Property Organization (WIPO) “Conversation on Intellectual Property and Frontier Technologies”, a global dialogue aimed at developing common understanding and guiding principles. Kenyan delegates have participated in sessions discussing the attribution of inventorship and authorship in AI-generated works. This engagement indicates institutional recognition of the issue, even though domestic legislation remains unchanged.
At the regional level, Kenya is also a member of the African Regional Intellectual Property Organization (ARIPO), which is currently exploring the legal implications of AI in collaboration with WIPO and national IP offices. While no binding protocols have been issued, ARIPO’s exploratory studies signal the possibility of future harmonized regional approaches.
- Role of KECOBO and KIPI
The Kenya Copyright Board (KECOBO) and the Kenya Industrial Property Institute (KIPI)—the primary regulators of copyright and patent rights respectively—have not yet released formal guidance on AI-generated works. However, KIPI has participated in discussions around AI, including at WIPO-led forums, and has acknowledged the growing pressure to clarify the position on AI-generated inventions. KECOBO, meanwhile, has issued public advisories on digital content protection, albeit without directly referencing AI-authored works.
- Legislative Outlook
As of 2025, no bill has been tabled before Parliament to amend the Industrial Property Act, 2001 or the Copyright Act, 2001 to include AI within the definitions of “inventor” or “author.” However, a Technology and Innovation Bill, still in early drafting stages under the Ministry of ICT and the State Department for Innovation and Digital Economy, reportedly includes provisions aimed at fostering AI development and protecting outputs from AI-assisted innovation. Whether these will extend to conferring IP rights on AI-generated works remains to be seen.
THE WAY FORWARD FOR KENYA
Kenya finds itself at a crossroads. On one hand, the country is rapidly digitizing and embracing technologies like AI in sectors such as finance, agriculture, and health. On the other hand, its IP regime remains rooted in traditional, human-centric assumptions. The absence of legal recognition for AI-generated IP could hinder innovation and limit the incentives for investment in AI development.
To address this, Kenya could consider:
- Legislative Reform: Amend the Industrial Property Act and Copyright Act to define the status of AI in relation to authorship and inventorship. This may include defining AI-assisted works and delineating ownership rights.
- Recognition of Human-AI Collaboration: Rather than recognizing AI as an autonomous inventor or author, laws could acknowledge collaborative creations where a human uses AI as a tool. In such cases, the human should retain authorship/inventorship rights.
- Administrative Guidelines: KIPI and KECOBO could issue interpretative guidelines to provide clarity for stakeholders in the absence of formal statutory amendments.
- Comparative Dialogue: Kenya should actively engage with international IP forums like WIPO to harmonize its approach with evolving global standards.
CONCLUSION
The legal recognition of AI in intellectual property law is one of the most pressing and complex issues of the digital age. For Kenya, the answer to whether AI can be an inventor or author is, under the current law, a firm no. Both statutory language and judicial interpretations require human agency for IP protection.
Kenya’s approach can be characterized as nascent and exploratory, marked by soft-law mechanisms, policy frameworks, and multilateral participation. While no concrete legislative changes have been enacted, the country’s positioning within global dialogues and its national digital strategy suggest a trajectory toward regulatory engagement. The absence of a binding legal framework leaves stakeholders—researchers, developers, and IP practitioners—in a precarious position. However, as AI continues to permeate creative and innovative processes, there is a growing need for the legal system to evolve. By proactively addressing these gaps through legislative and regulatory reform, Kenya can position itself as a forward-thinking jurisdiction that encourages technological innovation while upholding the integrity of its IP system.
In conclusion, AI may never be a “creator” in the human sense, but its outputs—and the rights to those outputs—demand serious legal attention. The future of innovation depends on getting this balance right.