March 24, 2026 • Tracy Okal

Introduction

The Artificial Intelligence Bill, 2026 sponsored by nominated Senator Karen Nyamu M.P was introduced before the Senate. It proposes a comprehensive legal and institutional framework for the development, deployment, and governance of Artificial Intelligence (hereinafter ‘AI’) systems in the country.

It defines Artificial Intelligence as a technology or system that uses data and rules to perform tasks on its own, producing results like predictions, decisions, recommendations or content. It includes systems designed to do things normally requiring human thinking, such as making decisions, understanding language, or recognizing images.

It also defines generative artificial intelligence” means an artificial intelligence system capable of generating text, images, audio, video, or other content based on learned patterns from data inputs.

Key Provisions of the Bill

1. Establishment of the Office of the AI Commissioner

The Bill proposes the establishment of the Office of AI Commissioner designated as a State Office tasked with overseeing the Act, as well as receiving and investigating complaints about AI systems. Further, the Commissioner will create policies and guidelines for AI governance, and manage regulatory sandboxes for safe AI testing, among other functions.

2. Establishment of an advisory committee on Artificial Intelligence

The proposed legislation establishes an Advisory Committee that will support the AI Commissioner by providing expert advice on emerging trends, risks, and opportunities in AI. It will also review and recommend improvements to AI guidelines and regulations.

3. Risk Based Classification of AI Systems

The Bill adopts a risk-based classification of AI systems according to the level of risk they pose to health, safety, fundamental rights, the environment or societal welfare.  It categorizes the risks either as unacceptable, high, limited or minimal.

4. Obligations for deployers of high-risk AI systems

Deployers of high-risk AI systems are mandated to conduct a risk assessment before deployment and implement mitigation measures, including human oversight. Further, where the system generates or manipulates images, voice or likeness, such as deepfakes, they are required to obtain explicit consent from the affected person or their legal representative, and ensure the output is clearly labelled as AI generated.

5. Responsibility of deployers of AI systems

Under this section, providers or deployers of an AI system must be transparent with users and affected individuals. They are required to disclose the system’s nature, purpose, and its limitations.

Additionally, they are required to explain the role of automated processes and any human involvement in decisions and ensure there are no biases for fairness.

6. Regulatory Sandboxes

The Commissioner is tasked with establishing regulatory sandboxes for testing artificial intelligence systems in a controlled environment and prescribe conditions for participating in the same

7. Ethical Guidelines

The Bill further mandates the Commissioner to develop and publish ethical guidelines for the development, deployment and use of artificial intelligence systems

8. Penalties

Non-compliance with the proposed Act can result in fines of up to five million shillings or to imprisonment for a term not exceeding two years, or to both.

Concerns arising from the Bill

There are concerns that the proposed legislation is premature as it lacks the necessary infrastructure to support its implementation. It is modelled on the European Union AI Act which as opposed to Kenya, is built for over 20 countries with national supervisory authorities and robust infrastructure to back compliance.

Pre-deployment requirements of high-risk systems such as pre-deployment risk assessment, human rights impact assessment, and maintenance of records of training data, are seemingly designed for powerful AI. When this is transposed to the Kenyan AI landscape with no major AI company, it ends up being a compliance nightmare to the small and local innovators who form majority, if not the entirety of the Kenyan AI sector

Further the Bill creates significant functional overlaps with the already existing Office of Data Protection Commissioner (ODPC) and Communications Authority (CA). It assigns the Commissioner, the AI Authority and its Advisory Council functions that parallel those of CA and ODPC but fails to clarify how the AI Commissioner’s enforcement powers will coordinate with these existing regulators. This raises the risk of regulatory overlap, dual compliance burdens, and potential forum shopping by regulated entities.

Conclusion

Be that as it may, the Bill builds on Kenya’s push to regulate AI systems, following the earlier regulatory proposals such as the National AI Strategy 2025-2030 and the Kenya Robotics and AI Society Bill 2023.

If enacted, the Artificial Intelligence Bill, 2026, would place Kenya among a growing number of nations adopting dedicated legal frameworks for artificial intelligence, reflecting the global push to balance innovation with accountability in an increasingly data-driven economy.

Ready to assess your organization's AI readiness? Let us talk.

? Telephone: +254 115 867 309 | +254 740 196 519
? Email: cybersecurity@southendtech.co.ke | info@southendtech.co.ke | dataprotection@southendtech.co.ke

South-End Tech Limited — Helping East African businesses build secure AI-ready foundations.

Comments (0)