AI AS THE SIDE KICK AND NOT SUPERHERO: Navigating AI in the legal field in Kenya

Introduction

The integration of AI into Kenya’s legal field is becoming increasingly evident. From students pursuing their undergraduate work to advocates defending their clients in court. A space which has traditionally been shaped by precedent through the doctrine of Stare decisis and meticulous analysis is now transformed by Artificial Intelligence. It is a fact that this is observed not only in Kenya but also worldwide. In the legal sector around the world, there is an increasing number of legal professionals using large language models(LLMs) to help summarise documents, cross-reference cases and conduct various research.¹ Examples of such LLMs are ChatGPT, DeepSeek, Gemini, amongst others.² Two recent judgments in the UK, the case of Ayinde v London Borough of Haringey and AI Haroun v Qatar National Bank, have revealed the dangers of mishandling of AI tools for case preparation and legal guidance.³ In both cases, the advocates used AI to the detriment of the case, whereby in the former, the court found that they had met the threshold for initiating contempt proceedings by deliberately including fake citations.⁴ In the latter case, the solicitor relied on AI-generated research done by the client without verifying, and the court held that they cannot delegate their professional work to AI.⁵ This, amongst other cases, is an example of how AI is not being used correctly in the field. The truth is that AI is here to stay, and as it penetrates the legal space it is important that we use it to complement, improve and make the trial process more efficient. AI is not here to replace lawyers but to enhance the lawyer’s craft.⁶ Therefore, we should move far from the AI-solutionarism approach, which is a popular belief that given enough data, machine learning algorithms can solve all humanity’s problems.⁷ This article examines Kenya’s use of AI in the legal sector. It will examine the benefits that AI can bring to this sector, how the country, through its policies, is addressing this emergence, and the general perceptions of AI in the Kenyan legal sector.

Current affairs

First, it is important to note that the backbone of this legal sector, the Judiciary, is currently developing a policy framework for the adoption of AI. This Framework is designed to manage and monitor AI adoption in Kenyan courts and tribunals by ensuring it is aligned with the Constitution and champions a people-centred approach to dispute resolution by expanding access to justice and fostering social harmony.⁸ This framework sits on various principles such as the principle of people-centred access, equity and non-discrimination, transparency and explainability, accountability and oversight, reliability and security, judicial independence and integrity and efficiency and effectiveness.⁹ The framework shows that AI is not here to replace but complement the trial process by addressing backlogs, reaching the underserved communities, enhancing accuracy and bolstering public trust and ultimately still retaining judicial discretion.¹⁰ The human-in-the-loop principle remains a constant within the framework. It ensures that judicial officers and judges will maintain the ultimate oversight over any AI-driven recommendations, guaranteeing constitutional safeguards and upholding Article 159 of the Constitution on judicial authority by preventing overreliance on AI tools.¹¹

Speaking at the engagement between the Supreme Court of Kenya and the Federal Supreme Court of Ethiopia on 11th August 2025, the Chief Justice Martha Koome stated that no one would be left behind in the digital shift, and they are currently establishing ICT help desks within Huduma centres and court stations so as to ensure that users navigate online services and e-filing systems.¹²

AI as a tool and not a replacement for human expertise

As AI continues to revolutionise the legal profession, it is crucial for legal practitioners to embrace this transformation while upholding the principles of good governance and the rule of law to ensure sustainable development in this space.¹³ These tools are becoming integrated in various services such as reviewing contracts and agreements, identifying potential legal issues and risks.¹⁴ Additionally, it streamlines research through Natural language processing and machine learning algorithms that analyse large datasets, thus identifying relevant informatio,n saving time.¹⁵ Also, by analysing past court decisions, client behaviour and certain trends,it enables them to make predictions on the likelihood of success or potential risks associated with certain causes of action. Lastly, it enhances client experience by providing more personalised services such as preparing legal documents. An example of such Kenyan AI tools includes M-wakili, which provides immediate and accurate legal answers to legal questions based on the Kenyan constitution and other relevant laws.¹⁶ Secondly, an example of a machine learning tool includes Wakili AI, which offers public legal help, NGO resources, professional tools and the ability to find a lawyer across Kenya.¹⁷

While AI has the potential to enhance and make the legal field more efficient, we must ensure that it remains a tool rather than a replacement. While this tool can provide relevant precedent, conduct research, and predict outcomes, it is the lawyer’s discretion, judgment, and strategic thinking that enable a lawyer to prevail in a case. These tools should complement their expertise and augment their decision-making process and not replace it entirely, as they should be the ultimate decision makers that uphold the best interest of their client.¹⁸ Of emphasis is the inclusion of the Human in the loop principle, where the advocates have a final role in the ultimate decision-making process and are required to validate Machine learning output.¹⁹ It’s important because each case presents distinct nuances, and with these come individuals’ rights and justice that must be served expeditiously. By retaining the essential human elements in the justice system, they remain²⁰ the final arbiters in any judicial decision.

With the rise of AI in this field comes the need for advocates to upskill and reskill. This debate featured in the Volume 9 publication by the Judiciary on the Future of Legal Education in Kenya: Integrating Technology, Law and Research.²¹ The article proposed three main recommendations to better equip students for the future and to advance legal education. First, Integrating Technology into the existing courses and modernising pedagogy.²² This includes incorporating topics on Law and technology in core subjects such as constitutional law and criminal law, thus shifting from traditional lecture-based teaching towards student-centred pedagogical methods.²³ Second, developing interdisciplinarity.²⁴ This involves courses on Law and technology, information technology and digital finance, electronic transactions that allow the students to understand the legal implications of these technological advancements.²⁵ Some of these recommendations are already visible in practice with Law schools such as the Strathmore Law School offering units such as ICT law. Additionally, on February 6th 2025, Kabarak took a bold step in revolutionising the digital space by launching the Emerging ICT digital skills and programming academy, which aims to empower students and professionals to have the requisite skills and digital competencies in this evolving technological space.²⁶ The impact of introducing courses on Law and Technology is that it will enable students to understand the legal implications of these technological advancements. Second, with the rise of technology driven sectors such as fintech and e-commerce, the legal profession must take into account the technological and legal aspects of these advancements.²⁷ Lastly, it emphasises on legal practical skills and ethics whereby there is emphasis on research and writing skills and the necessity of ethical conduct and preservation of the rule of law.²⁸ The article also notes that institutions such as the Centre for Intellectual property and Information Technology offer courses on Data protection, which is crucial in this revolution and in further building knowledge and transferable skills towards understanding ethical use and policy implications of technology in the legal profession

In the spirit of this discussion, a recent judgment passed by the Court of Appeal on 3rd October 2025 exposed the drawback of using AI tools without verification.²⁹ In the case of Kenya Agricultural and Processed Food Products Development Authority and Krish Commodities Limited, the court in its judgment relied on a non-existent act called the Geographical Indications Act in deciding the question of protection of Geographical Indications in Kenya.³⁰ This elicited considerable shock in the legal sector, demonstrating that the misuse of AI, once a hypothetical concern, is becoming a tangible reality. Therefore, lawyers and judges must be able to critically evaluate the output from AI tools, understanding the data used to train them and ensuring that it is not perpetuating biases or making decisions that violate ethical and legal standards.³¹

This poses a rising challenge on AI hallucinations whereby AI confidently generates false information including fabricating legal citations. This risks deciding the case in a manner that is not just and to the detriment of the disputing parties which beats the whole purpose of having courts that ensure fair and just trials. Given the possibility of such a case, Kenya should introduce mechanisms such as sanctions whereby such decisions are made in such AI hallucinations given these judges or even lawyers must have a duty to verify all the citations and references that are generated by AI.³²

Challenges

As this revolution continues, it is important that ethical standards are prioritised in the deployment of these AI tools so as to ensure fairness and equity and transparent and accountable decision making.³³ In ensuring this, lawyers must be well equipped with data protection measures and thus ensure that the rights of the clients and their privacy are secure.³⁴ This is evident in upskilling through short courses or in the incorporation of such courses into legal education.

Recently, another concern was raised by Justice Isaac Lenaola, a Supreme Court judge, who expressed dismay upon reading a case in which a judge was horrified to discover that the stellar submissions made by an advocate were based on fabricated authorities generated by AI. He warned lawyers not to use AI for legal filings until formal guidelines are issued, and he revealed that the judiciary has established a team to develop these guidelines for the use of AI in court processes. He encouraged the inclusion of the Law Society of Kenya. He emphasised the focus by the judiciary to ensure they become paperless and ensure sustained innovation in the world of technology.³⁵

Conclusion

In conclusion, the discussion of AI in the legal field is becoming increasingly exciting, as it will ensure rapid and more efficient access to justice. This conversation presents both challenges and opportunities in this profession, as discussed. However, with the judiciary taking steps to embrace AI and uphold the principles of good governance and the rule of law, this is a positive development. To continue this journey till the promised land, it’s important for AI experts, legal professionals, policy makers and other stakeholders to collaborate and create transparent guidelines, frameworks and even laws that ensure responsible and effective deployment of AI.³⁶ While this tool continues to evolve, the legal space must still be committed to upholding good governance and the rule of law.³⁷

Image used is from Shutterstock