Speeches

Remarks by the Minister of Justice and Constitutional Development, Minister Mmamoloko Kubayi, AI and the Law Conference, 09 July 2026

Programme Director – Prof Bongani Ngqulunga, Senior Executive Director, University of Johannesburg
Justice Dunstan Mlambo, Deputy Chief Justice of South Africa
Professor Tshilidzi Marwala, UNU principal
Professor Lethlokwa Mpedi, UJ Vice Chancellor
Distinguished Guests
Ladies and Gentlemen
Good Afternoon

It is a great privilege for me to address this afternoon at this very important conference whose focus has huge bearing on the practice of the law and more importantly how society will be governed in the not-so-distant future. Artificial Intelligence tools have reshaped and are reshaping almost every aspect of human social and economic existence. However, AI machines are yet to master the full capabilities of human cognitive abilities and hence humans and AI machines have to coexist and work side by side. Machines are still learning and modeling the complexities of human behavior but the pace at which they are learning has been increasing exponentially. As machines increase their knowledge and capabilities, they will create a ubiquitous system from which no sector, no economy and no country will be able to escape. Our task, as leaders of society, is to ensure that whatever that artificial intelligence brings to bear on human existence is not harmful but good for humanity.

I am pleased that the University of Johannesburg has taken up this important task of leading us in this important discussion and I would like to commend the visionary leadership of the university for organising this conference. The theme for this conference is: "Artificial Intelligence and the Law – Advancing the Sustainable Development Goals through Law, Governance and Practice" and it goes to the heart of what the world has to grapple with regarding Artificial intelligence. Goal 16 of the sustainable development goals promotes among other things access to justice for all and it is a goal that is central to our mandate as the Department of Justice and Constitutional Development. Access to justice for all is an aspiration that is codified in our constitution however after 30 years of constitutional democracy we are yet to achieve this goal. Significant progress has been made but challenges remain as some communities are still beyond the reach of justice facilities and services. Among the factors that have been a barrier to the achievement of universal access to justice are prohibitive legal costs, geographic remoteness from courts and legal services, lengthy delays in case resolution, and a shortage of legal practitioners willing or able to serve lower-income communities.

Undoubtedly, by automating repetitive processes, advancing legal research, drafting, increasing dispute resolution, and encouraging predictive legal analytics, the introduction of AI-powered solutions has drastically transformed the delivery of legal services. The effect of AI powered tools does not only reduce the time it takes to perform these tasks, but they also do these tasks at a fraction of the traditional cost. The effect of these developments is the rapid increase of access to justice.

There have been a number of cases in which lawyers who used large language models to draft court papers have been caught on the wrong side of the law mainly because the AI machine cited cases that it hallucinated because they do not exist. These cases have always been used by techno-skeptics as evidence for the belief that AI will never be able to master the legal field. For those who have followed closely the growth in computational capability of AI they would have noticed that Artificial Intelligence has now outpaced Moore's law. According to Moore's law computational capabilities double every two years, but with AI it is estimated that the computational capabilities double every seven months. What this means is that what an AI machine can do today, will have changed dramatically in seven months.

We must move from a premise that AI will drastically change the way the law is practiced, administered and enforced and it is here to stay. The issue that we must wrestle with is: what is it that we are going to do to ensure that the use of AI in law does not lead to the perpetuation of injustice?

It was Kofi Annan, the former secretary general of the United Nations, who said: "Good, healthy democratic societies are built on three pillars: there's peace and stability, economic development, and respect for rule of law and human rights. But often, we take stability - peace in terms of security and economic activity - to mean a country is doing well. We forget the third and important pillar of rule of law and respect for human rights, because no country can long remain prosperous without that third pillar."

Indeed, without the rule of law and respect for human rights no country can long remain prosperous and will soon descend into chaos and instability. It is within this context that one can understand the anxiety of those who are asking themselves whether it wise and prudent to leave such an important pillar of our country and democracy to machines.

This issue is particularly important because African participation in AI development is still largely missing as Professor Marwala has repeatedly emphasised. What this means is that African Data which carry our archives, African languages and cultural nuance do not find expression in the AI machines that are currently used in the various fields including law. Most AI is trained on datasets from North America and Asia. This exclusion perpetuates data poverty and algorithmic bias, leading to real-world consequences such as poor language translation machines especially for African languages.

The bigger risk for our country in respect of the law is the use of AI machines which are totally devoid of our historical context as a country. This year, we are celebrating the 30-year anniversary of the adoption of our constitution. Broadly speaking, our constitution was drafted to achieve peace, unity and stability after many years of painful and violent struggle and secondly, to lay a foundation for a transformation of the South African society so as to "Heal the divisions of the past and establish a society based on democratic values, social justice and fundamental human rights" and "Improve the quality of life of all citizens and free the potential of each person". Simply put, ours is a transformative constitution.

Former Chief Justice Pius Langa in his paper delivered at Stellenbosch university entitled "Transformative Constitutionalism" said the following about the task of the judges in relation to interpretation of the law in the context of a transformative constitution, he said: …transformative adjudication requires judges to acknowledge the effect of what has been referred to elsewhere as the "personal, intellectual, moral or intellectual preconceptions" on their decision-making. We all enter any decision with our own baggage, both on technical legal issues and on broader social issues. While the policy under apartheid legal culture was to deny these influences on decision-making, our constitutional legal culture requires that we expressly accept and embrace the role that our own beliefs, opinions and ideas play in our decisions. This is vital if respect for court decisions is to flow from the honesty and cogency of the reasons given for them rather than the authority with which they are given."

What the former Chief Justice was emphasizing was that context, especially the social and historical context, in the interpretation of the law is very critical. The men and women who have been central in the interpretation of our laws and the development of jurisprudence since the dawn of our constitutional democracy have demonstrated through their judgments that they understand the transformative nature of the constitution. Their lived experience has imposed an obligation on all of them to interpret our laws in a manner that reflects a deeper understanding of the impact apartheid has had on the majority of South Africans who were oppressed. It is this understanding that has placed social justice and socioeconomic transformation at the centre of our justice system.

The question is: How and when are we going to transfer this deep understanding of our past and social condition that our past created for the majority of our people into AI machines?

AI systems are as good as the data in which they are trained and currently the reality is that most of the data that used is riddled with racial bias. As a country we have seen and experienced the horrific consequences of a nation organised through racial segregation. We need to ask ourselves: what is it that we need to do that technological progress is not concomitant with the growth of digitized racial segregation?

The data that carries our history and our social context has been excluded in the development of the AI machines that are currently in the market. How then should we expect that the machines which are, in all likelihood, biased against us will be able to help us build a just society? Are we not, in our quest to increase access to justice through AI, at the risk of increasing access to injustice?

The researchers and the leaders of the universities should be able to help to answer this question and many other questions that have arisen with the development of artificial intelligence.

Transparency and explainability remains a major challenge in the use of Artificial Intelligence. Without a clear analytical framework through which we can understand how AI arrives at its conclusions, in this case judgements or legal opinions, we run the risk of creating so-called "black-box" systems where the algorithms automating decisions are inscrutable to those who are managing the systems. Artificial intelligence tools in their current limit mechanisms for holding them accountable for their decisions.

Another question of justice that has arisen very sharply with regards to AI machine development is the question of environmental justice. The increase in the scale and computational capacity of AI machines comes at a huge environmental cost because these machines require large amounts of energy and water both of which are scarce. Will AI machines destroy themselves to save the environment? Put differently, will AI machines be able to dispense environmental Justice if it means self-destruction? Or will humanity have to bear the cost?

Let me conclude with the of Antonio Gutteres, the Secretary General of the United Nations, who, speaking ahead of the Global Dialogue on AI three days ago, said that: "AI is advancing at runaway speed. The question is whether we will govern it together – or let it govern us. ... We must now turn global participation into global action – to make AI safer, fairer, more accessible and more ethical."

I have no doubt this conference will attend to all these very important issues so that we can be able to utilise AI to advance the Sustainable Development Goals through Law, Governance and Practice. I look forward to the outcomes of this conference. I wish you all the best with your deliberations.

I thank you.