AI Book Review: Mastering Sovereign Artificial Intelligence by Mark Nasila

A timely exploration of how sovereign AI could move Africa from data colony to data nation, and why leaders and parents cannot ignore it.

Article link: https://open.substack.com/pub/johanosteyn/p/ai-book-review-mastering-sovereign?r=73gqa&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true

Audio summary: https://youtu.be/8dBOO6XgHj0

Follow me on LinkedIn: https://www.linkedin.com/in/johanosteyn/

I have always been an avid reader, and over the past decade, my shelves have filled up with books on technology and Artificial Intelligence. These subjects are no longer abstract interests; they shape the work I do with organisations, policymakers and educators. Reading widely is one of the few ways to stay ahead of the curve in such a fast-moving field. Writing these reviews is my way of sharing that journey, and hopefully encouraging others to pick up the books themselves rather than relying only on headlines and soundbites.

In Mastering Sovereign Artificial Intelligence: Owning and Shaping the Future in the Age of AI, Dr Mark Nasila makes a bold argument: countries that merely export their data and import foreign AI models are setting themselves up as digital colonies. Published in 2025, the book treats data as a strategic resource, much like minerals or water, and warns that leaving its processing and interpretation to distant tech powers is a direct threat to sovereignty. Nasila calls on African societies in particular to become “data nations” instead: builders of their own AI capabilities, aligned with their laws, cultures and priorities. It is an ambitious book, dense at times, but it speaks directly to anyone worried about who will control the intelligence built on Africa’s data.

CONTEXT AND BACKGROUND

Nasila’s starting point is simple but uncomfortable. In the industrial era, power flowed to the countries that controlled factories, shipping routes and energy. Today, a new layer of power sits with those who own the data pipelines, the computing infrastructure and the models that make sense of all that information. A small group of global platforms, mostly headquartered in the global North, now mediate everything from payments and communication to security and healthcare. In that context, African states, banks and telcos risk becoming long-term tenants in someone else’s digital property.

The book situates sovereign AI within this emerging “digital empire”. Sovereignty, in Nasila’s view, is not just about having data physically stored within a country’s borders. It is about the ability to govern how data is collected, processed and used, to embed local legal and ethical frameworks into AI systems, and to ensure that value created from data flows back into local economies. For a continent that has long exported raw materials and imported finished products at a premium, the parallels are hard to ignore.

INSIGHT AND ANALYSIS

One of the strengths of Mastering Sovereign Artificial Intelligence is its insistence that sovereignty is a multidimensional problem. It is technical, certainly: you need data centres, connectivity, computing power and robust cyber security. But it is also legal, ethical and cultural. Nasila repeatedly stresses that AI systems trained on data and values from elsewhere may not reflect African realities, languages or social priorities. Sovereign AI, in his framing, is partly about the right to encode one’s own norms into the systems that will make more and more decisions about citizens’ lives.

The sections on “AI factories” are particularly useful for business leaders. Here, the book moves from rhetoric to operating model. An AI factory is not a magical machine but a set of processes, platforms and teams that systematically turn data into decisions: from ingesting information, to training models, to deploying them into products and services, and then monitoring their impact. This is the language that boards, risk committees and regulators can engage with. The message to African organisations is blunt: if you simply plug into foreign models without thinking about ownership, localisation and governance, you are building your future on sand.

At the same time, the book is more aspirational when it comes to politics and state capacity. It is easier to call for sovereign AI than to design and maintain credible institutions in environments marked by fiscal pressure, skills shortages and, in some cases, corruption. Nasila acknowledges these challenges, but readers may wish for a deeper engagement with the messy realities of implementation. Nevertheless, his high bar is valuable precisely because it exposes the gap between our rhetoric about “digital transformation” and the hard work required to own our destiny in the age of AI.

IMPLICATIONS

For policymakers, the implications are stark. AI cannot be treated as a side project delegated to a small team in a distant ministry. Investment in connectivity, data infrastructure, skills pipelines and regulatory capacity becomes part of the basic toolkit of modern sovereignty. Decisions taken today about cloud procurement, data localisation and public–private partnerships will shape the room for manoeuvre for decades.

For business leaders, the book is a warning against outsourcing their AI thinking to overseas vendors. Buying ready-made models may offer short-term efficiencies, but it can deepen strategic dependency and limit the ability to respond to local regulatory, cultural and market demands. Boards need to ask harder questions about where data is stored, who controls the models, and how decisions made by AI systems can be explained and challenged.

Parents and educators are also part of this story. As AI systems increasingly influence who gets a job, a loan or medical care, the question of whose values and assumptions sit inside those systems becomes deeply personal. Sovereign AI, at its best, is about ensuring that the future our children inherit is not entirely coded elsewhere.

CLOSING TAKEAWAY

Mastering Sovereign Artificial Intelligence is not an easy airport read, but it is an important contribution to the conversation about Africa’s digital future. Nasila forces us to confront a simple choice: will we continue exporting our data and importing intelligence, or will we develop the capabilities to shape AI on our own terms? The core message is compelling. If we ignore the sovereignty dimension of AI, we may wake up to find that the key decisions about our economies and our children’s lives are being taken in distant data centres we neither see nor control.

Author Bio: Johan Steyn is a prominent AI thought leader, speaker, and author with a deep understanding of artificial intelligence’s impact on business and society. He is passionate about ethical AI development and its role in shaping a better future. Find out more about Johan’s work at https://www.aiforbusiness.net