In my previous article, “The Silent Fragmentation: The Internet Is Quietly Breaking Apart, and Africa Must Choose Its Side,” I explored how the digital world is increasingly shaped by competing geopolitical interests. The internet, once a promise of openness and global connectivity, is gradually fragmenting into competing spheres of influence.

Nowhere is that fragmentation more significant than in artificial intelligence.

The question is no longer who builds AI.

The question is who governs it.

Across the world, governments are establishing frameworks to regulate artificial intelligence. The European Union has introduced the AI Act. The United States continues to expand its governance architecture through executive orders and risk management frameworks. China has implemented regulations governing algorithms and generative AI. UNESCO’s Recommendation on the Ethics of Artificial Intelligence has become the first globally agreed framework for responsible AI.

These are not merely technical documents.

They are expressions of values.

They define what societies consider fair, safe, and acceptable.

The challenge for Africa is that many of these rules are being written without sufficient African participation.

As a result, we risk becoming consumers of AI systems governed by standards we did not help create.

This is not simply a technology challenge.

It is a leadership challenge.

Artificial intelligence is becoming the new infrastructure of power in the twenty-first century.

It will influence how governments deliver services, how businesses assess risk, how banks make lending decisions, how healthcare systems allocate resources and how education systems personalise learning.

The countries shaping AI governance today are helping determine how future economies will operate.

According to the 2025 Stanford AI Index, governments worldwide are accelerating AI governance efforts, with regulatory activity, national AI strategies and public investment continuing to grow as nations compete for technological leadership.

This is why AI governance matters.

The conversation is often framed as innovation versus regulation. Ultimately, it’s about power, accountability, and trust.

Who decides how AI systems are deployed?

Who is responsible when harm occurs?

Whose values are embedded within the technology?

And whose voices are missing from the conversation?

Despite growing AI adoption across the continent, African participation in many global AI governance forums, standards bodies and technical working groups remains disproportionately low compared to Europe, North America and Asia.

For Africa, these questions are becoming increasingly urgent.

Ghana’s ongoing debate around the proposed National Information Technology Authority (NITA) Bill 2025 provides an important case study.

Supporters argue that stronger oversight is necessary to improve digital governance, cybersecurity and accountability across Ghana’s growing digital economy.

Critics raise concerns about regulatory burdens, innovation constraints and the concentration of authority.

Both perspectives matter.

Yet the most important question sits beneath the debate itself:

How do African nations build governance structures capable of supporting innovation while protecting citizens?

This is not a uniquely Ghanaian challenge.

It is a continental one.

The countries that attract investment in the AI era will not necessarily be those with the weakest regulation or the strongest regulation.

They will be those with the most trusted regulation.

Investors seek certainty.

Entrepreneurs need clarity.

Citizens require protection.

Governments need accountability.

Increasingly, boards and investors are evaluating organisations not only on their AI capabilities, but on their ability to manage AI risk, governance and public trust.

Balancing these priorities is the real test of AI governance.

One of the greatest misconceptions about AI ethics is that it is primarily a compliance exercise. It is not.

Ethics begins where compliance ends.

Organisations can comply with regulations while still deploying systems that cause harm.

We have already seen examples of facial recognition technologies performing significantly worse on darker skin tones because the datasets used to train them lacked diversity.

We have seen healthcare algorithms trained predominantly on Western patient populations produce questionable outcomes when applied elsewhere.

We have seen automated decision-making systems reinforce existing inequalities because historical bias was embedded within the data itself.

These failures are often described as technical shortcomings. I view them differently. They are governance failures.

When communities are absent from the design process, they often become invisible in deployment. For African business leaders, this should feel familiar.

Imagine an AI-powered recruitment platform trained primarily on data from Europe and North America. Would it recognise the value of entrepreneurial experience gained in informal markets? Would it understand local educational pathways? Would it fairly assess talent that does not fit imported definitions of success?

Technology does not become ethical because it functions efficiently. It becomes ethical when it produces fair outcomes.

There is another dimension of AI ethics that receives far less attention in boardrooms and government offices.

Surveillance.

Around the world, AI-powered technologies are increasingly being used to monitor behaviour, identify individuals, analyse movement patterns and predict future actions.

These capabilities can improve public safety and national security. They can also be misused.

Without strong governance frameworks, oversight mechanisms and independent accountability, surveillance technologies can threaten civil liberties, press freedom and public trust.

This is why AI ethics cannot be separated from governance.

Technology itself is neither ethical nor unethical.

The institutions that govern it determine its impact.

The question is not whether AI can monitor populations.

The question is whether the systems overseeing that power are sufficiently robust, transparent and accountable.

The most important AI debate facing Africa is not whether we will use artificial intelligence. We already are.

The real question is whether we will help shape the rules governing its use.

Ghana’s commitment to responsible AI offers a promising foundation. But strategies alone are not enough. Effective governance requires institutions, technical expertise, public participation and long-term political commitment.

As someone who has spent much of my career working at the intersection of technology, inclusion and emerging innovation, I have come to believe that ethical leadership is one of the most important forms of leadership in the AI era.

It requires leaders willing to challenge assumptions, question incentives, and consider the long-term consequences of today’s decisions.

Because if the rules governing AI continue to be written without African voices, African contexts, and African values, AI systems will continue to perform well for the world that designed them and poorly for the world left out of the room.

Ethical AI is not charity.

It is not reputation management.

It is not a diversity initiative.

It is competence.

And in the age of artificial intelligence, competence may become Africa’s most valuable strategic advantage.