WHY BOARDS MUST GOVERN ALGORITHMS, NOT JUST PEOPLE

South Africa’s transition to King V marks more than the release of another corporate governance code, it signals a shift in how organisations must think about leadership, accountability and risk in a digital world. But there is one emerging governance risk that boards cannot afford to underestimate: artificial intelligence (AI).

For years, corporate governance focused on people, executives, directors and employees, and the systems they controlled. Today, decisions are increasingly influenced by algorithms, automation tools, and machine-learning systems. These technologies are powerful and efficient, yet they introduce risks that traditional governance structures were never designed to address directly.

King V arrives at precisely the right time to confront this new reality.

At its core, King V strengthens expectations around ethical leadership, accountability, and responsible oversight. It moves governance beyond compliance and towards value creation, stakeholder trust and long-term sustainability. Boards are expected to take responsibility not only for financial performance, but also for how organisations use technology, manage risk and safeguard stakeholder interests.

This expectation is made explicit in Recommended Practice 109 of the King V Code, which calls on governing bodies to ensure the ethical and compliant acquisition, development and use of technology and information. This provision signals a clear shift: technology governance is no longer optional;, it is a board-level responsibility.

Artificial intelligence sits squarely within this responsibility.

AI systems are already embedded in many business processes, from financial forecasting and customer engagement to fraud detection and operational decision-making. In accounting and finance, for example, AI tools are being used to analyse transactions, generate reports and identify anomalies. In many cases, these tools produce results faster and more efficiently than traditional manual processes.

However, efficiency does not eliminate risk.

Recent international developments have already demonstrated the consequences of inadequate oversight. In Australia, a government-commissioned report valued at approximately $290,000 was found to contain multiple AI-generated inaccuracies, including fictitious references and non-existent academic sources, ultimately leading to reputational damage and repayment obligations for the consulting firm involved.

This example illustrates how even sophisticated organisations remain vulnerable when AI outputs are not properly governed.

Concerns about misuse extend beyond reporting errors. In a separate incident reported in February 2026, a partner at a major global consulting firm was fined A$10,000 for using artificial intelligence to cheat during an internal training course, while more than two dozen staff were found to have used AI tools improperly in internal assessments.

These incidents highlight a broader governance concern: technology does not remove responsibility; it amplifies the consequences of weak oversight.

Legal and governance specialists are also warning that the speed of AI adoption is fundamentally changing decision-making itself. As noted in A practical guide to AI governance for businesses, published by Cliffe Dekker Hofmeyr in April 2026, “AI is compressing decision cycles across every professional workflow. This acceleration has consequences for how human judgement is exercised, how accountability operates and how risk accumulates.”

This observation underscores a fundamental governance risk: when decisions happen faster, the opportunity for review and reflection becomes smaller — increasing the likelihood of unnoticed errors and unintended consequences.

Most concerning of all is the potential erosion of accountability. When decisions are influenced by algorithms, responsibility can become blurred. If an AI system produces misleading results, who is accountable — the developer, the executive, or the board that approved its use?

King V offers a powerful framework to address these challenges, even though it does not focus exclusively on artificial intelligence. Its emphasis on ethical leadership, information governance and risk management provides the foundation needed to govern emerging technologies responsibly.

Ethical leadership, a central principle of King V, requires organisations to act responsibly and transparently. This principle becomes even more important in the context of AI, where decisions generated by automated systems can be difficult to interrogate and explain. Boards must ensure that the technologies used within their organisations reflect the same ethical standards expected of employees and executives.

Risk governance under King V also takes on new meaning in an AI-driven environment. Traditional risk frameworks focused on financial and operational threats. Today, organisations must also assess risks such as data integrity failures, algorithmic bias, cyber vulnerabilities and unintended consequences from automated decisions.

Information governance is another critical area. AI systems rely heavily on data, and poor-quality data leads to poor-quality outcomes. Boards must ensure that data used to train and operate AI systems is accurate, relevant and ethically sourced. Without robust information governance, AI becomes a liability rather than an asset.

The challenge now is moving from awareness to action.

Boards should begin by asking a simple but powerful question: Where is AI already influencing decisions within our organisation? Many organisations underestimate how deeply automation has already been embedded into daily operations. Identifying these areas is the first step towards effective governance.

Next, boards should establish clear accountability structures for AI oversight. Responsibility for technology governance cannot be left solely to IT departments. Instead, AI governance should be integrated into existing governance frameworks, supported by defined reporting lines and oversight responsibilities.

This urgency is reinforced by South Africa’s recent release of the Draft South Africa National Artificial Intelligence Policy, approved by Cabinet in March 2026 and released for public comment in April 2026, signalling government recognition that AI governance must be addressed at a national level.

Ultimately, King V is not just about compliance, it is about foresight.

The organisations that succeed in the coming years will be those that recognise technology governance as a leadership responsibility, not merely a technical one. Boards must understand that algorithms are not neutral tools; they reflect the values, data and assumptions built into them.

Governance in the King V era will require boards to govern not only people and processes, but also the intelligent systems that increasingly shape organisational outcomes. Those that embrace this responsibility early will not only reduce risk, they will build trust, resilience and long-term value in an increasingly digital world.