The most honest number in Africa’s CEO survey is 14%
African leaders admit their AI gap while global peers claim to be keeping pace, and boards should value candour over confidence.
Sign up for my Substack daily AI newsletter https://johanosteyn.substack.com
Follow me on LinkedIn: https://www.linkedin.com/in/johanosteyn/
When KPMG published its 2026 Africa CEO Outlook, the headline almost wrote itself. Only 14% of African chief executives believe their organisations are keeping pace with or ahead of AI developments, against 66% globally. It reads like a continent falling behind. I want to suggest a different reading. The 14% may be the most useful figure in the survey, because it is the one a board can actually work with.
CONTEXT AND BACKGROUND
KPMG surveyed 1,400 chief executives worldwide, 176 of them in Africa, between 11 August and 21 September. African leaders are not indifferent to AI: 72% call it a top investment priority, close to the global 79%, and 84% expect their headcount to rise over the next three years. They are, however, far more patient about the payoff. Only 48% expect returns within one to three years, against 71% globally, and 42% expect them only after three years, against 12% globally.
South Africa is hardly a laggard by continental standards. The 2026 Global Outsourcing AI Readiness Index, compiled by Ataraxis Management, scores South Africa 66.5 out of 100, the highest of the eight African countries in its group of 25 outsourcing destinations. Leaders who score well at home and still say they are behind are probably measuring themselves against the global frontier, which is the right yardstick.
INSIGHT AND ANALYSIS
The global 66% deserves the harder look. Official figures suggest that AI is less embedded than executive confidence implies. The US Census Bureau's business survey found that 37% of firms with 250 or more employees used AI in their operations, and between 17% and 20% of all firms did. The populations differ, and chief executives of large companies run the firms most likely to use AI, so the comparison is not exact. The distance between two-thirds of chief executives claiming to keep pace and a minority of even large firms using AI at all should still give pause.
Directors cannot easily check the claims either. In PwC's survey of about 600 US public company directors, 59% rated the information they receive on AI investments and returns as fair or poor, and 66% said the same of metrics linking AI to business performance. Below the boardroom, the picture is inflated too. A Visier survey of 1,000 US employees found that 48% sometimes overstate their AI use or expertise to colleagues or leadership.
I have previously written about this from the other side, arguing that African firms risk being stuck in pilots while AI's gains concentrate in a disciplined minority of companies. That risk is real. A leader who acknowledges the gap is at least in a position to close it.
The fair objections deserve weight. The 14% may simply be accurate, since data, skills and the cost of computing are real constraints across the continent. Low confidence can harden into fatalism and become an excuse for spending too little. The African sample is small, and the same survey found 86% "confident or cautiously optimistic" about keeping pace, which shows how much the wording of a question shapes the answer. None of this rescues self-assessment as evidence. It only confirms that a feeling is a poor measure of progress.
IMPLICATIONS
For boards, the lesson travels well beyond Africa. Asking an executive whether the company is keeping pace with AI produces a mood, not a measurement. Directors should ask instead for three things. First, usage data showing which systems are used, by whom and for what. Second, outcomes measured at the level of the process, such as cycle times, error rates and customer results. Third, return expectations set against what has actually been delivered, reviewed each year.
In South Africa, King V already makes the board responsible for overseeing technology and information. Evidence of AI readiness belongs in that oversight, preferably tested by internal audit or an independent reviewer rather than reported by the people who sponsored the programme.
CLOSING TAKEAWAY
A chief executive who says the company is behind on AI has handed the board something valuable: a starting point. One who claims to be ahead without data has handed it a risk. African leaders may well be behind, but they appear to know it. The more pressing question for boards everywhere is whether their own confidence rests on anything they could show.
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
Comments