Company boards, not just technology teams, must take responsibility for governing artificial intelligence if businesses are to realise its full...
The argument that company boards, not IT departments, must govern AI is arriving late everywhere, but it arrives especially late in African boardrooms where most organisations have not yet articulated an AI policy at all. For South African corporates, this gap is acute: JSE disclosure requirements and the King IV governance framework have not been updated to explicitly address AI risk, leaving boards to self-define what responsible oversight looks like in the absence of regulatory pressure. The accountability deficit is sharper on the continent because AI adoption is accelerating through imported tools rather than domestically developed ones, meaning the governance frameworks are borrowed alongside the technology.
The call for boards, not just technical teams, to own AI governance is a familiar refrain in global corporate governance circles, but it carries particular weight on a continent where many companies are adopting AI tools faster than they're building the oversight structures to manage them. African enterprises have often leapfrogged legacy systems entirely, mobile money is the classic example, and there's a real risk AI adoption follows the same leapfrogging pattern without the governance maturity that slower, layered adoption elsewhere has forced into place.
The argument that boards must take responsibility rather than delegating it entirely to technology teams is really an argument about accountability architecture: technical teams can assess capability and risk, but only boards can weigh that risk against strategic exposure, regulatory uncertainty and reputational consequence. Most African corporate boards, like boards globally, simply don't yet have the AI literacy to do that weighing credibly.
The practical test isn't whether boards issue policy statements about AI governance, most are capable of that, but whether they build the recurring scrutiny structures (dedicated committees, regular technical briefings, independent audits) that make governance more than a one-time compliance exercise. Without that infrastructure, board-level AI governance risks becoming a credential rather than a practice.
READ THE SOURCE REPORT FROM CIO AFRICA →Enjoying Strata-AF™?
Sign in or sign up for a personalised feed and unlock Strata-AF™ Originals.
