As per World Wide Worx's South African Generative AI Roadmap: 67% of large SA enterprises are now using generative AI, up from 45% the previous year. When Arthur Goldstuck presented the findings, he flagged the tension embedded in them: pervasiveness isn't the same as strategic integration.

Many companies are experimenting without connecting the technology to business outcomes, governance structures or any clear measure of success.

Senzo Mbhele, managing director of Cloud on Demand, frames the gap clearly: The organisations seeing real results aren't chasing every new development. They started with specific, bounded questions. Where can automation remove friction? Where can data analysis sharpen decisions? Where can AI improve customer experience without introducing governance risk? 

Companies that skip those questions end up with AI that looks impressive in demos and shows up nowhere in results. OfferZen's 2026 State of SA's Developer Nation report lands in the same place: 97% of SA tech teams use AI in their workflows, but only 36% of developers say it has positively affected their earning potential. Usage is universal. Impact is not.

This is the pattern running through our SA AI adoption coverage. For founders building AI products and operators deploying AI in their teams, the practical question isn't whether you're using AI. It's whether you can point to where it's working. 

It’s partly why local ops like Kwanda have taken such a unique approach to AI enablement for SA. And, if you haven’t read it yet, Mbhele's full analysis is worth reading over at Cloud on Demand’s blog.

You might also like: Our piece on Africa AI adoption, how SA founders using AI are navigating the same gap, and our look at making AI adoption safe for enterprises.

Get more SA tech and business news and subscribe to The Open Letter.