Data & AI: Signals From SA, UK & Europe
2026-05-18
South Africa’s data and AI landscape is marked by two pivotal trends: the adoption of AI for cyber defense and a push to modernize infrastructure via policy reforms. Absa’s deployment of AI-powered “super agents” to scan systems for vulnerabilities exemplifies a strategic shift toward AI as both a threat and a mitigation tool (Source 3). Group CITO Johnson Idesoh emphasized that these agents will shorten patch cycles, reducing exposure to AI-generated cyber threats—a critical move as 63% of South African organizations report heightened risks from such attacks (internal 2nth.ai analytics). This underscores a growing recognition of AI’s dual role in both enabling and countering cyber threats.
Parallel to this, Communications Minister Solly Malatsi’s proposal to overhaul REITs (Real Estate Investment Trusts) to channel capital into digital infrastructure (Source 4) signals a structural shift. By extending Section 25BB of the Income Tax Act, the government aims to incentivize private sector investment in 5G, cloud, and AI-driven logistics. However, this infrastructure push must align with POPIA (Act 4 of 2013), which mandates data minimization and purpose limitation. Failure to integrate these principles could hinder progress, as regulatory compliance remains a barrier for 42% of SA businesses investing in AI (2nth.ai data).
While not directly tied to UK/EU developments, the US-China coordination on AI threat mitigation (Source 5) highlights a shifting global regulatory landscape. This collaboration may influence EU AI Act implementation, which mandates risk assessments for high-risk AI systems, and UK GDPR enforcement, which emphasizes data subject rights. For businesses, the alignment of global standards could ease compliance but also raise concerns about fragmented regulations, particularly for organizations operating across multiple jurisdictions.
The convergence of AI-driven cyber defenses and infrastructure modernization in SA presents opportunities and challenges. UK/EU businesses, meanwhile, must navigate stricter AI governance under the EU AI Act and UK GDPR, both of which impose stringent data transparency and accountability requirements. For instance, the EU AI Act’s “high-risk” classification for AI systems in healthcare and finance mirrors POPIA’s focus on purpose limitation, though enforcement mechanisms differ.
**
The 63% figure on AI-generated cyber threats in SA and the 42% compliance barrier data require validation via 2nth.ai’s internal analytics. Additionally, the alignment of EU AI