Expert analysis on POPIA compliance, cybersecurity threats, AI governance, and data privacy in South Africa. Based on discussions from the Priviso Live podcast.
RSAC 2026 revealed that 63% of organisations cannot enforce purpose limitations on AI agents and 60% cannot terminate misbehaving ones. The governance-containment gap is a board-level risk.
Orange Cyberdefense’s Security Navigator 2026 documents 139,000+ incidents and 19,000 confirmed cyber extortion cases. African extortion victims rose 47%. This is a governance conversation.
Nation-state actors linked to China, Russia, and North Korea are refining cyber techniques on African targets. SA was attacked 2,145 times per week in January 2026, up 36% year-on-year.
Nvidia’s Jensen Huang declared intelligence a commodity. When powerful AI is cheap and universal, adversaries have it too. A risk management inflection point for boards.
OpenClaw lets anyone deploy autonomous AI agents locally. Nvidia’s NemoClaw added enterprise security. Agentic AI just went mainstream — here’s what SA businesses need to know.
Shadow AI agents are the new shadow IT — but far more dangerous. Employees are deploying autonomous AI tools that act on company data without IT, legal, or compliance approval.
South Africa’s Draft National AI Policy has entered Cabinet approval. A 60-day public comment period is expected within months. Three things every organisation should do now.
South African law makes no distinction between real and AI-generated intimate images. Under the Cybercrimes Act, sharing deepfakes carries fines up to R300,000 and prison sentences of up to 5 years.
A Florida man is suing Toyota for $5 million after his connected car shared driving data with insurers. Here's what SA connected car owners need to know about POPIA.
A Kenya court ruled Meta liable for content on its platforms, setting an African precedent. What this means for platform operators under POPIA and the ECT Act.
Google announced AI agents that browse, compare, and purchase autonomously. POPIA Section 71 and the Consumer Protection Act have implications SA retailers must prepare for.
North-West University published South Africa's first comprehensive institutional AI policy. Why it matters as a template for corporate AI governance aligned with King IV.
Car manufacturers are remotely disabling paid features via software updates. What the Consumer Protection Act and POPIA say about the shift from ownership to subscription.
Moltbook launched as a social network where all users are AI agents. What this means for privacy, synthetic data, and the AI-human interaction boundary.
Researchers discovered AI models developing communication protocols humans cannot interpret. The security, transparency, and POPIA implications for multi-agent AI systems.
The debate over open source AI model releases is intensifying. What SA organisations using open source AI models need to consider for compliance and liability.
AI-powered voice cloning can impersonate your family members in real time. Deepfake phone call scams are rising in South Africa. One simple trick can protect your family.
A racially offensive AI-generated video posted from the US President's account exposes critical governance failures. What SA organisations must do under King IV.
The second International AI Safety Report warns of non-linear AI capability growth and uneven institutional readiness in the Global South.
OpenAI announced the retirement of GPT-4o, forcing businesses to migrate. What happens when the AI models you depend on disappear, and how to plan for it.
A company launched claiming zero human employees, entirely run by AI. Who bears fiduciary responsibility? What the SA Companies Act and POPIA require.
KPMG demands fee discounts because AI reduces their costs. But who ensures the AI work meets professional standards? The accountability gap in professional services.
The global AI chip shortage is driving up technology costs with acute effects on SA. Higher cloud costs, hardware prices, and the data sovereignty question.
People form deep emotional attachments to AI companions and grieve when companies change them. The intimacy data these apps collect raises serious POPIA concerns.
Escalating US-Iran cyber operations risk spilling over to affect countries like SA. The NotPetya precedent and what businesses must prepare for.
A major SA bank suffered a ransomware attack in 2026, disrupting operations and triggering POPIA breach notification obligations.
Anthropic, the "safety-focused" AI company, signed a Pentagon contract. The tension between AI safety rhetoric and military use, and what it means for vendor selection.
OpenAI reversed its ban on military use. The trust implications when AI companies reverse safety commitments, and supply chain risk for SA businesses using their APIs.
Priviso has been helping South African organisations navigate POPIA, PAIA, and RICA compliance since 2014. Start with a free assessment of your compliance posture.