With generative AI (GenAI) rapidly reshaping the insurance industry, SAS, a global leader in data and AI, is urging insurers across Africa to prioritise governance from the outset to avoid costly trust gaps and compliance failures. Building on recent global customer success, including Polish insurer PZU’s groundbreaking work with SAS AI Governance Advisory, SAS is now calling for African insurers to embed governance-first frameworks that support both innovation and integrity.
“AI is redefining what is possible in insurance, but the speed of adoption often leaves governance playing catch-up. Trust is not built after the fact. It must be engineered into every model, each customer interaction, and all decision systems,” says Itumeleng Nomlomo, Senior Business Solutions Manager at SAS in South Africa.
This local push comes as SAS continues to expand its portfolio of trustworthy AI offerings globally. For example, SAS assisted PZU, one of Europe’s largest insurers, in adopting ethical AI governance as a strategic advantage, aligning internal stakeholders, compliance teams and data scientists through a single governance model. The move is already enabling PZU to deploy GenAI in customer-facing scenarios with confidence and clarity.
For African insurers, the ask is no different, but the starting conditions are often more complex. With rising climate risk, coverage gaps, and digital fraud all increasing at pace, insurers cannot afford to treat AI governance as a compliance checkbox.
A recent global SAS survey of over 500 insurance leaders found that 78% view closing the $1.8 trillion global protection gap as an ethical obligation but only a fraction feel equipped to do so. The same research revealed outdated tech systems, siloed operations and low innovation readiness as top internal blockers.
“The reality is that many African insurers are still modernising core systems while navigating regulatory uncertainty. This makes governance even more essential not less. A unified approach to AI governance gives insurers the structure and foresight to innovate responsibly while protecting their customers and brand,” says Nomlomo.
To support this shift, SAS recently launched the AI Governance Map, a free online assessment that benchmarks an organisation’s governance maturity across four key domains: oversight, compliance, operations and culture. The tool provides tailored recommendations, enabling insurers to pinpoint gaps and take pragmatic next steps without overhauling existing infrastructure.
This initiative is underpinned by strong industry validation. SAS was named a category leader for model risk management in Chartis’ 2024 RiskTech Quadrant and recognised for its forward-thinking approach to governance in GenAI. The company has also released transparent “model cards”, similar to nutrition labels for AI, and a growing catalogue of tools designed to help insurers track, audit, and explain AI outputs.
Crucially, these capabilities are being contextualised for African realities. As digital and financial crime surges and regulators strengthen oversight, SAS believes insurance leaders must move beyond defensive risk management and towards proactive trust building.
“Insurance is fundamentally about trust. As we build more automated, AI-powered experiences for customers, from claims processing to personalised product design, we must ensure those experiences are explainable, inclusive, and bias-free. That is not a technical problem. It is a governance imperative,” says Nomlomo.
SAS’s broader strategy includes the rollout of domain-specific AI agents within its SAS Viya platform, enabling insurers to automate complex decisions while maintaining oversight through built-in ethical guardrails.
As the regulatory landscape matures with models like the EU AI Act and NIST frameworks gaining traction, SAS sees early investment in AI governance as a competitive differentiator. Insurers who lead on responsible innovation will be best positioned to close protection gaps, reduce fraud, and win long-term customer loyalty.
Learn more about AI governance at sas.com/ai-governance.
