As the insurance sector matures across Africa, the focus is shifting from digitalisation to the application of AI and hyper-personalisation, with efforts to preserve empathy at the claims stage, says insurance provider Hollard South Africa.

The company notes that, as technological capabilities expand, the fundamental question is whether an AI agent can really replace the empathy required during a crisis or loss, or if its role should be strictly confined to back-office and non-client-facing process optimisation.

Over the past five years the insurance experience has dramatically changed. There are now two distinct pathways: the functional journey and the emotional journey. The functional aspects – underwriting, policy administration and initial claims processing – are prime candidates for total digitalisation, says Hollard International CEO Pravin Kalpage.

Digital processes reduce friction, lower costs and improve accuracy. Claims can be assessed more quickly, and policy changes can be executed in real time. Underwriting models can become more precise.

“By automating these high-friction, data-heavy tasks, insurers can strip away the administrative delays that traditionally plague the industry. However, the objective of this efficiency is not to remove humans from the equation, but rather to create the capacity for them to intervene where it matters most,” he says.

The consensus among industry leaders is that these two realities should coexist side by side.

“Automation should handle the logic, leaving professionals to handle the heart”, with Hollard noting that this approach ensures that, while the payout is processed with the speed of an algorithm, the support is nuanced and underpinned by human empathy.

The next frontier for the local market is hyper-personalisation, that is, leveraging data to tailor product delivery to the individual.

“We are already seeing the early stages of this shift with firms like Naked Insurance, where consumers can toggle their risk profile and see an immediate impact on their premium. However, this hunger for consumer data brings the industry into a head-on collision with the growing demand for privacy and ethical AI usage,” cautions Kalpage.

African Market Lessons

In East Africa, particularly in Kenya, the market effectively leapfrogged legacy infrastructure, evolving around mobile solutions from the outset.

“There are businesses that have provided platforms where everything from micro- loans for day traders to insurance premiums is consumed [using] a mobile phone. These markets have benefitted from a regulatory environment that favours agility through proof-of-concept sandboxes, where concepts can be tested and failed fast without the immediate burden of heavy regulatory approval,” he adds.

Hollard says a “big learning” is that each country’s regulatory environment matters for how AI is used and how insurance products are adopted, as some countries have stringent institutional processes to ensure consumer protection. The trade-off, however, is that while the safety net prevents large- scale failures, it can stifle innovation.

“There is a need for creating room in regulatory processes to allow smaller-scale, lower-risk testing of new concepts, to allow the local market to breed innovation while maintaining its commitment to data security,” Kalpage avers.

By embracing automation for functional efficiency and fostering a regulatory environment that allows for controlled experimentation, the industry can ensure it remains fit for purpose.

“Ultimately, the future of insurance lies in ensuring that Africa adapts continental lessons to each country’s regulatory context. The goal should be to secure a better future for the consumer by combining the best of what technology can calculate with the best of what a human can feel.”

Insurance claims are inherently linked to moments of vulnerability for the customer, and therefore require empathy, reassurance and human judgement.

“Technology should create space for this human touch, not eliminate it. The future model, therefore, is coexistence. Automate the predictable and repetitive; preserve human engagement where emotion and complexity intersect,” Kalpage concludes.