Advances in AI, particularly Gen AI and Agentic AI, are creating significant opportunities, but they are also sharpening important questions for insurers. The question is, just because a process can be automated, should it be automated?
executive summary
Manuel Rodríguez Vera from Capgemini’s WNS division provides a practical approach for insurers to embed control, privacy and compliance into their AI systems, while enabling cross-jurisdictional collaboration and innovation. Approaches being discussed include restricting data and AI workloads to specific regions and a new approach to machine learning known as federated learning.
High-impact AI use cases offer the potential to reduce costs and risks. Improve productivity and efficiency. Drive revenue across sales, underwriting, claims, and services. However, as insurers accelerate the adoption of AI, they face increasing risks that exceed its effectiveness in critical areas such as cybersecurity, privacy, and explainability. In 2025 alone, data breaches will affect nearly 300 million individuals, highlighting how quickly trust can be eroded. At the same time, customer expectations and regulatory oversight regarding data usage and model transparency are increasing, leading to more and more direct questions about how data is used and how models operate.
Please enter your email address to read the full article.
Already a subscriber? Log in here
