Insurtech reshapes Vietnam s insurance sector

The article discusses how the insurance industry in Vietnam is being transformed by digital technologies, particularly AI, which raises concerns about transparency and cybersecurity. Insurers must now consider AI as a critical business risk rather than just an IT issue. To address these challenges, companies need to adopt resilient systems that can recover autonomously and maintain business continuity during crises.

One of the key challenges for insurers is the sheer volume of documents, paperwork and records they deal with. This can cause significant delays in processing, directly impacting the policyholder experience. GreenNode, a company specializing in AI-driven document processing, has developed GreenNode IDP, an advanced platform designed to handle the unstructured nature of insurance records in real-world operations. This platform ensures that sensitive data is processed securely, compliantly and at scale.

In Vietnam, BAOVIET IT Centre has adopted a hybrid AI approach, combining rule-based systems with AI for claims processing. AI handles rule-defined cases efficiently, while exception cases are routed to Agentic AI for deeper analysis. However, this AI deployment requires a robust governance framework, including explainability, audit trails, human oversight and strict data privacy standards.

Chubb Life Vietnam's chief customer and digital officer, Nguyen Ngoc Tu, identifies three values that technology cannot replace in life insurance: clarity, consistency and care. He emphasizes that while technology can enhance efficiency, trust is built through transparent communication about product benefits, costs, terms and conditions. Moreover, customer care remains essential during difficult times, as life insurance policies often require long-term commitment and emotional support.

Bach Ngoc An, Deputy CEO of Technology at OPES Insurance, stresses that the success of AI applications in insurance depends on their ability to create tangible business value. Insurers should carefully evaluate every AI project, ensuring it outperforms existing approaches. Data quality is crucial, as AI models require large volumes of high-quality data for effective training.

Finally, strong AI governance is necessary due to the inherent non-deterministic nature of AI models. Human oversight and risk management frameworks are essential for high-impact decisions, ensuring accountability and preventing complete AI replacement of human judgment.