Efficiency vs. customer experience: the double impact of AI in insurance

The automation of processes through artificial intelligence (AI) is a hot topic in the Swiss insurance industry. However, a report in the Handelszeitung newspaper points out that while the increasing use of AI in insurance brings efficiency and cost reductions, it also jeopardizes the customer experience and personal advice.

The Swiss insurance industry is characterized by a constant increase in the amount of data that needs to be processed. This trend is driven not only by advancing digitalization, but also by increasing demands for transparency and efficiency in risk assessment and processing.

AI can help to manage these data volumes and set up automated processes. Some leading insurers, such as AXA Switzerland and Zurich, have already implemented AI-supported systems to speed up the processing of claims and the calculation of risks.

Examples of such systems include:

• Regression analysis: A technique for predicting customer behavior or customer outcomes based on historical data.
• Machine learning: An algorithm that learns through self-study and self-improvement, such as the optimization of sales processes.
• Natural language understanding (NLU): A technique for analyzing and interpreting natural language in the form of emails, chats, or letters.

However, the introduction of AI in insurance also raises concerns about the impact on the customer experience and personal advice. Some experts warn against the predominance of efficiency considerations over the individual needs of customers.

The introduction of AI in insurance can lead to a reduction in customer satisfaction and loyalty. When customers feel that they are only perceived as a source of revenue, they often lose their loyalty to a particular company.

Switzerland is considered a pioneer in the implementation of AI-supported systems, but the question is whether this also leads to an improvement in the customer experience or only to further dependence on efficiency considerations.