
|
12 min reading
How many underwriting and claims decisions can we hand over to AI? It depends on how we design the process
AI in insurance generates plenty of interest, but the conversation often hits a wall as soon as decision automation comes up. Even though the technology can do much more, modernization efforts still tend to start with proven use cases such as document review, data extraction, and automated summaries. Insurers are far more cautious when it comes to using AI for more advanced tasks, including risk analysis, anomaly detection, building case context, and preparing recommendations. But perhaps we are spending too much time asking how many decisions we can hand over to AI, and not enough time asking how to design a process that uses the technology without giving up control, accountability, or auditability.
Magdalena Marczak





