Red Cross survey: 87.1% say AI helps disaster relief

The Japanese Red Cross Society conducted its survey to understand attitudes toward AI in responding to natural disasters such as earthquakes, typhoons and heavy rain. The background it cites is a fast advance in AI technology, with AI provided by the Digital Agency also used by municipalities during the Reiwa 8 Kumamoto Earthquake, while disaster relief touching lives and livelihoods demands careful consideration. The survey aimed to grasp expectations, trust, anxieties and the desirable division of roles between people and AI to inform future activities.

The results show a gap between usefulness and trust. While 87.1% said AI is useful for disaster relief, only 62.4% said they partly trust the information AI provides, the most common answer on trust. The top anxiety, named by more than half, was the possibility of AI giving wrong information, followed by not knowing how far to trust AI answers at 41.1% and unclear sources or grounds for information at 29.0%. On the preferred division of roles in disasters, the most common answer across various roles was humans at the center with AI assisting.

The Red Cross concludes that in disaster relief, while AI should be partially used, roles only people can fill remain important, such as staying close to survivors, giving them a sense of security, and responding to each person's situation. It argues AI should not replace people's roles, but be appropriately used as a means to complement human-centered support. How far AI is actually deployed in the field may hinge on whether concerns about wrong information and unclear sourcing are addressed, and on keeping human-led aid at the center.