Hurricane Ian Destroyed Their Homes. Algorithms Sent Them Money

Hurricane Ian wreaked havoc over Florida in late September, leaving a trail of destruction. Some people in three of the worst hit counties saw an unexpected beacon of hope a week after the storm.

Thousands of people in the three counties received a push notification on their phones offering $700 in cash assistance. Satellite images showed that people living in badly damaged neighborhoods needed some help.

This new way of targeting emergency aid is being tested by GiveDirectly. The people who received money were users of a benefits app. GiveDirectly was able to offer aid only to people who lived in areas devastated by Ian more quickly than manually sorting through the rolls of the app's users, thanks to the use of artificial intelligence.

In the months after the world's economy was crippled by the Pandemic, GiveDirectly tested a similar idea in Africa. Clues from cell phone bills and signs of poverty were used to help households.

The Florida project was powered by a mapping tool called Delphi, developed by four machine-learning experts who worked with GiveDirectly over six months. The software highlights communities in need after disasters by showing live maps of storm damage with data on poverty from sources such as the US Centers for Disease Control and Prevention. The storm damage data is provided by another tool that uses machine learning to analyze satellite imagery from before and after a disaster and estimate the severity of damage to buildings.

There is a map that shows where socio-economically vulnerable and where has been damaged. It can speed up delivery of aid.

Satellite images of a few hundred buildings in a disaster-struck area that are known to have been damaged are manually labeled by the damage assessment program. The software can detect damaged buildings at a rapid pace. According to a research paper presented at a 2020 academic workshop on Artificial Intelligence for Disaster Response, the auto-generated damage assessments match those of human experts with between 85% and 98% accuracy.

In Florida this month, GiveDirectly sent its push notification offering $700 to any user of the Providers app with a registered address in the areas where more than 50 percent of buildings had been damaged. Half of the people who took up the offer have been paid. The organization will give out more than two million dollars in direct financial aid if every recipient takes the offer.