Facebook Is Building Global Population Maps Using AI
This post is also available in: 
עברית (Hebrew)
Facebook’s goal of having the whole world go online is nothing new. Starting with its project to supply the whole world with airborne internet using its Aquila drone, facebook has been keen on global and ambitious projects to reach this end, and its new project is no different.
The company is now giving away for free new maps it is building that describe patterns of population density in the world’s poorest countries in unprecedented detail using high-res satellite images and machine learning to map every house in the photos.
Its Connectivity Lab is using image recognition technology to create population density maps that are much more accurate (to within 10m) than previous data sets.
Facebook wants those maps to help it figure out how to deploy the solar-powered drones and ground-based infrastructure it says can help get Internet access to a sizeable chunk of the more than four billion people who are not online today. The company says 10 percent of the world’s population lives in places where connectivity is not available. It is working with Columbia University to release its new maps so that other companies and organizations can use them, too.
“Our goal is to figure out how we can develop technologies and understanding how we can connect every person on the planet,” said Yael Maguire, the director and head of Facebook’s Connectivity Lab.
Robert Chen, director of Columbia’s Center for International Earth Science Information Network, believes the maps will have many uses beyond Internet access projects. “These higher resolution data will be useful in optimizing the location of health and sanitation facilities, planning energy and transportation networks, improving resource management and access, and facilitating humanitarian assistance,” he says.
The maps cover 20 countries and 21.6 million square kilometers, including India, Mexico, Sri Lanka, Nigeria, and several other African nations. Creating the maps took billions of satellite images and thousands of computers working for weeks. The social network is asking for feedback before generating maps for other countries where many people have poor or nonexistent access to Internet infrastructure.
