Free Trend Report Free 2018 Report & eBook

Get the top 100 trends happening right NOW -- plus a FREE copy of our award-winning book. Our Research Methodology

This article is one of 350,000 experiments. We use crowd filtering, big data and AI to identify insights.

The Penny AI Predicts Neighborhood Wealth Based on Satellite Images

 - Jun 28, 2017
References: penny.digitalglobe & wired
Walking through a neighborhood, one can usually get an intuitive sense of where it ranks on the continuum of affluence, but Penny AI is a system that aims to do the same thing with internet images alone. The software system can make predictions on the median income of a given neighborhood by analyzing satellite images of that neighborhood and judging those images against what it knows based on census information from other areas.

Currently, Penny AI is only functioning for images of New York City and St. Louis. To be sure, this doesn't mean that the AI simply posts readings of census data based on where it is told to assess. Rather, when its frame is dragged over a given area, it looks at that region based purely on the shapes and structures within that frame. In fact, Penny gives its assessment alongside the actual census grade for that region, showing how accurate its assessment was.