The World Population Project by World Data Lab
On a recent trip to Vienna, I had a chance to sit down with World Data Lab(WDL) to learn about their newest product: an AI system that models worldwide economic development from satellite photographs. WDL is the vision of several professional economists building a data model to measure and record changes in the world’s population.
WDL’s Geographic Information System processes satellite imagery to distinguish schools and homes for the elderly from grocery stores and shops, then meshes this information with data from other sources to develop key insights. In the case of North Korea, this enables WDL to approximate economic conditions on the ground despite very little data being available. WDL have also used this mapping model to provide economic analysis in Turkey and Austria.
The image-to-data model was an “a-ha” moment for me. It flips the traditional process of attempting to glean meta-insights from data in spreadsheets, to actually beginning with a meta-perspective through the images and then translating those insights into data.
This is not the first time WDL have blown my mind.Their World Poverty Clock and World Population Project crunch publicly available data sets on income, life expectancy, and population size, presenting the results in an accessible format that is actually fun to interpret. But their work with this data is not frivolous – it’s one of the few examples of an organization amassing and modeling large datasets for the benefit of humanity.
- In January 2018, WDL identified that India was transitioning out of its position as the country with the largest number of people living in extreme poverty (a position now held by Nigeria). How did WDL know, down to the exact month, when this significant milestone would occur? Their World Poverty Clock has the ability to measure and predict conditions in individual regions and countries and, in certain areas where it has refined the data, like Kenya, extrapolates data down to the person.
- Just this month, WDL noted a new world milestone. For the first time in history, one-half of the world’s population are living in households considered “middle class” (earning between $11 and $110 per day, on a World Data Lab 2011 purchasing power parity basis, a benchmark used by many organizations and governments, including India and Mexico).
This is the kind of data that many of us assume someone is already monitoring. It has enormous policy ramifications and should be considered at the tables where decisions are being made. However, even in our data-driven world, we still do not have sufficient infrastructure or resources committed to maintaining data for the public benefit.
It is estimated that 90% of the world’s data was generated in the last two years — from which entirely new inferences can be extracted and applied to help address some of today’s most vexing problems.
For all of data’s potential to address public challenges, most data generated today is collected by the private sector. Typically ensconced in corporate databases, and tightly held in order to maintain competitive advantage, this data contains tremendous possible insights and avenues for policy innovation. But because the analytical expertise brought to bear on it is narrow, and limited by private ownership and access restrictions, its vast potential often goes untapped.
– Stefaan G. Verhulst and Andrew Young, “How the Data That Internet Companies Collect Can Be Used for the Public Good”
When he left Facebook to found Cloudera in 2013, data scientist Jeffrey Hammerbacher infamously remarked, “The best minds of my generation are thinking about how to make people click ads… that sucks.” One could update Hammerbacher’s quote for 2018 by saying, “The valuable data of the world is being collected, hoarded, silo-ed, and used almost exclusively for commercial or political gain… that sucks.”
What isespecially tragic is that data maintained for public use and public benefit has the potential to offer common facts and mutually acknowledged truths for our thorniest political and social questions. At the present moment, many technology issues still exist in a rare “pre-partisan” stage, new enough that political and interest groups have not yet settled on intractable positions. We may yet be able to discuss the possibilities of a digitally enabled, data-driven future as humans beforewe engage as conservatives or liberals, and citizens of this or that country or continent.
World Data Lab, as a mission-driven for-profit company, is providing a wonderful example of how data can help us better see and understand ourselves as members of humanity – an important step on the journey to making technology – and policy – more “people-centered”.