The Economist ran a story last week “The world’s most valuable resource is no longer oil, but data,” calling for new regulatory measures for the Internet giants. While it’s true that Google, Facebook, and Amazon control most of the Internet’s data, ride-sharing companies such as Uber and Lyft also yield tremendous power as we anticipate the advent of self-driving cars. News this week from Didi and Uber perfectly illustrates why folk at The Economist were so concerned.
Didi — Using Data for Good
It’s not hard to imagine terrible traffic problems in the world’s most populous country. According to a 2015 report by the Ministry of Public Security in China, there are 40 cities that have car ownership of more than 1 million.
Didi Chuxing, the Chinese equivalent of Uber, is combating traffic congestion by partnering up with local authorities and providing access to its real-time road data.
The city of Jinan, which boasts a population of more than 7 million people, installed smart traffic signals to sync traffic lights based on the number of cars on the road in real-time. But sensors alone cannot fuel this beast. The real-time traffic data is fueled by Didi’s “big data” backend to predict which roads will likely become congested. Initial results have reported an 11% improvement in delays caused by heavy traffic.
This traffic data platform named Didi Traffic is also being used by regulators in even bigger cities like Shenzhen, where car ownership exceeds 2 million. With 20 million ride requests per day, amounting to 2,000 terabytes of traffic data, Didi is in prime position to collaborate with authorities to build a smart city application.
Of course, Didi isn’t the first one to offer this kind of collaboration. Back in 2015, Uber partnered with the city of Boston to analyze its data and to ease traffic congestion. And speaking of Uber…
Uber — Using Data For Bad
Uber just can’t catch a break this year. It’s now under a criminal investigation by the U.S. Department of Justice for the use of the software known as Greyball.
The controversial tool allows Uber drivers to identify and evade local regulators in cities where the service hasn’t been approved yet. A Northern California grand jury issued a subpoena to review how this tool functioned and was deployed, so now Uber is facing even more legal challenges to top off its growing list of PR nightmares.
Uber claims that Greyball was part of a fraud detection and prevention program called Violation of Terms of Service. The software combined credit card data, device identifier, and location to classify each request as legitimate or bogus. If the request was deemed illegitimate, Uber said that it displayed false information about the cars around the area and denied service to the requester.
The problem with the software came when it was used to target officials and law enforcement personnel who might be looking to ticket drivers in cities, like Portland, where its services hadn’t been authorized.
The DoJ investigation will determine whether or not Uber mined credit card information to cross-reference ride requesters who work in the Bureau of Transportation, for example, to deny them service or to hide any Uber drivers in the area.
According to Reuters, Uber said “it used the Greyball technology in December 2014, while it was operating without approval, because it was ‘deeply concerned that its driver-partners would be penalized financially’ or otherwise for their driving.”
Even for a startup known for its aggressive strategies, the use of Greyball illustrates how much power these companies have in hoarding all the data. Perhaps it’s a bad idea to let companies unilaterally decide whether or not they want to use their data for good or bad.