Skip to Content
Data privacy
Security & privacy

Think your mobile data is anonymous? Think again, according to new study

Look around the next time you’re out for a stroll in the city. You’ll probably notice cameras on everything from streetlights to parking garages. You might even see cell towers or other types of sensors, even if they’re supposed to blend in with the scenery.

In your pocket or bag, your smartphone is also at work, constantly generating location-tagged data. Sending a text, opening an app and using a credit card or public transportation smart card is enough to provide a general idea of your whereabouts to advertisers and tech companies, but also urban planners and designers working to plan and understand cities. And this all takes place under the assumption that your data is anonymized, removing any identifiers that could reveal who you are.

But you might not be so anonymous after all. A group of scientists and urban planners say in a new study it’s shockingly easy to find out who you are, and with high accuracy.

Anonymous data combined can identify you

The study was conducted by MIT researchers, who concluded that multiple, anonymized datasets on you and others’ movement patterns in cities is a “double-edged sword.” They’re meant to shed light on human behavior (for research purposes), so groups can study issues like urban poverty and transportation in order to plan cities better. But they can also come with the potential side-effect of putting your personal data at risk.

In an MIT news release, the study’s co-author explained how you data can be deanonymized quickly and used to identify you. The group took two anonymized datasets (mobile phone logs and transit trips) of people in Singapore and combined them. Each dataset had “location stamps,” only capturing the time and place of each. But when they put in place an algorithm to match users whose phone and transit data overlapped closely, it became much easier to identify specific individuals. In fact, it only took 11 weeks for the group to achieve a 95% accuracy rate – and much less time if adding smartphones’ GPS data into the mix.

Bonus: Documents reveal Facebook scraped call and text data without user permission

The researchers weren’t looking to make specific identifications. They simply wanted to illustrate that someone with bad intentions could use a similar approach to identify and even put people under surveillance.

Raising awareness to prevent data misuse

MIT Senseable City Lab founder Carlo Ratti said in the news release that the study, which she co-authored, made the group feel ” … a bit like ‘white hat’ or ‘ethical hackers.'” (Read the full news release by clicking here) But they hope to increase public awareness regarding the risks to consumer data. “We need to keep thinking about the challenges in processing large-scale data, about individuals, and the right way to provide adequate guarantees to preserve privacy.”

“The world today is awash with big data. In 2015, mankind produced as much information as was created in all previous years of human civilization,” explained Daniel Kondor, a postdoc in the Future Urban Mobility Group at the Singapore-MIT Alliance for Research and Technology. “Although data means a better knowledge of the urban environment, currently much of this wealth of information is held by just a few companies and public institutions that know a lot about us, while we know so little about them. We need to take care to avoid data monopolies and misuse.”

Stop robocalls once and for all

Robocalls are not only annoying, but they scam Americans out of millions every year. Learn Kim's tricks for stopping them for good in this handy guide.

Get the eBook