According to new research by two Stanford doctoral students, the National Security Agency’s (NSA) mass surveillance of telephone metadata could potentially reveal much more detailed information about consumers’ private lives than what the federal government has claimed.
Based just on the phone records of 546 volunteers, Jonathan Mayer J.D. ‘12 Ph.D. ‘16 and Patrick Mutchler Ph.D. ‘16 were able to draw a number of conclusions about those volunteers’ activities based solely on the phone metadata, which includes the phone number of the caller and recipient, the serial numbers of the phones involved, as well as the time, duration and potential location of each person when the call occurred.
The concern about metadata has surged since last summer’s revelations about the NSA’s surveillance of American citizens. Beginning in September of last year, according to Mutchler, the two researchers set out to find out exactly how much could be learned from the metadata that was the agency had access to.
“What we wanted to do was basically figure out exactly what you could learn about someone, or at least a lower bound of what you could learn about someone with this information,” Mutchler said. “So we built an app that allowed people to volunteer their metadata to us and basically had their phone logs going back a couple months.”
Mayer and Mutchler were able to draw a number of facts from the volunteered information by conducting an analysis of individual calls made by the volunteers.
Through the crowdsourced data, it was determined that 57 percent of the volunteers made at least one medical call and that 40 percent of the calls were related to financial services.
“We were able to identify a number of patterns that were highly indicative of sensitive activities or traits,” Mayer said, offering an example of one participant that had contacted several local neurology groups, a specialty pharmacy, a rare-condition management service and a pharmaceutical hotline used for multiple sclerosis.
According to Mayer, the research shows that metadata from phone calls can offer a multitude of information about family, political, professional, religious and even sexual associations, among other details.
However, the fact that the researchers were able to find out that much information wasn’t a surprise, according to Mutchler. He mentioned that neither of them had reached out to their participants about the results of their research.
“We’ve adjusted our app a bit to give them a little bit of automated feedback about what businesses and people who were able to identify in their call logs, but we haven’t actually asked the people how they felt about these things,” Mutchler said. “Though, if I were to guess, they would all be upset because the people that sign up for our app tend to be people that are very privacy-aware and see the NSA as invading their privacy, so I would imagine that these particular people would be very upset. But I don’t know if that necessarily that would generalize to the [whole] population.”
Regardless, the research contributes to the public knowledge about the potential of phone metadata.
“I think that this have and will continue to inform people about the nature of metadata and what it means to collect data,” said John Mitchell, the researchers’ faculty advisor. “I think most people have a sense that showing your phone bill to someone else might say something about what your issues were and what you were actively lately concerned about, that you can reasonably draw conclusions [from the metadata].”
Mutchler also expressed similar sentiments about public opinion.
“The goal here really is that if we’re going to have a national discussion about this topic, it’s very important to have facts about it,” Mutchler said. “This makes it harder for people that are trying to claim things that are untrue — sort of get away with it — and it also makes it so that people are more informed. They’re able to make the more informed decisions about how they feel about these programs.”
Contact Catherine Zaw at czaw13 ‘at’ stanford ‘dot’ edu.