Staying off social media is not enough to protect your privacy, study says

Even if you're not a user of a social media platform, it's possible to create a 95% accurate profile of you from your friends' accounts, according to new research.

When users sign up to Facebook or Twitter, they might believe they're only handing over their own data, but actually they're selling out their friends too, according to Professor James Bagrow.

The study suggests that simply "opting out" and not subscribing to social media platforms is not enough to protect an individuals' privacy.

Professor Bagrow and his team at the universities of Vermont and Adelaide have published a paper in the journal Nature Human Behaviour examining how this works.

His team found it was able to predict the content of a person's tweets using data collected from just eight of their contacts, and to do so as accurately as if they were looking at that person's own Twitter feed.

According to the research, the profile used to predict the content of a tweet is no different from the profile that social media platforms typically build.

The scientists, dealing with a branch of mathematics called information theory, believe they have shown that as long as an individual can be placed within a social network of friends, it is possible to profile them - even if they're not actually a member of the social media platform.

This profile is the sort which social media companies constantly work to build on everyone by establishing who their friends are, whether by accessing the contacts book on their phone or their email accounts, and their profiles could cover any number of interests, beliefs, or desires.

"All of these social media platforms are profiling people are using those profiles for targeted advertising or to recommend friendships," Professor Bagrow explained to Sky News.

"Think political party, favourite products, [and] religious commitments," say the researchers, noting profiled details could be established even if the friend had never been a member of the social media platform or had deleted their account.

They explain that the key detail their work shows is that there is a mathematical limit on the predictive power of these profiles, but also that it makes little difference if the person being profiled is a user of the platform or not.

"There's no place to hide in a social network," said Dr Lewis Mitchell, a co-author in the paper, who was a researcher in Vermont and now a senior lecturer in Adelaide.

"You alone don't control your privacy on social media platforms," added Professor Bagrow. "Your friends have a say too."