Social media companies are mapping the 'data signatures' of young children using information from family and household profiles, research suggests.
The finding comes from work by Dr Veronica Barassi of Goldsmiths, University of London analysing the business models, data policies, patent requests and privacy rows surrounding Facebook, YouTube, Snapchat, and Twitter.
Dr Barassi, who leads the British Academy funded Child Data Citizen project at Goldsmiths, said: “While none of the four companies I studied allow children under 13 to own a social media profile they collect large quantities of children’s data through adult accounts and are trying to harness the data of under-13s.
“As a parent as well as a researcher I find it worrying just how little we know about how social media platforms are indirectly collecting and storing children’s data. It’s easy to blame ‘sharenting’ and parents sharing too much about their children online but actually my research shows that social media companies like Google (Youtube) and Facebook may be actively trying to create unique profiles of individuals from birth onwards.”
Dr Barassi examined the data policies of the four firms concerned and identified the different data flows all the companies admit to gathering: registration log-in details, activity, content, and device data from their platforms as well as off-line data from third-party partners.
As well as making plans to directly gather children’s data through technologies such as YouTube Kids or Facebook Messenger for Kids, companies such as Google and Facebook are developing ways to indirectly harness children’s data through family profiles. Google Family Link, for instance, enables parents to add under-13s to their Google account, which enables the company to have direct access to large quantities of children’s data with parental consent.
In 2018 Facebook filed a patent for ‘Predicting Household Demographics Based on Image Data’ that will use facial recognition technology to profile “photos posted by the user and photos posted by other users socially connected with the user” as well as data from captions and other text to “build more information about the user and his/her household in the online system, and provide improved and targeted content delivery to the user and the user’s household”.
Dr Barassi said: “The danger of social media data is not the fact that a parent posts the image of her child eating ice-cream but the fact that social media companies gather large amounts of individually identifiable data from birth (for example, through facial recognition), and that increasingly more institutions, businesses, and AIs rely on social media data to make automated decisions about individual lives.”