BBC audience members could soon be using all the data from their social media and online accounts to fine tune the content they listen to and view.
The BBC is developing a personal data store that analyses information from multiple sources to filter content.
Early prototypes of the BBC Box draw on profiles people have built up on Spotify, Instagram and the BBC iPlayer.
The BBC will not store data for users. Instead, preferences will be kept in the Box so they can be reused.
The project is seen as “disruptive” because individuals will decide what they use their data for themselves.
The Box is part of a larger European project seeking to give people more control over their data.
The BBC Box is currently being trialled with 25 employees.
It draws on data from lots of different social networks and content providers to recommend content that matches users’ interests.
A “profiler” application analyses the data in the Box.
And this profile recommends events, holidays, music and activities, as well as BBC shows.
But the BBC will have no access to the data in the Box.
“No third party, not even the BBC itself, can access any data in the Box unless it is authorised by the person using it,” it says.
Dr Hamed Haddadi, from Imperial College London, said: “The aim of this platform is to do analytics across data sets that no-one else in the world has access to.”
Dr Haddadi led a separate three-year project to develop the “Databox” the BBC has adapted for its trial.
The impetus for the Databox came from European governments and regulators keen to give citizens control over how their personal data was used, he said.
“Legally now you have to provide the means for people to engage with their data,” he told BBC News.
“The more people that have awareness of this and have the tools to engage with it, the more they get curious about imposing limits on how much they share.”