Facebook tests new system that blocks drunk selfies before they’re posted

The festive season is peak time for waking up in the morning with a dawning sense of horror about the drunk selfies you posted on Facebook the previous night.

But it might not be for much longer - the social network is testing a feature  which would warn users when they attempted to post embarrassing pictures while drunk.

The feature is the brainchild of one of Facebook’s ‘deep learning’ teams, headed by New York University researcher Yann LeCun.

LeCun explains that the software would use algorithms to distinguish between your ‘two selves’ in pictures - your drunk self and your sober self.

When ‘drunk self’ crops up, the algorithm would offer a private, automated warning before posting to the network, which LeCun described in an interview with Wired as saying, ‘Uh, is being posted publicly. Are you sure you want your boss and your mother to see this?’

LeCun said that the algorithms would look for embarassing posts uploaded by other people - and could offer users a warning that content they might not want to be seen was being posted.

[ Best space pictures of 2014 ]

LeCun says, ‘Imagine that you had an intelligent digital assistant which would mediate your interaction with your friends, and also with content on Facebook.’

[ Year in Review 2014: Top 10 tech innovations ]

Facebook’’s ‘deep learning’ algorithms are already used to recognise people on the site - and its research projects have gone far beyond that.

[ BlackBerry launches Classic in last-ditch effort ]

One of those, Deepface can match two previously unseen photos of the same face with 97.25% accuracy. Humans can do the same with around 97.5% accuracy.