0
energy
What is energy?
0 comments
In 2019, deepfakes evolved from a relatively jovial pastime into a serious ethical threat, slandering celebrities and disrupting politics.
In case you’re unfamiliar with the term ‘deepfake’, it essentially refers to the video manipulation and replication of human features using software centered around the principles of artificial intelligence, face tracking, and deep learning.
In laymen’s terms, it’s the process of mimicking one’s facial features and/or voice to look/sound real.
Networks consisting of interconnected nodes autonomously run calculations on input data. With enough user ‘training’ these nodes begin to arrange themselves to complete a user-specific task. This is deep learning, and it’s primarily being used to marry a certain nature of video with celebrity imagery, convincingly manifesting bogus content on the fly.
While programs of this nature may have started out as a harmless pastimes on the back pages of the internet, they’ve now been catapulted into the mainstream; re-purposed as a defamatory weapon to target prominent celebs and political figures.
So where did it all start?
The term ‘deepfake’ originates from an unnamed Reddit user by the same alias who rose to prominence in 2017 due to a series of fake celebrity pornographic videos. Aside from the obvious can of worms opened by this invasive and disturbing content, it would provide a glimpse into wider ramifications of the technology – still being tackled today.
It marked the first instance an individual was able to quickly and simply create a high-quality and convincing human imitation with readily available means. It certainly wasn’t the first ‘superimposed’ content to breach the internet, but it was the first instance of one being created so rapidly using consumer-grade graphics.
‘deepfake’ housed these celebrity imitations within a single subreddit that he created, and shortly after its conception members began contributing their own edits, mostly involving amusing movie clips. A whole section was dedicated to manifesting ever-memeable actor Nicholas Cage in random movies and TV shows.
The dangerous shift
As many predicted during the initial rise of deepfakes, the technology would quickly go on to incite more ethically dubious practices.
In June 2018 a computer program for deepfake creation, dubbed FakeApp, rose to prominence after being made available for download – it used the same software originally peddled by Reddit’s deepfake.
Shortly after, several videos emerged involving politicians that intended to deceive people into believing fictitious events and statements. Notably, a Flemish socialist party named sp.a posted a video on their Twitter and Facebook account showing Donald Trump appearing to taunt Belgium for remaining in the Paris climate agreement.
While the forgery was sniffed out by most viewers, those unaware of deepfaking believed the video to be real. Following this revelation numerous social platforms, including Twitter, Discord, and Gfycat, explicitly banned deepfakes and all associated communities, with Reddit following suit shortly after… and Reddit rarely says ‘no’ to anything.
