Samsung’s latest use of deepfake technology using Artificial Intelligence has brought Leonardo da Vinci’s famous Mona Lisa painting to life.
A group of researchers at the company’s AI research laboratory in Moscow have used algorithms on a public database of 7,000 images of celebrities from YouTube. The AI system mapped facial features and movements on to a photo to bring it to life.
This video from The Telegraph shows the technology in action.
Apart from Mona Lisa, artificial intelligence technology also made videos from photos of Salvador Dali, Albert Einstein and Marilyn Monroe. The Samsung team, while outlining the work, called the creations “realistic neutral talking heads.”
Samsung AI brings the Mona Lisa (or any other picture) to lifehttps://t.co/2oN2CSTCxp#ArtificialIntelligence #future #technology pic.twitter.com/tq3fj7KVVQ
— Fribio Software Solutions Ltd. (@fribiocom) May 24, 2019
However, researchers at Tel Aviv University in Israel came up with a similar system in 2017. Dr Supasorn Suwajanakorn created a fake video of President Barack Obama.
They believe that this technology could advance to helping grieving relatives create avatars of dead family members. Usually, this technology had been used in the past to manipulate celebrities into pornographic videos.
However, all these have aroused concerns that the rise of deepfake technology can fall in the wrong hands. Some people think there is a huge potential for misuse. This is especially following the recent fake video of drunk American House Speaker Nancy Pelosi that went viral on Facebook.
Facebook and YouTube have since taken down the video. Unlike this technology, they only slowed down the video a little to make her speech appear like a drunken slur.
Is Nancy Pelosi drunk or are her dentures falling out? ⬇️pic.twitter.com/uA965XJ7eA
— Baylor Cook (@j_baylorcook) May 22, 2019