In the latest news, Google has released database of 3000 deepfakes. Deepfake technology takes audio clips and video of real people, usually of celebrities and other public figures such as politicians. By means of artificial intelligence, these are altered in often offensive ways. For instance, making people say things they never did or transposing their head on to the body of another person, etc.
Such videos are made in several ways by means of publicly available tools which alter their faces. Such videos are risky as they promote false propaganda and conspiracy theories.
The latest news confirms Google’s attempt to help researchers with tools which will further aid in pulling down the “harmful” fake videos.
The technology first surfaced in 2017, ever since it has emerged rapidly. In a statement based on the latest development, Google said, “Since the field is moving quickly, we’ll add to this dataset as deepfake technology evolves over time and we’ll continue to work with partners in this space.
“We firmly believe in supporting a thriving research community around mitigating potential harms from misuses of synthetic media. While many are likely intended to be humorous, others could be harmful to individuals and society.”
According to sources, the database will be incorporated at the Technical University of Munich and the University of Naples Federico II. The universities have created data by using four common face manipulation techniques for 1000 YouTube videos. With the help of both these databases, a train automation detection tool will further help in screening fakery.
Previous reports on the subject draw attention on Facebook’s fund of $10m (£8.1m) safeguarded primarily to detect deepfakes, Mark Zuckerberg himself being the victim of such fakery.
A video showing Zuckerberg crediting a secretive organization for the success of social network has previously made rounds.