Deepfakes are often used by cybercriminals to cover up crimes, but banning artificially created content can only exacerbate this problem.
Recently, deepfakes and other "artificially created" media content are becoming more common on the web. The co-founder of Pinscreen, a company that uses artificial intelligence to create digital avatars, told Business Insider that the number of deepfakes on the web is doubling every six months. As deepfakes proliferate on the Web, Lee suggests, it will soon become much more difficult to separate fact from fiction.
“What worries me most is not the abuse of deepfakes, but the world in which any image, video or audio can be manipulated. If something in a world like this can be fake, then nothing should be real”, artificial intelligence and deepfake researcher Hany Farid of the UC Berkeley School of Information told the publication.
According to experts, attackers are increasingly using deepfakes as a cover for their crimes, a problem that becomes more dangerous as society gets used to deepfakes. However, a complete ban on deepfakes can exacerbate the problem. For example, China is currently the only country where deepfakes are prohibited by law.
“Let's say some very problematic footage of Uyghurs in internment camps came from Xinjiang province. Now the central government in China has the right to declare that this is an illegal deepfake”, the experts explained.
Combined with the Chinese authorities' control over the country's internet, this can be a very effective enforcement tool.
The researchers believe that a variety of technological, legal, regulatory and educational approaches are needed to address this problem. Users are advised to be cautious and skeptical about content on social networks, especially if the information is not confirmed by several reliable sources.