Deep fakes

Google said Tuesday that it plans to tweak Google Search to make it clearer to users which images in search results were generated by AI or edited by AI tools. Over the next few months, Google will mark AI-generated and edited images in the "About this image" window of Search, Google Lens and the Circle to Search feature on Android, the company said, adding that similar alerts could appear on other Google platforms such as YouTube. Similar prompts may appear on other Google platforms, such as YouTube. Google says it will share more information about this later this year. Notably...

According to Yonhap News Agency, the South Korean police announced that they have launched a pre-investigation into Telegram's suspected permissiveness of Deepfake sexual crimes. Yu Jong-soo, director of the National Investigation Headquarters of the National Police Agency of South Korea, said at a regular press conference on the 2nd that similar to the situation in France, the Seoul Metropolitan Police Agency has launched a pre-investigation into Telegram, and pointed out: "This suspicion is related to the permissiveness of this crime (fake videos, etc.)." The minister also admitted that there were difficulties in the investigation process. "Telegram not only did not provide us with...

Recently, Google is vigorously launching new online safety initiatives aimed at more effectively removing explicit deep fake content from search on a large scale and preventing it from occupying a prominent position in search results. It is understood that when a user successfully requests to remove explicit false content depicting themselves without consent from search, the Google system will not only process this specific content, but will also be committed to filtering out all similar explicit results related to it, and even deleting duplicate images. Google product manager Emma Higham said that these protections have achieved remarkable results in dealing with other types of non-consensual images, and now…

The U.S. Senate unanimously passed a new bill, The Disrupt Explicit Forged Images and Non-Consensual Edits Act, aimed at combating pornographic deepfakes created without consent. The bill will provide victims with legal means to hold those who create or disseminate these false images accountable. The rapid development of deepfake technology in recent years has made the illegal production of pornographic images a growing problem. Especially when intimate images of some celebrities and ordinary people are maliciously created and disseminated…

As one of the fastest growing forms of adversarial AI, losses related to deepfakes are expected to soar from $12.3 billion in 2023 to $40 billion in 2027, a staggering 32% compound annual growth rate. Deloitte expects deepfakes to surge in the coming years, with banking and financial services being the main targets. Deepfakes are at the forefront of adversarial AI attacks, growing by 3000% last year alone. Deepfakes incidents are expected to increase by 50% to 60% by 2024, with 100 million incidents this year …

A group of British MPs have proposed a law to ban the use of artificial intelligence to create deep fakes of music stars in order to protect their voices and images from being infringed. Recently, more than 200 musicians signed an open letter calling for stronger protection against the abuse of AI, arguing that AI poses a potential threat to the voices and images of artists. Among them are well-known British artists such as Jess Green, the Monfort Brothers, Sam Smith and Zayn Malik. The bill proposed by the MPs also includes other measures to protect musicians from the possible threats posed by AI and ensure that AI becomes a force for innovation and inspires human creativity,…