研究人员创造工具,从人工智能生成的内容中去除水印,引起对深层假发现的关切。 Researchers create tool that removes watermarks from AI-generated content, raising deepfake detection concerns.

加拿大研究人员开发了一个名为 " Unmarker " 的工具,可以将水印从人工智能生成的内容中去除,从而破坏探测深层假象的努力。 Canadian researchers have developed a tool called UnMarker that can remove watermarks from AI-generated content, undermining efforts to detect deepfakes. 水印旨在识别人工智能生成的内容,但Unmarker可以在不需要任何关于水印或创建水印的系统的信息的情况下将其去除。 Watermarks were meant to identify AI-generated content, but UnMarker can remove them without needing any information about the watermark or the system that created it. 这引起了对划水标记作为防范深层假象的防御手段的有效性的关切,促使人们呼吁采用其他解决办法。 This raises concerns about the effectiveness of watermarking as a defense against deepfakes, prompting calls for alternative solutions.