AI-generated deepfakes challenge ability to distinguish truth

Recent studies and expert discussions in Denmark and Norway highlight growing public difficulty in distinguishing between authentic content and AI-generated media. In Denmark, an Epinion survey for DR found that 41 percent of respondents likely or definitely would not be able to recognize manipulated sexual material, often referred to as deepfake pornography, while 32 percent expressed doubt regarding their ability to identify such content.

Ask Hesby Holm, director of the organization Digitalt Ansvar, noted that the rapid evolution of AI tools makes it increasingly difficult to navigate facts in public debate. He emphasized that the inability to recognize manipulated sexual material presents a particularly critical problem for individuals.

In Norway, concerns have been raised regarding the broader societal impact of AI-generated content. With reports suggesting that over half of internet content may be AI-generated, media professor Rune Karlsen from the University of Oslo warns that this could lead to a systemic loss of trust in news, traditional media, and interpersonal communication as the line between real and synthetic becomes blurred.