"These days, artificial intelligence (AI) images are rampant, and looking at people drawn by AI makes me uncomfortable. It feels bizarre and eerie."
AI-based services are rapidly spreading from content creation to customer consultation, but consumers are expressing awkwardness and discomfort due to the limitations of AI.
In particular, it is analyzed that the so-called 'unpleasant valley' phenomenon occurs when AI fails to properly understand human emotions and context. The unpleasant valley theory, first proposed by Japanese roboticist Masahiro Mori in 1970, refers to the psychological rejection that people feel when technology mimicking humans is not perfect.
According to a recent survey conducted by the market research company Nielsen on 6,000 consumers worldwide on the 29th, more than 60% of respondents can detect AI-generated content, which leads to negative emotions. Among those who detected AI-generated content, 55% felt discomfort, and 48% indicated they do not trust brands that advertise on platforms with high AI dependency.
While technological completeness has improved with generative AI, consumers perceive inhuman elements in the content provided by AI, creating a sense of distance. This explanation suggests that AI fails to properly reflect cultural sensitivity and identity beyond simple technological limitations.
For instance, 60% of Black respondents felt that AI-generated content distorts their culture, while 56% of Native American respondents reported low trust in brands with high AI dependency. Patricia Ratulangi, vice president of global communications at Nielsen, noted, "This survey reveals the gap between consumer expectations about AI and the anxiety felt when using it in media and advertising."
Research results also show a decline in trust towards corporations utilizing AI. According to the 'AI Era Customer Experience Status Report' released by Salesforce on the 15th, among a survey of 16,500 respondents across 18 countries worldwide, 71% indicated that their trust in corporations using AI has decreased compared to a year ago.
Salesforce stated, "71% of respondents believe it is important for humans to verify AI-generated results, and corporations must adopt transparent AI usage practices to earn consumer trust."
AI systems are effective in handling repetitive and simple civil complaints. However, concerns have arisen that AI humans may remain mechanical in response to complex problems or situations requiring careful communication, potentially leading to consumer dissatisfaction.
Eunhee Lee, a professor at Inha University in the department of consumer studies, said, "The discomfort consumers feel towards AI-generated content arises not merely from technical completeness but from a lack of human elements, emotions, and authenticity. This is connected to the 'unpleasant valley' phenomenon, where the feeling that AI is encroaching on human domains can evoke discomfort and psychological resistance among consumers."