A recent study highlights the limitations of AI language models in mimicking human emotional expression, revealing that these models achieve only 80% accuracy in distinguishing AI-generated language from human speech. While they excel at replicating conversational patterns found on platforms like X, they struggle with emotional nuances, particularly on Bluesky and Reddit. The research shows that affective language—expressions of positive emotions, affection, and optimism—remains a significant marker of human communication that AI fails to capture. Even when AI-generated text is adjusted for length and complexity to seem more human-like, distinguishing features in emotional tone still emerge. The study indicates that AI can imitate the structure of dialogue but falters in conveying the genuine feelings present in human interactions. Despite the clear signs of AI usage on social media, users continue to engage with AI-generated content. For instance, over half of LinkedIn long-form posts are reportedly AI-generated or heavily edited, reflecting a marked increase in such content since the launch of ChatGPT in 2022.