Brief

Researchers have developed a method to detect AI-generated fake news that can generalize across different prompting strategies used to create the content. By analyzing interpretable linguistic features such as lexical diversity, readability, and emotional intensity, a random forest classifier achieved consistently high performance, with AUC values ranging from 0.988 to 1.000. The study found that AI-generated text generally shows higher lexical diversity, lower readability, and less emotional intensity compared to real news, and these characteristics remain stable enough to be effective even when the AI's generation prompts vary. AI