AI now can reveal your sarcasm - or maybe not - Tech Friday

While people are used to paying attention to the tone and meaning of the words that have been heard, developed AI-based algorithms are being taught to detect connections between words themselves and to see the irony and intentional lies. Now intelligence officers/ agencies are ready to prove the ability of artificial intelligence to detect the human art of sarcasm. The tool, funded in part by the U.S. military, lets them analyze the social media content (including comments, posts, forum talks) while ignoring posts that shouldn't be taken seriously. A couple of scientists from the University of Central Florida conclude the following: "Сertain words in specific combinations can be a predictable indicator of sarcasm in a social media post, even if there isn't much context". To simplify, there are some words - "triggers", that get more attention from the system and help it to identify sarcasm. Thus, words as'just', 'again', 'totally', or exclamation mark make it easier to recognize the suggestion of sarcasm in text.